Carsten Ullrich's Blog

A Chinese-English dictionary for Technology-Enhanced Learning

2011-09-19T06:57:00.000+08:00

One result of this year's Summer School Matchmaking Event on Technology-Enhanced Learning: A Chinese-English dictionary for Technology-Enhanced Learning. Take at look if you speak Chinese and let us know what you think.

Our modern skulls house a malleable mind

2011-09-07T09:58:00.002+08:00

"Darwin in Mind: New Opportunities for Evolutionary Psychology" is a fascinating article discussing the often heard claim that "our modern skulls house a Stone Age mind". This statement comes from the field of evolutionary psychology, a field with the thesis that "genetic evolution simply could not keep pace fully with the extraordinary rate at which human technology transformed environments.".
The authors assemble an impressive collection of research that quite convincingly indicates that this thesis does not hold. Some extracts:

"There have been substantial human genetic changes in the last 50,000 years, with possibly as much as 10% of human genes affected"
"Events in the Holocene (the last 10,000 years) ... were a major source of selection on our species, and possibly accelerated human evolution"
"If humans exhibit equivalent rates [than seen in other species], then significant genetic evolution would occur over the course of a few hundred years."
"Recent trends in developmental psychology and neuroscience have ... stressed the malleability of the human brain, emphasizing how experience tunes and regulates synaptic connectivity, neural circuitry and gene expression in the brain, leading to remarkable plasticity in the brain’s structural and functional organization"
"the human brain has too much architectural complexity for it to be plausible that genes specify its wiring in detail; therefore, developmental processes carry much of the burden of establishing neural connections"
"there is no evidence for modularity in central systems such as those involved in learning and memory."
"... accounts of the evolution of brain and cognition cannot in themselves explain the brain’s underlying working mechanisms"

What I find fascinating about this research is that it highlights the potential of us human. We are not driven and out into a final shape by nature, but carry potential and possibility to develop.

The article is freely available on PLOSBiology.
Darwin in Mind: New Opportunities for Evolutionary Psychology by Johan J. Bolhuis, Gillian R. Brown, Robert C. Richardson, Kevin N. Laland

Quote from Mencius

2011-08-30T12:03:00.000+08:00

Another great quote from a Chinese teacher/philosopher:

"One who believes all of a book would be better off without books."
Mencius

"尽信书，则不如无书 "
孟子

See one of my earliest blog entries for a similar wise quote from Xun Zi.

Call for Papers: International Workshop on Enabling Successful Self-Regulation in Open Learning Environments (S-ROLE 2011)

2011-07-29T10:17:00.000+08:00

Dear readers, please consider contributing to this workshop I am co-organizing:

International Workshop on Enabling Successful Self-Regulation in Open
Learning Environments (S-ROLE 2011)
8-10 December, 2011, Hong Kong, in conjunction with
ICWL 2011 - International Conference on Web-based Learning

http://dbis.rwth-aachen.de/S-ROLE2011/

Workshop Submission Deadline: September 11, 2011

WORKSHOP TOPIC

This workshop focuses on the design of personal learning environments
and its underlying psychological and pedagogical rational. While a
significant amount of research currently investigates PLEs, in-depth
investigations on how to successfully enable self-regulation in
practice are rare. For instance, most successful PLE usage examples
were driven by digitally literate and self-motivated learners. The
workshop welcomes contributions that elaborate on conditions which are
necessary that a learning environment supports self-regulated learning
and that a learner can use the personal learning environment in a
meaningful way. Furthermore, guidelines and principles should be
elaborated how a learner can compile her own learning environment and
how the compilation can be supported. Case studies from test-beds that
involve "average" learners (e.g., adult learners with limited
opportunities to study and low digital literacy) are particularly
welcome.

Research questions to be addressed include:
• Which factors are relevant that a learning environment supports
self-regulated learning?
• How can a learning environment be personalized to the needs of learners?
• Which guidelines can be made to support the assembly of a learning
environment?
• Which recommendation strategies and systems can support the assembly
of learning environments?
• What case studies about self-regulated learning in open and personal
learning environments are available?
• Which evaluation strategies are possible in the context of learning
environments?

The expected duration of the workshop will be a full day.

SUBMISSION PROCESS

Workshop papers should have a length of 10 pages and must be formatted
according to the LNCS author guidelines. In order to submit a workshop
paper, please use the workshop submission system installation at
EasyChair.

Each workshop paper will be reviewed by 3 reviewers (members of the
Programme Committee). The accepted papers will be published in a
separate volume in Springer Lecture Notes in Computer Science (LNCS)
as a second post-proceeding volume after the meeting.

WORKSHOP ORGANIZERS

Milos Kravcik RWTH Aachen University, Germany
Alexander Nussbaumer Graz University of Technology, Austria
Carsten Ullrich Shanghai Jiao Tong University, China

PROGRAMME COMMITTEE (to be confirmed)

available on the website

WORKSHOP FORMAT

The full-day workshop format will foster interactive presentations and
constructive work. Workshop papers and demonstrations have to address
at least one of the research questions specified above.

The schedule of the workshop will include sessions for paper
presentation, discussion of the individual papers, plenary discussion
of the workshop topics, and collaborative elaboration of key aspects
raised during the workshop.

Further information: s-role2011@easychair.org

Learning Object Repositories: Requirements from the Practice

2011-07-25T11:05:00.000+08:00

Maybe I'm having the wrong expectations about Learning Objects / Educational Resources. I started working on that topic more than 10 years ago, and the first 8 years as a researcher investigating what you can do with them, under "optimal" conditions. Recently, I "switched" sides and wanted to use them from a teacher's perspective, but what I find is to 99% not useful at all. Let me explain.
From 2000 to 2007 I worked in the ActiveMath group where we investigated reuse of learning objects, basically what you can do with fine-grained LOs, self-containing, paragraph long, properly annotated with metadata (such as dependencies, difficulty level, type). If you have plenty of those, amazing things become possible, such as automatic courseware generation. The drawback is obviously that such resources are expensive to author, different viewpoints on the metadata, etc. But if you get over such problems within a project, it opens up a lot of possibilities.
Now, since about two years, besides doing research in technology-enhanced learning, I started to teach French in the distant university of Shanghai Jiao Tong University. In these few years as a teacher I learned a lot about what you can expect from students and teachers in real life (pdf). In my teaching I face a problem that I also encounter as a language learner (of Chinese). I have a book I need to follow when I teach. So what I am looking for are not complete units, but rather individual content items to supplement my teaching. Additional exercises, activities. I experienced this from my own language learning: very quickly you know the texts and exercises from the text book by heart. So, in order to continue to practice, I need more exercises, specifically training a concept from the target language. A lot of such activities are out there, readily available as Web pages.
But to be honest, for my course, I did not find many helpful resource in repositories (actually I don't remember finding a single one). I searched through a lot of repositories, but for this task, none of them was helpful. Most often, the search functions are too general. Try it for yourself: go to your favorite repository and try to find a resource for absolute beginners in French that trains "articles".
In contrast, very helpful were link lists, like this le point du fle. There, I can search very quickly for activities training the target language concept.

Am I expecting something wrong? I want a search for a domain (French) the proficiency level (beginners), and a domain concept (articles). That doesn't seem to be to difficult metadata, doesn't it? I'm currently writing a little script that generates this metadata automatically.
What is your view on this? Is this a too specialized use case? Did I miss the right repository? Do I need to change my way of searching?

Encouraging Collaboration between Chinese and Western Students in a Matchmaking Event on Technology-Enhanced Learning

2011-07-15T16:13:00.000+08:00

Last week my lab, the e-learning lab of Shanghai Jiao Tong University hosted the Chinese-European Summer School Matchmaking Event on Technology Enhanced Learning, in cooperation with the European Network of Excellence Stellar.
Find a report on the event at the e-learning lab site, including the slides of our invited experts:

Organizing such an international event (12 Chinese students, 6 "Western" students, mostly from Europe) was quite a learning experience. Our Chinese students were a bit shy, due to differences in command of English and research experience, but nevertheless, in the end, the group work was excellent.

Here is a list of things that really helped to bring our students together and improve interactivity and collaboration:

Shared rooms. Having one Chinese and one Western student in one room is an excellent ice breaker. We could see how these "pairs" stuck together during the first days.
Multicultural groups. Before the event, we assigned the students in groups (2 Chinese + 1 Western student). Each group was responsible to lead the discussion of the expert talks (including introduction to the speaker and leading the discussion (which means to prepare questions about the talk)). We changed groups after two days to bring a different mix of students together.
Counteract passivity/shyness. In our groups, Chinese students tended to have the Western students present the outcomes. Very easy to avoid such behavior: select the student to present just before the presentation itself. This forces everybody to contribute and be aware of the outcomes, since there is the "danger" of becoming the presenter.
Collect the outcomes of groupwork before the presentations. After each groupwork, we had a presentation session. To avoid that the groups continue to work on their presentations while the other groups are presenting, simply collect the outcomes (ppt, txt, whatever) from all groups before the presentations and be very clear about the fact that you will not accept any revision. I had to do this once, and afterward it was not problem anymore.
Make the students use templates. In our sessions (especially the students' introductions) we had the problems that some students talked much longer than the assigned 5 minutes. To circumvent this, create a slide template in which the slides change automatically after 1 minute.
Made the students explain their social activities. Our Western students wanted to have a drink in the evening, our Chinese students wanted to go to KTV. Do you know what KTV is? Neither did the Chinese students know what having a drink means. KTV means Karaoke, "having a drink" does NOT mean you have to drink lots of beer, a coke or orange juice is also okay. Once this was clarified, our students had a great time together in the evening.
Mix experts and students during meals. Chinese students tend to sit at a table separate from the experts. Don't let that happen. Let the Chinese students sit first, then the experts, even if this means that a few people have to change seats.

We did not succeed in receiving criticism from our Chinese participant. The Western students don't have much of a problem of stating what they liked/did not like. Our Chinese students praised the workshop, but didn't talk about problems. Maybe an anonymous questionnaire would work, or maybe I just have to become better in reading between the lines.
All in all, it was a really successful event. The students achieved what we hoped for: they interacted, learned from each other and became friends. Even tears were flowing the day they had to say good-bye to each other.

Value creation in networks and communities

2011-06-27T10:21:00.000+08:00

Very interesting document by Etienne Wenger, Beverly Trayner and Maarten de Laat on what kind values of created in networks and communities and how to assess them: Promoting and assessing value creation in communities and networks: a conceptual framework.

From Section 4 onward they list cycles of value creation in networks and communities:

Immediate value, such as a good tip provided by a colleague. Activities and interactions can produce value in and of themselves. They can be fun and inspiring.
Potential value: Knowledge capital

Personal assets (human capital)
Relationships and connections
Resources
Collective intangible assets (reputational capital).
Transformed ability to learn

Applied value: Changes in practice
Realized value: Performance improvement
Reframing value: Redefining success

They also "suggest a series of questions to investigate as a way to reflect on the value that communities and networking produce" and provide a set of templates for this task.

Via Stephen Downes.

Free place for EU PhD Student at China/Europe Summer School Matchmaking Event

2011-06-20T19:50:00.000+08:00

We have one free place for a European PhD student at the joint Chinese/European Summer School Matchmaking Event. The first Joint Chinese-European Summer School Matchmaking Event on Technology-Enhanced Learning will take place from Friday, July 8th to Tuesday, July 12th in Shanghai, China. It is jointly organized by the e-learning lab of Shanghai Jiao Tong University and the European Network of Excellence STELLAR.
The Summer School Matchmaking Event aims at bringing together PhD students from China and Europe to foster international cooperation, identifying common research questions and broadening the perspectives by becoming aware of cultural differences and similarities. Several internationally renowned experts in TEL will introduce participants into the current state-of-the-art and help them to gain an understanding of the challenges of TEL from multi-cultural perspectives. The participants will perform matchmaking with their Chinese/European counterparts and jointly work and discuss on their topics.
There is still one place left for a PhD student from Europe. We will cover for accommodation and food in Shanghai, but cannot support travel. There is also the possibility to attend the "International Summer School of Doctoral Candidates in the Major of Educational Technology 2011" at Beijing Normal University.
The events will take place soon, so send me your application as soon as possible. An informal email stating why
you would be a good candidate is sufficient. Send it to ullrich.c+sjtuevent* a t *gmail.com.
The following speaker will give presentations at the event:

Pierre Dillenbourg, Professor of pedagogy and learning technologies at EPFL, Lausanne, Switzerland: Interactive furniture in collaborative Learning
Denis Gillet, Director of the Real-Time Coordination and Distributed Interaction Systems (React) Group, EPFL, Lausanne, Switzerland:
Ulrich Hoppe, Professor for “Cooperative and Learning Support Systems”, University of Duisburg-Essen, Germany
Ronghuai Huang, Dean of the School of Educational Technology, Beijing Normal University, Beijing
Weikai Xie, Deputy Director of the E-learning Laboratory of Shanghai Jiao Tong University University, Shanghai: P2P technology in education

Call for participation for Chinese PhD students for the Chinese-European Summer School Matchmaking Event on Technology Enhanced Learning

2011-05-03T16:09:00.001+08:00

In the last weeks I worked hard to organize the Chinese-European Summer School Matchmaking Event on Technology Enhanced Learning. I think it was worth the effort, we have put together a great program with even better speakers (Pierre Dillenbourg, Denis Gillet, Ulrich Hoppe, Ronghuai Huang, Weikai Xie):

The first Joint Chinese-European Summer School Matchmaking Event on Technology-Enhanced Learning will take place from Friday, July 8th to Tuesday, July 12th in Shanghai, China. It is jointly organized by the e-learning lab of Shanghai Jiao Tong University and the European Network of Excellence STELLAR. The Summer School Matchmaking Event aims at bringing together PhD students from China and Europe to foster international cooperation, identifying common research questions and broadening the perspectives by becoming aware of cultural differences and similarities. Several internationally renowned experts in TEL will introduce participants into the current state-of-the-art and help them to gain an understanding of the challenges of TEL from multi-cultural perspectives. The participants will perform matchmaking with their Chinese/European counterparts and jointly work and discuss on their topics.
Check out the complete call for participation at the Summer School Matchmaking Event and the application form.

The call is targeted at Chinese PhD students (in China). Please help to spread the news!

Teaching award for my French class

2011-04-13T14:53:00.000+08:00

Today I received the "致远 (Far-reaching) teaching award" of the Shanghai Jiao Tong University School of Continuing Education (what a name!) for my French class. I don't think it is such a big deal, but I'm glad anyway. In my class I did lots of experiments with Personal Learning Environments... and most these experiments didn't really turn out the way I expected. See my paper "Not Yet Ready for Everyone: An Experience Report about a Personal Learning Environment for Language Learning" for some results. So, I'm really happy that my students liked my way of teaching, which heavily relies on using Web-based tools, such as online translators, text-to-speech, collaborative text editors.
A big thank you to my students who voted for me.

Selected list of conferences on Technology-Enhanced Learning

2011-03-28T15:56:00.001+08:00

A few months back, I was asked by my department to draft a list of conferences about Technology-Enhanced Learning, so that they could decide whether or not to encourage our researchers to submit and attend. I asked my readers for feedback (link to Google Doc), and based on your help, personal experience and rankings I came up with these:

Also mentioned but not recommended were CATE and ED-Media.

This list is by no means exhaustive. For that, see the list by Clayton R. Wright via Stephen Downes.

Thank you to all 6 of my readers who completed the survey ;) and to Pierre-Alain Dorange for the picture.

Rembering Erica Melis

2011-03-24T11:12:00.001+08:00

It is with great sorrow that I have to announce that my former PhD supervisor and friend Erica Melis passed away in the end of February 2011.

Erica was internationally renowned for her research that lead to the development of the ActiveMath system. Coming from the area of Artificial Intelligence for Theorem Proving where, working together with Jörg Siekmann, she achieved a breakthrough in the field of knowledge-based proof planning, she started working in the area of Artificial Intelligence in Education in 2000.

Within only a couple of years, she established herself as highly respected researcher in the field and presented her work on adaptive Web-based learning environments and semantics for learning at all major conferences and journals. She was able to acquire major project funding and to supervise several students in the 10 years that followed.

Erica was also an enthusiastic painter, as you can see on her Web page. The picture above is one of my favorites.

I had the privilege of working with her during these years, first as a MSc student and then as a PhD student. Erica was not only an exemplary scientist, but also a very kind, trusting and open hearted supervisor, a great teacher and a mentor. She taught those entrusted to her what it means to be a scientist, by living it, and supported us in any way she could to live up to our potential. She will be greatly missed.

Napoleon's Vocabulary Book

2011-02-16T16:53:00.000+08:00

I realize this is the second post on Napoleon in a row. It wasn't planned, funny how things turn out. While I was visiting my parents in Bonn, we went to the exhibition "Napoleon and Europe. Dream and Trauma". Well done, worth a look.
One very interesting item on display was Napoleon's vocabulary book. On exile on Saint Helena, he only had an English newspaper to read. To my surprise he couldn't read English and had to start to learn. The page on display looks like he started eagerly but got more and more distracted and annoyed, ending up in drawing of what looks like a fortress.

Sorry for the bad quality---I tried to take a picture of the real exhibit but got stopped by a guard. The photo shows a screenshot of the exhibition book.

Stanley Kubrick, Napoleon and Data Classification/Visualization

2011-01-14T13:55:00.000+08:00

I've been reading Stanley Kubrick's script for his unmade Napoleon movie and came across his production notes at the end of the script:

PREPARATION THUS FAR

A great deal of preliminary preparation has already taken
place and I would like to briefly outline what this has
been.

1. A picture file of approximately 15,000 Napoleonic
subjects has been collected, cataloged and indexed, on
IBM aperture cards. The retrieval system is based on
subject classification, but a special visual signaling
method allows cross indexing to any degree of complexity.

...

6. A master biographical file on the principal 50
characters in the story has been prepared by graduate
history students of Oxford University. They have taken
the highlights of each person's life, putting a single
event and its date on a single 3 x 5 index card. These
cards have all been integrated in a date order file with
special signals indicating the names of the characters.
The system allows you to instantly determine what any of
the 50 people were doing on any given date.

I would love to know whether these files (in the literal meaning!) still exist. Would be quite intriguing to play with them. In the production notes he also describes that some of the battle scenes will involve 15.000 men. I wonder whether he would have settled for digital technology.

Talks at ICWL2010: PLE, language learning, adaptivity

2011-01-12T10:05:00.000+08:00

The International Conference on Web-based Learning 2010 took place beginning of December. I had quite a busy schedule as I took the chance of having a local conference to submit quite a few papers. I talked about our research on Personal Learning Environments, language learning, and adaptivity. Click on the title of the paper to download it as a pdf.

Not Yet Ready for Everyone: an Experience Report about a Personal Learning Environment for Language Learning. Abstract: A Personal Learning Environment (PLE) is a mash-up of learning
services. It enables students and teachers to assemble a work
environment that is adapted to a domain and specific individual
needs. In this article, we report on our experiences on using a PLE
for Language Learning in five French lectures at the Shanghai Jiao
Tong University Continuing Education School. We found that while a
PLE has the potential to simplify access to and usage of Web sites
and services for language learning, students will use it only if
properly motivated. Furthermore, at the time being, difficulties
that result from the user interface and technical implementation
make the interactions with PLEs difficult. The problems need to be
overcome in order for PLEs to become adopted by the average, not
technically highly literate students and teachers.
Not Yet Ready for Everyone: PLE for language learning
View more presentations from Carsten Ullrich.
Cross-Cultural Multimedia Language Learning:
Case Study and Analysis. Abstract: This paper describes a pedagogical pattern for cross-cultural language learning and its application in a case study involving learners in Australia and China. The pattern uses a multimedia discussion tool as its main component, supported by video conferencing and chat. We analyze how students interacted with the tools, especially the documents they created and the commenting behavior. Based on this analysis, we give a set of recommendations relevant to practitioners and researchers in technology-enhanced language learning.
Cross-Cultural Multimedia Language Learning
View more presentations from Carsten Ullrich.
Supporting Flexible Competency Frameworks. Abstract: Since Bloom's initial work on competencies in 1956, various
competency systems have been designed and used to assess students'
competencies. Different pedagogical researchers and stakeholders
prefer different systems. We have been collaborating with them. Such
systems are essential for the adaptation by adaptive intelligent
tutoring systems. Now, this paper presents how ActiveMath
integrates several competency systems to bridge the gap between
different competency systems and thereby facilitating the reuse of
learning objects across system boundaries. The combination of
competency-related data is achieved by mapping a new competency
system to the internal one.
Supporting Flexible Competency Frameworks
View more presentations from Carsten Ullrich.

Survey on Technology-Enhanced Learning conferences

2010-12-16T16:50:00.000+08:00

I was asked to make a list of conferences on technology-enhanced learning our researchers should be encouraged and supported to attend. There are some ratings of TEL conferences on the Web, but I'm very much interested in what you think. Please take 2 minutes and complete the survey below. I will blog about the results soon.
Loading...

Talk Announcement: Latent Semantics and Social Interaction

2010-12-03T14:35:00.001+08:00

It is my pleasure to announce a talk on Latent Semantics and Social Interaction by Fridolin Wild from Open University UK. Everyone is welcome to attend.
Date: 7.12.2010, 14:00-15:00
Location:
SJTU e-learning Lab
浩然大厦6楼
华山路1954号
Haoran Building, 6/F
1954 Hua Shan Road
http://maps.google.com/maps/ms?ie=UTF&msa=0&msid=103700951278915477597.00048166e58a9d8fcfe24

Abstract:
Latent Semantics and Social Interaction

Processing the meaning of texts with math and statistics has a long tradition. The 70ies have given birth to the vector space model as a new search engine technology, the 80ies applied means of factor-analysis to it, and the 90ies have diversified both applications and methods.

In recent years, a new focus has been added to enrich techniques for calculating semantic representations from texts with social data -- i.e. data about the social provenance of texts. This way, new models and techniques have been developed that allow to analyse social interactions of persons with content or that allow to compare the knowledge invested into the professional languages of different communities. The talk will introduce into the basics of a family of analysis techniques based on 'latent semantic analysis (LSA)' and its software support in R. It will complement the core LSA-process with means of network analysis and will thereby introduce the new technology 'meaningful interaction analysis (MIA)'.

Theoretical fundaments will be illustrated with concrete software implementations in the context of the FP7 ICT project 'language technologies for lifelong learning' and the EC-funded open network of excellence in technology-enhanced learning STELLAR.

Speaker:
Fridolin Wild is a research associate at the Knowledge Media Institute of the Open University of the UK. Fridolin studied information science in Regensburg, Hildesheim, and Munich (all Germany). He founded axon-e interactive media in 1998, and managed numerous commercial and governmental projects there until he left the company in 2003. In 2004 he graduated as magister artium in information science and politology.

Between 2004 and 2009, Fridolin has been working as a scientist at the Vienna University of Economics and Business in Austria within the European projects PROLEARN, iCAMP, LTfLL, and ROLE thereby serving as the overall technical manager with an additional work package lead in interoperability of social software learning tools in iCamp, leading the work package ‘infrastructure’ within LTfLL, and leading a work package on community building and sustaining in the ROLE project. At the Open University, Fridolin is since May 2009 the deputy coordinator of the European network of excellence STELLAR.

Additionally, he serves as the principal investigator of the LTfLL project, herein coordinating the infrastructure work package, and he shadows the lead of the work package on community building and sustaining in the ROLE project. Fridolin is the voted treasurer of the European Association of Technology Enhanced Learning (EATEL).

Mash-up: visualizing the top 500 authors in technology-enhanced learning

2010-11-08T17:18:00.005+08:00

Some time ago I was asked to analyze the "Top authors in Computer Education" list by Microsoft Academic Search. In my view, there are problems with the list (and rankings in general), but in this post I don't want to discuss this, but highlight the powers of mash-ups.
Update: Tony Hirst suggest to user Google Fusion Tables, and he is right, this makes it even simpler. See at the bottom of the post for more info.

The power of visualization is that it makes information hidden in the data visible. For this list, I wanted to see it on a map, to see at one glance where the top ranked scientist come from (Pittsburgh by the way) and which are near me (not so many, sadly).
Here is the visualization:

View Larger Map
The amazing thing is that I could generate this map by reusing free services: Google Spreadsheets for extracting the information and generating a RSS feed, Yahoo! Pipes to manipulate the feed and YQL to add additional information. In detail:
Data extraction: the data in the table is plain html, which makes access difficult. In addition, the top 10 entries are formatted differently than the remaining entries. To convert the data into a more processing friendly format, I used Google Spreadsheets. This spreadsheet extracts the author names and their organizations using these two formulas:

=importXML("academic.research.microsoft.com/CSDirectory/author_category_23.htm"; "//td/a[contains(@href,'Author')]//text()")

=importXML("academic.research.microsoft.com/CSDirectory/author_category_23.htm"; "//a[contains(@href,'Organization')]")

importXML is a very powerful function that allows you to extract data using xpath.

I then shared this document as a Web page and copied the RSS link. This makes the list of authors available for further processing.

Yahoo! Pipes is an amazing service for mashing-up data. I created a pipe that takes my RSS output and adds geo-information, that is, it tries to find a location for the organization of each researcher. The pipes first cleans up the data from the spreadsheet and strips off unnecessary elements. It also limits the input to the first 500 items, as otherwise I had some problems with query limitations.

The biggest problem is to add the geo-information. The table only contains the name of the organization, but no precise information on the location. Yahoo! Pipes offers the Location Builder module for converting text into locations, but it never worked for me. Luckily, hapdaniel has build a pipe that uses more intelligence to find a location from a piece of text. His pipe uses YQL, the Yahoo! Query Language, another great service from Yahoo! to process Web services (originally, I wanted to use YQL instead of Google Spreadsheet to extract the data, but I couldn't combine the results from the two queries necessary to extract author and organization into one feed). My pipe passes the organization name to his pipe and as a result gets back location information, when available. Obviously, this does not work every time, but often enough.

The pipe then adds the location information in the way it should be done for further processing and outputs the result: here is the result of the top authors in computer education as a KML file. You can load this link in Google Earth or input it into the search field of Google Maps to get the above map.

It is really amazing what you can do today with free services. Yahoo! has developed a suite of great tools, and I sincerely hope they will remain available.

Update: By importing the Google Spreadsheet into Google Fusion Tables, we get the geocoding for free, without having to do any RSS manipulation. Just import the data, that is all. Almost too easy! Here is the map (access the table containing the data about the researchers):

Sharism Forum at "Get It Louder" Festival

2010-10-29T09:14:00.074+08:00

Last week I attended the Sharism Forum at the "Get It Louder" Festival in Shanghai. Get It Louder is one of these events why I love living in Shanghai: a showcase of creativity.
I learned of the Sharism Forum only very late, and due to other appointments I only had time for a few sessions.
The first talk was the introduction to set the overall context of the forum. Isaac Mao gave an overview on Sharism. 40 minutes later I still had trouble understanding what Sharism precisely is. Isaac's talk was in Chinese, and I guess that quite a lot of meaning was lost in the very good simultaneous translation (a big thanks to the organizers for that) due to the many different areas that were touched upon in the talk.
After the talk I tried to explain Sharism to a colleague who missed the talk, but wasn't able to.
One axiom of Sharism is "the more you give the more you get". But there are millions of blogs out there, that only have a handful of readers. There are thousands of pictures on Flickr that have less than three views. So, I wonder, how could these people make the step from sharing to sharism (so that they become not Shamans, but Sharemans)? Having such suggestions would help to clarify the concept. I also wonder how to reconcile the fact that on the one hand according to Isaac in his talk we can only manage networks with about 150 people and on the other hand the massive amount of content that will come from a Sharism world? Can we really devote shared content the necessary time it deserves?
I have to admit I found some of the example given by Isaac not particularly relevant (assuming I have understood them correctly). According to Isaac, one recent campaign by Nike (the shoe manufacturer) could count as an example in Sharism as the public was able to influence the design of the shoes. Well, for me this sounds like a rather cheap and socially irrelevant ad campaign. Changing the design doesn't do any good. I want to be able to influence where and under which conditions the shoes are produced, not choose the colors of a shoe. But maybe in the world of advertisement this is a small step forward and in the future companies will embrace Sharim enough to be courageous to let their buyers determine important things.

The second talk by Hu Yong on "Citizen Jounalism and the Revolutionary Scene of the Netizen". A very interesting presentation, explaining the current state and problems of citizen journalism. I tried to memorize the three most pressing problems of citizen journalism in China, but I fear I already forgot. One point was that the traditional media opposes citizen journalism. Another point might have been that there is no support. And the third one, I don't remember. Guys, upload your slides somewhere, then people can continue thinking about your topics.

After the lunch break, Mike Linksvayer continued with a talk about Creative Commons. It was a good introduction into the principles and ideas of Creative Commons. I think such talks are tremendously important to increase awareness in this area in China.

Sadly I had to leave afterwards and missed quite a lot of promising talks. A big thanks to rejon (Jon Phillips) and the other organizers. I'm looking forward to the next Sharism Forum.

Seminar at Finland Pavilion "Sharing Inspiration in Higher Education"

2010-10-26T10:03:00.003+08:00

Our lab was invited by Prof. Erkki Sutinen to participate at the seminar "Sharing Inspiration in Higher Education" that took lace at the Finland Pavilion at the Shanghai World Expo site.
Erkki and we shared the session "Technologies for Development and Learning: innovations grow from concrete challenges. In his talk, Erkki had a few great quotes: "Innovation is socially meaningful invention put into practice" (I've been thinking about what innovation is recently, and I find this a very good answer) and "Risk taking means giving oneself a permission, sometimes requiring apologizing later".
Dr. Xie Weikai, the Deputy Director of our lab presented work in our lab and at the online college of SJTU in which we address specific challenges arising from the educational situation here in China, namely "Highly unbalanced education resource deployment", "Lack of accessible form of life-long learning for working adults" and "Shortcoming of traditional learning management systems".
I'm directly involved in the latter part with my research on Personal Learning/Teaching Environment, especially within the ROLE project. Have a look slides and please let me know any questions you might have:

Technologies for development and learning

View more presentations from Carsten Ullrich.

Some pictures from the event:

Chinese researcher on China becoming leading science nation: will take a long time

2010-10-25T08:44:00.003+08:00

The German journal "Die Zeit" has published an interview (in German) with Jiangang Li, a vice director of the Hefei Institutes of Physical Science. The interview discusses China's plan for nuclear fusion and, my focus here, contains a few comments regarding China's ambitions in science.
When asked about his view on the claim that China is good in copying but bad in developing its own technologies, Li answers that "it will surely take 100 years until China will become leading in science worldwide". The interview continues (my own translation):
Zeit: "But didn't China declare the goal to become worldwide leader in 40 years?"
Li: "To be honest, I find that unrealistic."
Zeit: "What are the major problems?"
Lu: "It is not enough to have a few intelligent people. Education must reach everyone. Even though the numbers of student are increasing, only 10% to 20% of the young people make it to university. Already today, China has a excellent top executives. But this level of education is not reached by most people."
...
Zeit: "What is the major difference between the science world in China and Europe?"
Li: "I think that politics in Germany is often a bit slow, because consensus has to be reached in all questions before decisions are taken. In China, if the government decides that a research project should begin, then it can start within one month. That is an advantage of our system."

I find Li's "insider" comments interesting, as they confirm my impression regarding the current state of education and level of science. It opens up a with questions, too: what does leader in science mean? The output of papers? I'm rather sure that China will rather quickly become number one in publishing, especially regarding the enormous amount of conferences that take place in China every year. These conferences, often bearing the IEEE logo and indexed by the important indexes, have important sounding names, but are absolutely worthless regarding the scientific quality. I've been to one, it was rather depressing and eye opening experience.
I find it also interesting that at least in these few sentences Li finds it more important to increase the number of students than changing the way they are educated (which I find more pressing).
And regarding his last point, the major difference between science in China and Europe being the speed with which decisions are taken, well, if there is the will, then Germany can take quick and intransparent decisions too, as seen with the financial crises, where even today a substantial amount of information is not available to the public...

Keynote about Semantic Web at Bayer Innovation Fair

2010-10-13T09:14:00.003+08:00

I was invited to be a keynote speaker at the Bayer Business Services (BBS) Innovation Fair 2010 in Singapore. The fair aims at stimulating and presenting novel approaches within BBS and jointly between the different branches of Bayer. It was very interesting to get insights into such a huge company and to see from what directions innovation is approached, be it making interaction with customers and traveling easier using tablets, data integration and e-learning.
My keynote was a very high level introduction to the Semantic Web. I introduced the basic ideas and motivation behind it, compared it to the existing Web, introduced RDF (very briefly, based on the excellent slides by Fabien Gandon http://www.slideshare.net/fabien_gandon/rdf-in-a-nutshell-v1), and gave lots of examples. I also spend two minutes on the real time Web, another important development of the current Web to be aware of.

While authoring my talk, I thought hard about an example that shows the relevance of Semantic Web technology in such a business scenario. I finally came up with the following: the management gives you the goal to increase sales by 20%. How can you do that? Imagine you have a dashboard where you can easily integrate different datasets, e.g., marketing budget, sales force, data about hospitals, etc. Your dashboard also has a nice visualization that gives you clues and makes predictions where to invest.

In my slides, I made a quite ugly mock-up what the above scenario can look like. Imagine my surprise when I saw that at two of the booths people were actually working on such dashboards. And indeed, one problem they were facing was the data integration. Data had to be provided in a standard format, using predefined tables etc. Exactly one shortcoming Semantic Web technology is addressing.

Find my slides below, but keep in mind that a huge part was quite interactive with live demos.

The Potential of Web 3.0

View more presentations from Carsten Ullrich.

Charter of Human Rights and Principles for the Internet

2010-09-15T09:54:00.000+08:00

The Internet Rights and Principles Dynamic Coalition (IRP DC) has published a Charter of Human Rights and Principles for the Internet. Will such a charter really have an effect? Even if no government will follow it, it is important that we rise such issues and make clear that we want them discussed. Thanks for doing this work, IRP DC.
I agree with the overall document, some comments:
"5) Right to Development". I'm not quite sure whether this section is needed. What is the meaning of "a) Enjoyment of all Rights on the Internet". Do "Environmental Sustainability" and "Poverty Reduction and Human Development" really need to stressed in this charter. Of course, I subscribe to these goals, but in what way are they related to the charter?
"7.b. Limitations on Racist Speech and freedom from hate speech" and "9.d. Freedom from Defamation" touch on a very important and similar issue (I wonder why they are separated). Shall we forbid hate speech and defamation in the charter? In the ideal world that the charter advocates, indeed, there is no place for such speech. And while it is difficult to define hate speech on paper, I think in reality it is not difficult to spot. We know it when we see it. It is fine with me to remove such pages. Actually, I would also include animal cruelty here, or better, to make it more general.
"11) Right to Education a. Right to Education on and through the Internet":
"Virtual learning environments and other sorts of multimedia, learning and teaching platforms should take into account local and regional variations in terms of pedagogy and knowledge-traditions". Hm, while this is an important point, I don't think it has to be included in this charter (which is quite long already, anyway). Also while "Open Educational Resources" are of course an important point, I feel it is subsumed in "12 b. Freedom from Restrictions of Access to Knowledge by Licensing and Copyright". Actually in that paragraph, I feel that the mention of a specific license model ("Creators and users should use licensing models such as Creative Commons") is inappropriate. The charter aims at identifying universal points, valid for more than a few years. Mentioning a specific license model waters the charter down in my view.

China graduates 12 times as many engineers as the United States. So what?

2010-09-03T10:44:00.000+08:00

The article in IEEE Spectrum "Outsourcing's Education Gap" is a good reminder to put China's as always impressive numbers into perspective:

First, the numbers shrink when taking a closer look:

Popular reports of India and China graduating 12 times as many professional engineers as the United States are way off. Considering only four-year professional degrees, the researchers found that the United States awarded 137 437 compared to 112 000 in India, while China reported 351 537, using broad criteria.

Then, elite status university in China is still on a different level than in the US and Europe:

The two countries have excellent institutions producing world-class engineers. Only the top 3 percent scorers in the national entrance exam get into the elite Indian Institutes of Technology. About 40 universities in China enjoy elite status thanks to the government’s 985 Project, which funneled billions of yuan to establish research centers and improve education. China awards the second-highest number of science and engineering doctorates in the world and is rapidly catching up with the United States, according to the National Science Foundation.... However, says Wadhwa, these elite institutions’ graduates are no better than average American graduates from his own university, Duke. And as you move down the ladder from these top establishments, education quality declines steeply.

And again, a typically China problem, where the party has to control everything:

In China, where nearly all the universities are public, a top-down push for ”mass education” in 1999 traded numbers for quality. The number of engineering bachelor’s degrees awarded jumped by 225 percent in a decade. ”In many cases they didn’t add enough teachers or the teachers weren’t qualified, or the funding didn’t increase for all schools and universities equally,” Wadhwa says. ”The government wanted more engineers, so schools gave them engineers…even technicians were called engineers.”

Talk Announcement: Computer-Supported-Collaborative Learning Scripts for Knowledge Convergence

2010-08-31T16:45:00.002+08:00

It is my pleasure to announce a talk on Computer-Supported-Collaborative-Learning by Prof. Weinberger from Saarland University. Everyone is welcome to attend.
Location:

SJTU e-learning Lab
浩然大厦6楼
华山路1954号
Haoran Building, 6/F
1954 Hua Shan Road
http://maps.google.com/maps/ms?ie=UTF&msa=0&msid=103700951278915477597.00048166e58a9d8fcfe24

Date/Time: Monday, Sep, 6th, 2010, 13:30.

CSCL Scripts for Knowledge Convergence
Do learners in Computer-Supported Collaborative Learning (CSCL) environments share and converge with respect to their knowledge by learning together? Sharing knowledge is a central concept in the area of CSCL, suffering, however, from two problems. First, research on knowledge convergence currently suffers from a lack of systematic conceptualisation and thus, operationalization, of the convergence construct. Second, studies show, that cognitive convergence in terms of sharing knowledge after collaboration is typically surprisingly low. When and how is knowledge convergence then the „engine” of collaborative learning?
In this talk, the first problem will be addressed by conceptualizing and operationalizing different types of knowledge convergence before, during, and after a CSCL experience, namely knowledge equivalence, shared knowledge, and transactivity. Knowledge equivalence refers to learning partners being similar with regard to the extent of their individual knowledge, e.g., having the same grades in a specific test. Shared knowledge refers to learning partners actually having the same knowledge, e.g. being able to solve the same knowledge tasks. In addition to analyzing knowledge levels, the transactivity approach suggests analyzing the degree to which learners refer to and build on others’ knowledge contributions.
With respect to the second problem of learners hardly attaining knowledge convergence, the instructional approach of collaboration scripts will be presented. Collaboration scripts are instructional interventions that specify and cluster learning activities, organize them in roles and assign and sequence these roles in groups. Collaboration scripts may guide learners across different learning arrangements including individual learning phases, small group learning, and activities in the classroom and beyond. Collaboration scripts have been found to strongly influence processes and outcomes of CSCL and may pose an instructional approach to facilitate knowledge convergence. Results of an empirical study will be presented showing how a specific script fosters knowledge divergence during and knowledge convergence after a CSCL session.

The speaker: Armin Weinberger is University Professor of Educational Technology and Knowledge Management at Saarland University,

Germany

.
Armin studied educational science M.A. (subsidiary subjects: organizational psychology and intercultural communication) at the UT (Erasmus studies) and the Ludwig-Maximilians-Universität München,

Germany

(LMU Munich), and received a PhD in 2003 at the LMU Munich on the topic ‘Scripts for computer-supported collaborative learning’.

Starting out with his professional career in 1999, Armin was designing, teaching, and evaluating online courses on empirical research methods and knowledge management at the LMU Munich, the Virtual University of Bavaria, and Siemens Qualification and Training.

From 2000 to 2004, Armin investigated argumentative knowledge construction in computer-supported collaborative learning in a DFG project (Deutsche Forschungsgemeinschaft) with Prof. Dr. Frank Fischer and Prof. Dr. Heinz Mandl at the LMU Munich.

From 2004 to 2006, Armin was leading the European Research Team CoSSICLE (Computer-supported Scripting of Interaction in Collaborative Learning Environments) in the EU Network of Excellence Kaleidoscope at the Knowledge Media Research Center (KMRC), Tübingen,

Germany

, and gave introduction lectures to educational psychology at the University of Tübingen.

From 2006 to 2008, Armin was again working at the Department of Psychology at the LMU Munich, as academic councilor and study advisor giving seminars and lectures on empirical research methods and intercultural aspects of learning and instruction, including a research visit in 2007 at the University of Jyväskylä,

Finland

From 2008 to 2010, Armin was Associate Professor at the Chair of Instructional Technology, Faculty of Behavioral Sciences, University of Twente (UT), The Netherlands, working on the 7^th framework project SCY.

Disney English in Beijing/Shanghai

2010-08-30T08:35:00.000+08:00

The Economist published an article on Disney English in Beijing and Shanghai (the first two cities where they have opened schools). You see the advertisement quite often in Shanghai, so it is interesting to read how they are doing:

They started not so long ago: "Thousands of Chinese children have signed up for Disney’s schools since the first one was opened in October 2008."

"Tuition is $1,800 a year". For an average Chinese, that is a lot of money. For a rich Shanghainese, it is not so much.

Slow expansion due to staff: "The main constraint is not customers—the older schools already have waiting lists—but training and staff. New schools must therefore be in places where they can easily be supervised". Yes, finding good staff is a big problem in China.

Very longterm and risky product: "The initial development costs, which Disney has not disclosed, must have been huge. Within a decade the programme will have a material impact on Disney’s results, predicts Andrew Sugerman, who runs it". That is quite risky as you never know with what regulation the Chinese government will come up next to protect its indigenous market.

But the potential is high, as always in China with its gigantic numbers: "Studies commissioned by Disney estimate that the market for children’s English-language education in China is growing by 12% annually and will reach $3.7 billion by 2012. That may be too modest. Adele Mao, an analyst at OLP Global, a research and consulting firm, reckons the market is already nearly $6 billion a year and is growing by 20%."

Photo by jplust

Highly recommended reading about China's Research and Development

2010-08-27T14:21:00.000+08:00

The U.S. Chamber of Commerce has published the highly interesting report "China’s Drive for 'Indigenous Innovation' - A Web of Industrial Policies".
It contains a detailed description on the politics behind China's Research and Development sector: We "attempt to explain the details of China’s still unfolding indigenous innovation industrial policies while placing them in historical, political, social and economic context."

Some excerpts and summaries:

China's plans for R&D are described in "The National Medium- and Long-Term Plan for the Development of Science and Technology (2006-2020)" (MLP)
Re-innovation (what a nice word) is a must: "The report states: One should be clearly aware that the importation of technologies without emphasizing the assimilation, absorption and re-innovation is bound to weaken the nation’s indigenous research and development capacity.'"
Why is R&D so important: "But the wide wealth gap between the cities and countryside, epidemic pollution, endemic corruption, several hundred protest demonstrations on any given day and an uncomfortable dependency on imported raw materials were just a few of the things that could keep the leaders awake at night. They heralded 'scientific development' as the elixir for China's structural problems and the main theme of their administration". (This also explains why you see the term "scientific" coming up over and over again, often in connotations that sound weird for Western readers)
Bringing back talent is difficult: "The “100 Talents Program” of Chinese Academy of Science brought back 778 foreign scientists between 1998 and 2004, but fewer than half had doctorates and almostnone had tenured appointments abroad. Some top level scientists were lured back, but most of them kept positions in both places, maintaining foreign university tenure while taking advantage of Chinese government funding and facilities for research projects. China was simply unable to bring back top talent due to uncertainty about academic freedom, the ability to conduct quality research, and weak IPR protection."
During the drafting of the MLP, conflicts arouse between those who favored peer review and smaller projects and those who preferred so-called "megaprojects". The conflicts were "solved" by censoring: "Overseas Chinese scientists appealed to senior leaders to avoid heading down the megaproject path. In the summer of 2004, a group of 11 ethnic Chinese scientists who were members of the Society of Chinese Bio-scientists in America wrote an open letter to Premier Wen saying that the big biology projects in the plan would stifle competition among scientists and hamper the prospects of genuine innovation. Nature magazine followed with a special Fall 2004 issue containing a collection of essays from prominent Chinese scientists, from inside and outside of China, which criticized the draft plan for giving bureaucrats too much power over scientists. Published in Chinese and English, the essays argued that the power to distribute research money should be taken away from MOST and funneled through peer review organizations. They argued that if megaprojects remained the central focus, money would be allocated to mediocre projects based on personal connections instead of pursuing real science. In one article, two US-based ethnic Chinese neuroscientists joined China’s senior life scientist from CAS to suggest that the power of MOST over research funding should be reduced, and perhaps the ministry should be disbanded altogether. When Chinese newspapers and magazines jumped into the debate, MOST and its supporters convinced the government’s General Administration of Press and Publications to ban distribution of the Nature supplement, and the Party propaganda department warned editors to drop the topic and avoid playing into the hands of “foreign forces”.
Megaprojects prevailed: "The most ambitious components of the plan are 16 megaprojects. They are vehicles for an import substitution action plan aimed at creating Chinese indigenous innovations through “co-innovation” and “re-innovation” of foreign technologies supplied by companies seeking to profit from the massive government outlays on the megaprojects."
And measures were taken to make sure they were taken seriously: "To ensure that the megaproject plans didn’t get lost in the Chinese bureaucracy, Party leaders assigned responsibility for the 99 supporting policies to ministers, vice-ministers and other senior officials by name and with deadlines attached... This unprecedented high-level hands-on micromanagement demonstrates that the indigenous innovation program is the government’s highest strategic economic priority. But it also should remind international government leaders and foreign technology company executives that the same ministers they meet in Beijing for friendly trade talks are also directing plans for creating Chinese technologies and companies to replace them."
Big winners are selected by the government, actual R&D doesn't matter so much: "Many SOEs [State-Owned Enterprises] are flush with cash, especially several dozen largest SOEs that have been chosen to become national champions and currently enjoy monopolies or controlled competition. When the plan was unveiled in 2006, the Party mouthpiece People’s Daily complained that SOEs “are not taking research and development seriously” and noted that some 75 percent of SOEs didn’t even employ a single R&D staffer."
Protectionism via regulations: "With the government controlling standards, certification and testing regimes can be formidable tools for protectionism. Under the indigenous innovation campaign many restrictions, requirements and constantly changing standards have been coming out of China’s national and local regulators. A key piece of this is the Chinese Compulsory Certification system which issues the “CCC Mark” of safety approval for China’s tech and industrial products. The CCC system is estimated to affect some 20 percent of US exports to China. In almost all cases, the products are already approved by qualified international organizations. But China requires redundant testing by Chinese government-owned labs and recertification, which often starts with the foreign manufacturers paying international travel expenses for Chinese inspectors to visit their factories. Incredibly detailed requirements often delay foreign products as Chinese competitors capture the market. For example, cosmetics companies must submit each new shade of lipstick or nail polish for months of separate testing and certification."
How to encourage patents: "Patent filing is part of SASAC’s [State-Owned Assets and Supervision Management Commission] performance evaluation for SOEs. Local governments provide instant profits for companies by giving subsidies to pay for patent filing costs that often exceed budgeted costs."
And in consequence: "So nearly three-quarters of Chinese patent filings were in the “junk”category."
Re-innovation can be successful: "In 2004, foreign wind turbines had a 75 percent market share in China. By 2009, the three largest domestic players, Sinovel, Goldwind and Dongfang alone had 60 percent of the market -- and the foreign share was down to 14 percent."
Distrust of Chinese scholars trained in the West: "Chinese scientists with many years overseas are sometimes viewed as a threat by less-experienced but influential government scientists as well as by the MOST and NDRC bureaucrats who want full control of the scientific projects that they fund."

Interesting view behind the scenes. I don't have to personal experience to judge whether everything is correct (and keep in mind the report was written for the Chamber of Commerce), but those parts related to research feel right.

Amazing technology: livescribe smartpen

2010-08-18T10:06:00.000+08:00

It's difficult to describe what this amazing piece of technology can do. Image a pen recording audio while you are writing, and then enabling to playback what you have written in sync with the audio, not only playback but also free navigation.
Here is an example of a lecture recorded with the tool:

Lesson 98 Teaching Text Geometry
brought to you by Livescribe

If you don't have sound, let it load a bit, and close and reopen the site. Then it should work.

In Robert Scobble's video interview with the creator of the pen, you will see some more examples of what it can do.

Particularly fascinating are the applications: for instance, you draw a piano and then play on it.

A very interesting piece for mobile learning that is not a phone. I'm looking forward to the next generation with Wifi!

Foursquare Blockage in China as an Example how to Game the Censors?

2010-06-08T15:34:00.000+08:00

The Chinese Firewall blocked Foursquare because a large amount of users check into Tian'anmen Square on the 21 anniversary of the crackdown. I wonder whether this is an example of how demonstrations can work in the digital age, especially in dictatorships that are crazy about controlling attention.

My observation is that by shutting down access to a service, its users wonder what happened and asks questions: attention is drawn to the event the shutdown was supposed to suppress. I noticed this clearly last year, when on the 20th anniversary of Tian'anmen all major forums and Web 2.0 sites in China were down for maintenance. When I asked one of my students about this, he said he nearly forgot about that event, but the shutdown made him remember.

Now to Foursquare. Because a few users checked into Tian'anmen Square, all users in China could suddenly no longer access it, and I suspect that quite a few became aware again of the crackdown. Exactly what a demonstration should result in: a few people take action, a lot are affected and some of those start to think about the issue.

So, let's see whether we will see a similar game next year. Ideally, like this year, another newly popular service gets selected, people check-in, upload, share, or whatever is done on that service. It gets blocked, people pay attention. Targeting a Chinese service would create more attention, of course, but might be more difficult since content creation is often already controlled. But taking down more and more Western services creates attention, too.

What Chinese Students Want to Learn: Independent Thinking

2010-06-04T10:48:00.001+08:00

Chinageeks has translated an article from China Youth Online that discusses a survey on the topic China’s university education system. The results of the survey shed light the problems Chinese student see regarding Chinese higher education.

When beginning to study, students are motivated, but this quickly wears off when confronted with the lectures:

At a Henan province teacher’s college […] most first-year students begin with a good first impression of their school. “The campus environment is very clean, and the facilities are excellent. The classroom buildings and dormitories are all brand new,” says Zhao Jie, a local student. However, since she started her classes, Zhao Jie has only become more and more depressed as time has gone on. “Many teachers [do their jobs] half-heartedly, they rarely interact with us. They don’t have any personal charm or charisma […] My classmates are drowsy in class, and often fall asleep en masse. We’ve gradually stopped going to class as we feel it’s a waste of our time. We’re better off finding an internship.”

Students are looking for role models, but their teachers cannot live up to these expectations:

“Some teachers are too occupied with finding money-making opportunities [outside of school]. [As a result], they don’t seriously pursue [continuing education in] their field of expertise […] Many students get together and compare the salaries of their internships. They’re too concerned with how much money they’ll make in the future, and very few put any heart at all into their studies,” said Renmin University of China student Tu Lingbo. “The impetuous and rash spirit of the whole of society and utilitarianism has long spread into [the university system].”

The Chinese party controls all aspects of the university system:

Yang Deguang, former headmaster at Shanghai Teacher’s College, and current Vice President of China’s Committee on Higher Education, was recently interviewed by China Youth Online. During the interview, Yang explained the government’s absolute administrative rule over the university system. [According to Yang,] the Communist Party has absolute authority over the management, economics and evaluation of schools, nominating school administrators, assigning budgets and expenditures, opening classes, and conferring degrees [….] The survey also revealed that 72.2% of respondents believe that higher education’s [bias to Party political lines] is obvious. These respondents believed this to be the university education system’s second biggest problem.

What should Chinese learn during their education? Being able to think independently comes first, being able to learn comes second. Only comes domain knowledge. Earning a degree comes far behind:

[Finally], the survey also asked the public what they believe university students should get out of their education. First is the ability to think independently (78.2%). Second, learning fundamentals (58.1%). Third, knowledge specific to their major (54.6%). Other important traits to be learned include: building strong character (49.2%), the ability to live on one’s own (36.9%), professionalism (37.0%), meet friends (22.1%), earning a degree (14.8%), and prospects for a good future (14.8%).

The public thinks that in order for the students to exploit their potential in life they need to be able to think independently and continue to learn. But the current system doesn't teach them these skills. Sadly, this part of the questionnaire is slightly confusing. I assume that "the public" means the interviewed, that is the students.

Related to this matter is this article on how the Web censorship in China makes Hong Kong a very attractive place to study for Chinese students. According to the Internet consultant Paul Denlinger, "during the dotcom era [of the 1990s], head-hunters were looking for talent from universities in Shanghai and Beijing. Now they are coming to Hong Kong." The article also gives numbers that show that coming back to the mainland is not attractive. Only 3% of the graduates of the mainland China graduates from University of Hong Kong return to the mainland for a job. On a larger scope, 70% of Chinese students studying abroad do not return after graduation. The Chinese government tries to make the return attractive by privileges, such as better access to medical care. But this does not seem to be what Chinese students are looking for according to the above survey. They don't want privileges, but being able to think independently - something I don't see the Chinese government being favorable of.

A Multimedia Discussion Tool (VoiceThread) in a Cross-Cultural Language Learning Course: Student Feedback

2010-06-03T11:28:00.001+08:00

Recently we explored using a multimedia discussion tool (VoiceThread) in a cross-cultural course, involving students from Shanghai Jiao Tong Distance Education College in China, Monash University in Australia. We used VoiceThread, a Website that allows asynchronous group discussions around one topic or thread. A thread can consist of videos, photos and PowerPoint presentations (presented like the photos as a slide show). Once a thread is uploaded and shared, other users of VoiceThread can either use their keyboards to type a comment; use their microphones to make an audio comment or use a web cam to create an audio-visual comment. Students received the homework to create multimedia documents on any topic. They were also encouraged to comment on their peers' threads.

Our students like this activity and created a significant amount of high-quality documents (only one presentation was clearly plagiarized).

Here are the word clouds results from two open feedback questions: “What do you like about using VoiceThread in the course?” and “What do you not like about using VoiceThread in the course?”. The figures render the students' answers as word clouds in which the frequency of a word is shown with the font size (common words like "a", "the" are removed).

According to the feedback, students highly value being able to communicate ("communication") in "English" with their "friends" (this includes their classmates and the "foreign" students). They find VoiceThread "interesting" and like the multimedia features ("voice", "pictures", "video"). The emphasis of on "make" and "share" indicates that the students like to perform these activities often associated with constructivism and enabled by Web 2.0.

When asked about what they did not like, most complaints were about the slow connection speed of VoiceThread, from the class-room but also from at home and work. While watching the threads was possible most of the time, commenting, especially with voice fre-quently did not succeed. One student stated that the registration was too complicated (even though it required only to select a user name, pass-word and to give an email address).

While VoiceThread had advantages, a very often too slow connection, disappearing comments, incorrect statistics and difficult group management made the learning unnecessarily complicated and sometimes frustrating. An alternative would be to build up a multimedia discussion tool from several Web 2.0 services, such as video sharing sites (YouTube or YouKu, the Chinese counterpart), photo sharing sites (e.g., Flickr), video blogs, etc., and to integrate them into a Personal Learning and Teaching Environment. That is what we are currently exploring in the context of the ROLE project.

The word clouds images were created with Tagxedo, a very helpful service.

We (Kerstin Borau, Daniel Jackson and Scott Grant) submitted a more detailed discussion of the study to ICWL. In case you are interested in more details, let me know and I will post some more.

Collaboration with Google Wave

2010-05-26T15:09:00.005+08:00

I have taken a close look on how Google Wave supports collaborative project management (for a contribution to a research paper). Like all of you, a year ago I was so excited when I received my Google Wave invitation. And like most of you, I found it somewhat... lonely there. I started some Waves with the few people I knew who also had an account, but well, since we had no real purpose on being there, I stopped using it.
In the last days, I used it again and explored what it can do today. I think Google Wave has amazing potential for teamwork. But you really have to use it focused on a task. Just playing around will make you bored very quickly.
A very good introduction into Google Wave is "The Complete Guide to Google Wave".
Below, find the public Wave that contains my contribution to the paper. It is a public Wave, so you can edit it. Please do so and let me know whether you agree/disagree. Can you actually write in it without an account?
Update: Hm, the embedded Google Wave has problems with the Disqus commenting function and appears twice. :(

EU Project "Language Technologies for Lifelong Learning"

2010-05-18T09:27:00.000+08:00

The EU Project "Language Technologies for Lifelong Learning" (LTfLL ) is working on some interesting research. I just read the Project Deliverable Report "LTfLL consortium’s approach to integration" which contains an overview on the goals LTfLL wants to achieve. I was attracted to it because I thought it would focus on language learning, but the project aims at lifelong learning in general supported via language technologies, e.g., LSA. The deliverable lists several interesting services:

"a service that aids the learner and tutor to discover the learner's position with respect to courses and learning objects in a domain of study";
"a service [that] analyzes a learner’s textual learning evidence (e.g., essays, blogs) and generates a topic representation, which shows the concepts and their relationships from the textual evidence";
a service that aims at "giving support and feedback for students who learn collaboratively using chat conversations and discussion forums; assessing and abstracting these collaborative activities that the tutor can then use for grading or supporting the student".
a service that targets "assessing and giving automated feedback for students' written activities".
and several more services.

I wonder to what extent such services can be applied to absolute beginners in a domain, for instance my students who are beginners in French. They cannot really write texts yet, nor have conversations. If LTfLL would grant me a free wish , I would love to have a service that finds texts on their level: I feed it a vocabulary list or the texts from the textbook and I would get texts on that level. However, this does not depend only the vocabulary but also on the grammar level. Texts in past tense wouldn't really help.
Do any of my readers know about such services? Please let me know!

"PLE" for Chinese Language Learning

2010-05-04T10:55:00.000+08:00

I built a Netvibes tab that assembles my favorite Chinese language learning tools (click here to have it set up). It includes:

Google Translate, English/Chinese, both directions;
Google Dictionary for more info about a Chinese word;
nciku mini for looking up a Chinese word and having a list of example sentences;
Pinyin display for Chinese texts from Chinese-tools.

Only Google Translate is a "real" widget, the other three services are included by embedding the whole Web page.
I did not include Anki, my favorite vocabulary trainer, because I mainly use the desktop version and the mobile full page version when I'm learning while being mobile.
Do you have suggestion for other widget/sites to include?

How to add a Facebook Like Button to your homepage, and some semantic info, too

2010-04-28T09:10:00.000+08:00

Facebook just released some interesting features, the universal "Like Button" and, even more interesting, the Open Graph Protocol. They are really easy to use. In this post, I will explain how I added Open Graph information and the "Like Button" to my wife Kerstin's homepage in about 5 minutes.

If you want to follow this example as a tutorial, you need to have access to the HTML source code of the page you want to add the Like Button to.

We have to do two independent steps: describe the entity on the page and add the button.

Describing the entity on the page requires adding information in the header of the html page. The Open Graph Protocol site and Facebook's Open Graph page contain a few examples. In my case, I wanted to describe my wife Kerstin, so I added the following in the HEAD


    ...
    <meta property="og:site_name" content="Kerstin's homepage"/>
    <meta property="og:description"
          content="The homepage of Kerstin Borau."/>
    <meta property="og:title" content="Kerstin Borau"/>
    <meta property="og:type" content="author"/>
    <meta property="og:url" content="http://www.kerstinborau.net/"/>
    <meta property="og:image" content="http://www.kerstinborau.net/images/kerstin_borau2.jpg"/>
    <meta property="fb:admins" content="carsten.ullrich"/>
    ...
</head>

Most items are self-explaining. I describe Kerstin Borau, provide a brief description in natural text and, very important, say what Kerstin is, i.e., specify the type of Kerstin. Here we see one problem of the Open Graph Protocol: the initial set of types is very restricted. A person can be an actor, athlete, author, director, musician, politician or public_figure. Facebook's Open Graph does not know researchers, what a pity. The Semantic Web standards on which Open Graph is based allow to define new types, to exchange the description of types, etc. All that got abstracted away, it seems. So, I decided to make Kerstin an author. All this information can be used by any application, search engine, etc. This is the Open part of the Open Graph protocol. I also specified myself as an admin of the Kerstin Borau entity I described (my Facebook user name in the last item). This is Facebook specific.

Adding the Like button is similarly simple. We need to load Facebook's Javascript library by adding the following piece of code to the HTML page (more details on the Facebook JavaScript SDK page).


<div id="fb-root">
</div>
<script>
  window.fbAsyncInit = function() {
    FB.init({appId: 'your app id', status: true, cookie: true,
             xfbml: true});
  };
  (function() {
    var e = document.createElement('script'); e.async = true;
    e.src = document.location.protocol +
      '//connect.facebook.net/en_US/all.js';
    document.getElementById('fb-root').appendChild(e);
  }());
</script>

In order for this piece of code to run, I need to create a Facebook application id by completing this form. Replace 'your app id' with the id you just created.

The final step consists of adding the Like button. Facebook offers a form to configure it. Add the Javascript code in the location where you want the like button to appear, ideally close to the entity you want to be liked. In my case, I added it next to a photo of Kerstin.

Ok, that's it. What did I do: I described an entity semantically. The good thing is that this part is truly open and not restricted to Facebook. Any application can use this data and try to do something interesting with it. If you want to see what you described, use og:it.

We also told an application to use our information, namely the Facebook Like Button. Now, please be aware that you give Facebook quite some interesting information whenever a logged-in Facebook user visits your page. Facebook knows exactly who visits the page, how long he/she stays, etc. This is a bit different from Google Analytics. This service only knows the IP address of the visiting computer. Facebook knows the name. Of course, if Google where to couple Google Analytics with the Google account, they would have the same information. I don't think they do this, but I didn't read the privacy statement (and there are also ways to uniquely identify a user just based on his/her browser).

Fake news: Determined to fight child abuse, EU will monitor and block Catholic Churches

2010-04-07T09:19:00.000+08:00

The text of a press release we won't get from the EU. I basically took this press release and changed a few words (a more permanent, slightly different version is here). The quote at the bottom is adapted from a newspaper article (information about the original article, German translation).

With this post I don't want to make a statement against prosecuting child pornography nor against the Church. I just wonder why certain measures are announced and taken and others that follow the same logic aren't.

07.04.2010 - For stronger sanctions against child sexual abuse, sexual exploitation and child pornography

Given the vast numbers of recently reported cases of sexual abuse of children by members of the Catholic Church, the European Commission proposes new rules obliging EU countries to impose more severe punishment on those who sexually abuse children. Today's proposal would make it easier to fight these crimes through different tools:

by providing severe criminal sanctions across the EU for sexual abuse and exploitation, as they are serious crimes. New forms of abuse will also be covered, like supporting abuse by not reporting it - specifically covering cases that become known within the Catholic Church but are not reported to the authorities.
Prevention by monitoring church activities. Activities involving children in closed premises within churches will have to announced to and approved by local authorities. Furthermore, these activities will be recorded by technical means, such as a CCTV system in mess halls, priest quarters and confessionals.
Member States will be obliged to ensure that access to churches in which children were abused can be blocked, as they are very difficult to take down at the source, especially if other buildings are standing nearby. The proposal will leave it to Member States to decide exactly how the blocking should be implemented but legal safeguards will always apply.

They will now be discussed in the European Parliament and the EU Council of Ministers and once approved should be translated into national legislations.

European Commissioner Cecilia Malmström stressed the importance of these measures: "If we do not act, we risk that people visiting uncontrolled churches will regard such activities as normal and acceptable.”

OSSI: Open Source Satellite Initiative

2010-04-02T08:49:00.004+08:00

Yesterday I attended a fascinating talk by Hojun Song on his Open Source Satellite Initiative. The goal of this initiative is to enable everyone to launch a satellite by finding low-cost alternatives to the hardware, launch, etc. and publishing the information. A great opportunity for making your physics course interesting.

Hojung Song in a Korean artist with an engineering background. He really knows his stuff and that makes his artistic work so refreshing. Actually, the satellite will be launched this or beginning of next year.

I absolutely love his poster presented at the 2009 cubesat developers' workshop (Click here for the large resolution version.):

It look unusual for a scientific poster at first, but if you take a close look it follows the usual structure: Motivation, Problem, Solution, Outlook. But presented in an amazing way. Ah, in case you wonder: GOD stands for Global Orbiting Device.

Once the satellite is in orbit it will use ultra-bright LEDs to broadcast messages down to earth.

Hojun Song plans to finance the launch costs by selling especially designed T-Shirts. The shop is not yet open, sadly, but the designs he presented look great. I'm looking forward to get one of those.

Follow Hojun on Twitter: http://twitter.com/opensat.

Thanks to Michele Travierso for making this talk happen.

Most Helpful Review Ever

2010-03-23T09:33:00.003+08:00

Honestly, a paper of mine submitted to a conference received the following review:

The author should prepare the final version of the paper as per review instructions:
-the abstract of the paper should satisfactorily show the aims, methods and result of the paper
-the materials and methods described in the paper should adequately support the arguments
-the language used should adequately inform the reader
-the paper should have sufficient length to adequately satisfy its aims
-the graphics used in the paper should sufficiently annotated or captioned
-the references shown should be relevant and cited in the paper

The notification proudly stated that "about 650 papers have been selected for presentation and publication". Well, make that 649.

Please feel free to share, reuse and remix the review.

Picture by alexliivet.

Templates for Usability Testing

2010-03-18T14:38:00.009+08:00

If you need a quick introduction into usability, check out http://www.usability.gov. The site contains very helpful concise information to help you find out whether users actually find your site usable. A clickable diagram illustrates the complete design&testing process. Use the Template section to design your own test (especially the templates in the Usability section). I found the following ones (doc files) the most helpful:

A template for a test plan. Especially interesting are Section 7 onwards, which contain information about Usability Metrics (what to measure) and how to categorize problems.
An instruction sheet that contains information the test facilitator should tell the test participant.
A guide for the test facilitator that contains questions to ask, and forms to fill during the test.
An article about SUS - A quick and dirty usability scale. Page 4 contains a list of questions to ask the test participants.
A template for a short test report for you to summarize what you tested and the results.

The usability tests require you to design scenarios that cover typical use cases. Who is typical user of your site and what is a specific goal they want to achieve. These scenarios are then given to the participant who should use your site to reach the given goal. Scenarios can also be more open: "Now, we’d like to hear about how you might use this site. Please identify an item that you would be interested in finding on this site. Please state your question and then search for the answer. Let me know when you’ve found the information."

Can you recommend other sites with information about usability testing?

Talk announcement: Stian Håklev on "The Chinese Jingpin Kecheng project -- lending and borrowing, or international misunderstanding?"

2010-03-12T16:37:00.005+08:00

Update:

Here are the slides:

The Chinese Jingpin Kecheng project - a story of lending and borrowing, or international misunderstanding?

View more presentations from houshuang.

The audio is also available.

It's my pleasure to announce a talk at the e-learning lab of Shanghai Jiao Tong University:

Speaker: Stian Håklev
Topic: The Chinese Jingpin Kecheng project -- a story of lending and borrowing, or international misunderstanding?
中国精品课程项目 -- 是从国外借鉴的故事，还是国际的误解？
Date and time: Tuesday, 16.03.2010. 15:30-17:00 including discussion

Location:
SJTU e-learning Lab
浩然大厦6楼
华山路1954号
Haoran Building, 8/F changed!
1954 Hua Shan Road
http://maps.google.com/maps/ms?ie=UTF&msa=0&msid=103700951278915477597.00048166e58a9d8fcfe24

Stian Håklev (侯爽) is an open education and open access activist and researcher. He is currently affiliated with the Ontario Institute for Studies in Education, University of Toronto. His work looks at Open Educational Resources (OER) and open teaching and learning from a variety of perspectives, both comparative, social and pedagogical, and is driven by a belief in promoting equal access to education. Stian is also one of the co-founders of the Peer2Peer University.

Abstract:
The Chinese Jingpin Kecheng project -- a story of lending and borrowing, or international misunderstanding?
中国精品课程项目 -- 是从国外借鉴的故事，还是国际的误解？

MIT is world famous for beginning the trend with "OpenCourseWare" -- making all their courses available online for free, and with an open license. Representing just one of many kinds of OER projects, this specific form is the most international, with more than 30 countries joining the OpenCourseWare Consortium. In Asia, there are Taiwanese, Korean and Japanese universities with active OpenCourseWare projects, and national and even regional organizations to coordinate. A classical example of a policy being invented somewhere (in the US) and transferred/borrowed to new locales.

China is similar, and with it's jingpin kecheng project, it's one of the consortium's most active contributors. More than 700 universities have participated in producing more than 12,000 courses in 7 years -- all freely available online. But is this the real story? Has MIT's OCW really had such a strong impact on Chinese education policy? Or is something else going on? And how can comparative education theories help us understand what is going on?

During the talk, I will also briefly introduce the Peer2Peer University -- a platform for collaborative teaching and learning using OER -- and how you can participate, as a student, course organizer, or researcher.

More info about Stian:
http://p2pu.org/users/houshuang
http://reganmian.net/blog
http://twitter.com/houshuang

Google and the Chinese Government: The dog barks, the wall stands

2010-01-15T19:23:00.004+08:00

As you all presumably know by now, this Tuesday Google took an amazing step and announced that they will change their approach to China, and in a first step stop censoring search results on their Chinese Google search . They obviously are aware that "this may well mean having to shut down Google.cn, and potentially [their] offices in China". By the way, they are not announcing that they will stop offering Google services in China as some headlines suggested.

For some Chinese views, check out the posts by Jenny Zhu and Jian Shuo Wang and also ChinaSMACK. Read also "Will Google stand up to France and Italy, too?" by Rebecca MacKinnon, a piece that reminds us that it is not only China who has problems with the Internet.

Just for the fun of it, I will make some predictions on how this issue continues. Will the Chinese Government block all Google Services? Will Google retreat? What about the long term of the relationship between China and the Internet?

Google has kicked the ball back to the Chinese government by announcing that Google.cn will no longer censor it results. To what extend previously removed results already appear on the site is not clear. I was able to perform queries that resulted in images of tanks on Tian An Men, but since I didn't do these searches, say, last week I don't have anything I can compare them to. However, I don't think that uncensoring the index is just a matter of a single switch. I guess over time, more and more "sensitive" results will appear.

But in any case, Google.cn does no longer comply to Chinese law. As a consequence, I am certain that the site will be shut down. However, I don't think that the Chinese government will kick back and block all Google services. This would be perfect ammunition for all those who argue that China is a threat to the Western world. Instead, the usual harassment will continue, getting a bit worse. From time to time www.google.com will not be reachable, as will GMail and other Google services. Secure access to Google Docs doesn't work anyway now, Google Spreadsheet is completely blocked.

Yet, making Google's life a bit more difficult is just for fun. I think that the actual goal of the Chinese government is far more ambitious. In a nutshell, they aim at redefining the standards and protocols of the Internet and the Web itself. The report "Does China Hope to Remap the Internet in its Own Image?" by Richard Winfield and Kristin Mendoza gives a good summary of China's activities towards this goal.

From this perspective, the problems with Google are just a momentary nuisance. Like the Chinese proverb that I just invented says: "The dog barks, the wall stands". Let Google bark a bit. Business with China will continue and enough companies will act like Opera and Microsoft did: quiet cooperation. Why should the Chinese government block access to Google services? This would just cause a stir among Chinese netizen and put a spotlight on the Chinese censorship, an issue that the the government rather prefers to keep outside the news.

Better have a bit of fun and kick the dog from time to time. The possibilities are infinite. Why not block Google's Web crawlers from accessing the Chinese Web? What a perfect way to increase Baidu's international market share. I'm sure "they" will come up with lots of other ideas to sweeten their days.

From Google's perspective, I don't see why the statement that they are willing "to shut down Google.cn, and potentially [their] offices in China" if they have to, should mean that their will disable access to Google services from China. This doesn't make any sense. What would the advantage of Google be if GMail, Google Reader, etc were no longer accessible? The idea of Google using GMail access as a leverage against the Chinese government would imply that they really wanted to start a fight with the Chinese government. This wouldn't do any good for freedom of expression and access to information, goals that Google supports. No, I don't see any danger here.

To summarize, my prediction is that google.cn will be shut down during the next two weeks. Access to other Google services will still be possible, but getting more unreliable. The Chinese government wants to control the Internet and has long ranging plans.

Reshaping the Internet is obviously not a goal that only the Chinese government has. The US has allegedly plans in the drawer to reshape the Internets basic protocols once a digital 9-11 takes place. The EU isn't better, either, see Frances 3-strike model.

Let's be dogs and piss against walls, where ever they are.

Image uploaded to Flickr by universaltheory.

Standardizing the audio industry means taking away control from you

2009-12-14T09:13:00.005+08:00

Lenovo disabled recording your laptop's audio on their newer models. Make no mistake, this is a major limitation in case you want to use your computer for creating - and another example of how your control is taken away from you.
I stumbled on this while I was trying to record my students Skype conversations for later analysis. In theory, it is easy, even without special software. Use the Windows XP audio settings to put the microphone input onto the audio output (thus mixing my students input with their colleagues output). Then use any audio software, e.g., the open source Audacity and record from "Stereo-Mix", which the sound that comes from your speakers.
But I couldn't find the Stereo Mix option on my Lenovo X61s. After some browsing I found this thread, describing my problem in detail. To make a long story short: Lenovo has disabled the option to record what comes out of your speakers. Not only Lenovo, but other manufacturers, too, e.g. Dell. Dell however was able to roll back and re-enabled it after its customers complained. Leonovo, it seems disabled that feature on hardware level. Their official explanation (published only after lots of customer requested an explanation) states that this was

caused by a design change agreement between Microsoft and the audio chip hardware supplier. The new change drops the support of mixed-stereo function and direct playback of microphone. The change in design calls to have these function implemented in application level in future. The goal of this design change is to begin standardizing the audio industry. The design change affected all PC vendors worldwide. In the future, Microsoft and audio vendors expect individual applications developers pick up these functions.

Why would they do such a thing? Well, the only reason why anyone would like to record their computers audio is obviously to illegally copy music. DRM can be circumvented by recording the song while it is played. So, let's disable it. And these guys that use their computer for mixing, podcasting, producing radio show, copying VCR tapes to DVD, making music, or whatever, well, bad luck. You can pay for an "application level" solution if it is ever implemented.
What can we do about this taking away of control? Talk about it is one way of action. That's why I'm posting about this event though it's already old news. Buying hardware that doesn't come with such restrictions. But how can you make an informed decision if you aren't aware of such "features"? Judging from Lenovo's explanation, we will see this more and more often. Let's hope that there will be alternatives available in the future.
The most ridiculous fact is that these measures, just like Internet blocking, can be circumvented. Buy an external sound card or install a virtual audio cable, etc. But this costs time and effort.

Image by siyublog.

Winter Barcamp Shanghai

2009-12-07T19:49:00.007+08:00

Another example why we need clones, now. Three parallel sessions, great discussions all around, information overdose at BarCamp Shanghai! It is so refreshing to see such a crowd of people full of ideas and building great projects. A big mix of male, female, Asian, Western, all open-minded and interested in learning from each other.

Kerstin, Scott and I started the session with a talk on our experiences on cross-cultural learning, done together with Daniel Jackson. Earlier this year, our Chinese students and Scott's Australian students worked together using VoiceThread, a multimedia discussion tool. Students from both sites used VoiceThread for presenting and commenting. Some students did amazing work, see this presentation on Shanzai and this one about the differences between the 70/80/90 generations in China. We were lucky with our audience, Eric Pang from italki and Miles Metcalfe added their views on the topic, and Micah twittered our presentation. Our slides:

Active Learning with the Web

View more presentations from Carsten Ullrich.

Then, David, a student of Scott presented Monash's Chinese island on Second Life. Scott and his team have build an amazing Chinese city, and they showed us the temple, hospital and the airport. Their students are using it for emerged learning, practicing situation where they have to book a ticket, see a doctor, etc. NPC (non-player characters, bots) handle the interactions with the students, freeing the teacher of having to play the doctors/teller roles.

The discussion continued after the sessions and that's why I missed one of the most interesting talks of the Barcamp. Zola Zhou talked about online journalism, barriers and how to overcome them.

Other talks I attended, at least with half an ear:
Maxime Guilbot talked about "Minimum Marketable Features". A valuable collection of tips for start-ups.

Nini Sum presented idlebeat her screen printing studio that creates beautiful handmade T-shirts and Prints.

David Li talked about his project "Scratch for kids", where he and his team teach programing to kids, starting at five. The kids do peer programming (each team consists of a 5 and a 10 year old), each team supported by two tutors. David said that it is amazing to see how the 5 year old kids profit by learning from their peer.

Miles Metcalfe also talked about learning, namely "Learning 2.0 and OpenID". If you though that OpenId is a dry topic, come and listen to Miles. He managed to discuss these issues in a truly humorous way.

Finally, Gang Lu talked about his ChinaMode awards, a grass-root competition for the best Chinese Web sites. Great idea, great concept. Sadly, Kerstin and I had to leave in the middle of his talk, we had to have some hot food, as we slowly froze to death.

In between, Liu Yan presented the idea behind Xindawei, the coworking space in which the barcamp took place. Liu and her partners recognized the need for a place in which people could realize their ideas and within a few months build up Xindawei. Judging from the projects she presented and the barcamp, they have done a terrific job. Check out their Website, if you are in Shanghai. What I find particularly interesting about Xindawei is that they are so successful. Shanghai has created lots and lots of creative centers. Some breathtaking beautiful, some rather ugly, but most of them... deserted. I wonder what Xindawei's magic is, maybe bottom-up vs. top-down?

A big thanks to the organizers Scott Ballantyne, Toffler Niemuth, and Mark Evans. Ah, and for delicious Chinese coffee, check out Flying Turtle!

Let's close with a great quote by Michele Travierso: The only thing I hate about #barcampsh : too much interesting content being displayed at the same time.

Learning about Metacognition/Reflection

2009-12-04T14:12:00.001+08:00

I just read the preprint "Reflection amplifiers in online courses: a classification framework" by Dominique Verpoorten, Wim Westera and Marcus Specht.

The annex is particularly interesting as it contains a very concise overview on 35 approaches on metacognitive/reflection. I am currently teaching French in the SJTU online college and this list gives me some food for thoughts on how I can embed reflection in a meaningful way. Meaningful meaning that my students (and I) actually see a value in it, not just as something that take their time away from learning "the real stuff".

These ones sound interesting (for the full references, please refer to the paper):

Indicators of understanding: Learners are asked to qualify their understanding with simple indicators like "lost/foggy/got it" or equivalent. Stadtler & Bromme, 2008
Formative assessment: The course offers assessment intended to generate feedback on performance to improve, helping learners to assess their own learning. Nicol & MacFarlane-Dick, 2006
Where and Why Is It Wrong? Learners receive pieces of work for which they are asked to say what is wrong and why. Mitrovic & Martin, 2002.
Practice of evocation (pausing to reflect) Learners are requested to recall important or puzzling facts/ideas/concepts from the previous learning episode. de La Garanderie, 1989

The Mitrovic & Martin publications is an example that less able students can profit from being helped to select the next exercise to work on, in this case by showing them the systems estimation about their knowledge state. Something teachers should do, too: explaining their students why they should work on a specific topic, and ideally teaching them this skill.

I am looking for more overviews on metacognition/reflection, in particular for language learning. Any suggestions?

Opera betrays its Chinese Users

2009-11-21T10:11:00.006+08:00

Opera decided to lock out its Chinese users. Until today, Opera Mini was a great browser to use. Fast access to web sites, and the additional benefit for Chinese users was access to blocked sites. Since the traffic runs over Opera servers, they bypassed the Great Firewall. Until today, that is. Now, every user in China is greeted with this message (update: I replace the original screenshot with one I took):

Let me guess what has happened here. The Chinese government has put pressure on Opera to close down that free access. And like most companies, they complied.

And they betrayed not only their Chinese users, but also themselves. Yeah, yeah, Opera, making the Web better, a friend of its users. Plain bullshit, as we can see now.

And you don't even dare to admit it? No press release, no posting in a blog about that you are now blocking usage of the international version to your Chinese usage.

And the worst thing: you even pretend that this is a good thing! You dare to say that this was done for "better browsing experience". Bah, disgusting.

Opera guys, if you were up to your image, then you would first of all inform you users about what has happened. You would inform you users that their data is now accessible to the Chinese government (I'm guessing that this is now the case --- and it makes me wonder what other agencies can access Opera data). Of course you cannot tear down the firewall. My guess is if you hadn't complied than your proxies would have been blocked in China. Well, it was your choice to comply to censorship. But the worst part is that you don't even inform your users.
Update: I'm rather certain that Opera is gagged. But you know, there are sites like Wikileaks. And someone has to speak up, it has gone to far.

Thanks to xlight and mranti for pointing me at this.

Chinese Netizen Compare Chinese and US Education System using Cinderella

2009-11-13T08:32:00.002+08:00

Again a worth reading post from chinaSMACK, this time a comparison between the Chinese and US education system: "Teaching The Cinderella Fairytale: China vs. America". The first paragraphs from both versions:

A Chinese teacher tells the story of Cinderella

Bell rings, students and teacher walk into the classroom. Guess which is which.
Teacher: Which characters did you like in the story? Disliked? Why?
Student: I like Cinderella, and the Prince. I don’t like Cinderella’s step-mother and her step-sisters. Cinderella is kind, adorable, and beautiful. Her step-mother and step-sisters were mean to her.
Teacher: If at midnight, Cinderella didn’t get into her pumpkin carriage in time, what would happen?
Student: She would turn back into a servant with dirty clothes. Aiya, that would be horrible.
Teacher: Therefore, you must be punctual, otherwise you might land yourself in trouble. Also, look around, all of you are very clean and pretty, you need to make sure to be clean or your friends will be scared of you. Girls, you need to be extra careful. When you grow up and go out on a date, if you’re not careful and your boyfriend sees you when you’re very ugly, he might be so scared that he faints. (The teacher pretends to faint, class laughs).
...

Teacher: Greetings classmates.
Students: Greeeee—tiiiinngs—teeeeeacher—(elongated vowels)

Teacher: Today’s class, we’re discussing the story of Cinderella. Did everybody study in advance?
Student: Study? Who doesn’t know the story.
Teacher: Is Cinderella a story from the Grimm Brothers or Hans Andersen? Who was the author? When was the author born? What were the major events in the author’s life?
Student: …..(In whispers) It’s all in the book, can’t you read it yourself?
Teacher: What is the theme of the story?
Student: This must be an exam question.
...

Several sites unblocked this morning UPDATE: NO MORE

2009-11-12T09:25:00.006+08:00

This is my first post since months that I can write without using a VPN. This morning, a lot of sites are suddenly unblocked: blogspot, ow.ly, bit.ly, ... Even Facebook has become accessible even though this changes every two minutes. YouTube works in an interesting way: the Web page is shown, but I cannot watch the videos ("something wrong with the connection"). The secure connection to docs.google.com is also working again.

Let's see how long this last! For the time being, I'm happy.

Update: I received several comments on Twitter (which was accessible, but currently now longer isn't). Most people in China still cannot access these sites. My current guess is that currently CERNET (the Chinese network for universities) might unblock some sites.

The "unblock" now lasts for several hours, starting around 9:30 this morning.

Let's see whether the curtain will drop again or whether the other networks will follow. I wonder whether it has do to something with Obama's visit to China next week? To show him that nothing is blocked here...

Update 2: After one week, the magic stopped, everything is back to normal, meaning blocked.

Lobbying in German Wikipedia

2009-11-09T16:42:00.002+08:00

Currently there is a rather big discussion going on about deleting criteria in the German Wikipedia. It seems that the administrators of the German Wikipedia are quite eager to delete articles that they deem irrelevant. My point of view is rather simple: as there are no page limits in digital media, why not have articles about Herr Müller aus der Moltkestraße 3?

Regardless, what is even more interesting is that these deleting criteria are symptoms of general problems of the German Wikipedia. A rather small, very connected group of admins "rules" Wikipedia and group dynamics seem to be pushing away members with different viewpoint. I'm not involved at all in Wikipedia, but the comments of this article in the excellent German blog "Spiegelfechter" look like they were written by someone who knows this matter. It convincingly illustrates how think-tanks and lobby groups are working within Wikipedia:

4. Thinktanks und pressure-groups haben Relevanz von WP erkannt, und entsprechend dann darin Fuß gefasst, tw. auch “ganz oben”. Sie verhalten sich einfach im Sinne von 1. und 2., machen sich so beliebt, knüpfen ihre sozialen Kontakte, und fallen dann nicht weiter auf, bzw. nur “positiv”. Kommt es in “kritischen” Artikeln dann zu inhaltlichen Diskussionen und Konflikten, wird natürlich ihrem Urteil vertraut, nicht dem der anonymen IP, die etwas anderes sagt. Als “Admin” können sie kritik mit formalen Argumenten abbügeln (”edit-war”), Autoren und Artikel sperren, löschen – es fällt nicht weiter auf. Wer hat als gut ausgebildeter Akademiker eigentlich soviel Zeit, die nötigen edits und Sozialkontakte zum Aufstieg in der WP-Hierarchie zu sammeln, ohne seinen eigentlichen Beruf zu vernachlässigen? Oder anders gefragt – wer finanziert und steht hinter manchen “Hauptamtlichen”? Solche Fragen werden aber unwirsch abgebügelt, siehe Punkt 3.

Summary: Think tanks have understood the relevance of Wikpedia and established themselves by exploiting the group dynamics. They follow the established unwritten rules, establish contacts, behave as wanted by the core group. When discussions arise in specific articles, it is they who are believed, not the anonymous IP. Once they are admins, they can silence critics, close and remove authors and articles.

I think this is highly relevant. Is it naive to believe that new tools will by themselves change the established landscape of power structures. Lobby groups are very able to adapt to these new tools. Again, we are pointed to the fact that we have to able to think critically.

Stimulating Mathematical Creativity in China

2009-09-10T07:12:00.003+08:00

The Chinese school system with its focus on the preparation for the university entrance exams leaves too little place for stimulating creativity and autonomous problem solving. The latest Templeton Newsletter presents an initiative that tries to address these issues in the field of mathematics: the Shing-Tung Yau High School Mathematics Awards. Shing-Tung Yau is a winner of the Fields Medal, chairman of the mathematics department at Harvard, and director of the Center of Mathematical Sciences at Zhejiang University.
From the newsletter:

Modeled after the long-standing Westinghouse competition in the U.S., the Shing-Tung Yau High School Mathematics Awards have been developed with the support of a three-year, $1.2 million grant from the John Templeton Foundation. Unlike other well-known math competitions, the Yau program does not pose questions and ask students to provide pre-determined answers. It is a research-based competition. With the guidance of teachers or mentors, students are asked to come up with an original issue for investigation and then to create the mathematical tools necessary to address their topic.
In its first year, winners for which were announced last December, the competition received entries from every province in China. Yau singles out the achievement of the team from the city of Wenzhou that won the gold medal. He notes that the city is not well known for its higher education and that the students did not have a lot of resources available to them. And yet, he says, they “came up with an outstanding problem and solution in number theory.”

It is a good thing that such competitions make people more aware about the importance of creativity in education. I hope that the initiative does not stop there, though, and helps teachers and schools to learn how to stimulate creativity in class, so that more than a few students benefit. In their resource center, they have links to tools and books; I could not find out whether they go to schools directly.

Citation Criteria: Accessibility

2009-09-05T07:57:00.003+08:00

Even the side-remarks of Steve Talbott's NETFUTURE newletter are worth reading. In the current issue, he motivates the reference section:

With a view toward the needs of the readership, I have preferred to cite review articles, where they are available and, in general, have made little effort to reflect in my citations the priority claims of the various investigators of any particular phenomenon. Public (online) accessibility of papers and ease of access to the relevant information are primary criteria for my selection ...

I like the idea of citing not who claims to be the first who made a discovery but those publications that can be accessed. I'm not convinced that crediting only a single team or person makes complete sense. Discoveries always occur in a context. So, why not use accessibility as a criteria?
By the way, you find all my publications on my homepage ;)
Note: Starting with this post, I'm trying out the comment system Disqus that aggregates comments from different social networking sites. Please leave lots of comments everywhere! H/T to Ignatia Webs.
Photo by eschipul.

Report: Why Children Need to Play in School

2009-08-27T07:56:00.003+08:00

Steve Talbott in his excellent NETFUTURE newsletter mentioned the new Alliance for Childhood publication "Crisis in the Kindergarten - Why Children Need to Play in School". The report provides evidence for what I personally believed all along, namely that children need to play. Instead of making kindergarten a low-level first class that drills reading and mathematics, it is better for the children to create an environment that enables meaningful play.

Some quotes from the preface:

Skepticism about the value of play is compounded by the widespread assumption—promoted by hundreds of “smart baby” products—that the earlier children begin to master the basic elements of reading, such as phonics and letter recognition, the more likely they are to succeed in school. And so kindergarten education has become heavily focused on teaching literacy and other academic skills, and preschool is rapidly following suit. The common misconceptions about young children’s play fall apart when we look closely at what is really going on. We begin to be able to differentiate between superficial play and the complex make-believe play that can engage five-year-olds for an hour or more, fueled by their own original ideas and rich use of language. ...

Research shows that children who engage in complex forms of socio-dramatic play have greater language skills than nonplayers, better social skills, more empathy, more imagination, and more of the subtle capacity to know what others mean. They are less aggressive and show more self-control and higher levels of thinking. ...

Long-term research casts doubt on the assumption that starting earlier on the teaching of phonics and other discrete skills leads to better results. For example, most of the play-based kindergartens in Germany were changed into centers for cognitive achievement during a wave of educational “reform” in the 1970s. But research comparing 50 play-based classes with 50 early-learning centers found that by age ten the children who had played excelled over the others in a host of ways. They were more advanced in reading and mathematics and they were better adjusted socially and emotionally in school. They excelled in creativity and intelligence, oral expression, and “industry.” As a result of this study German kindergartens returned to being play-based again.

I never understood why some people say that playing is a waste of time for kids. If you take a closer look at what and how they play it is amazing and surely very intensive learning.

Picture: Frances Foy: Child Playing, 1934, Smithsonian American Art Museum.

International Conference on Web-based Learning (ICWL) 2009 Trip Report

2009-08-24T15:46:00.000+08:00

Last week Kerstin and I attended the International Conference on Web-based Learning (ICWL) 2009. It was great to be in Germany again and to meet old and new friends. The conference was well organized and located in the beautiful city of Aachen.
Like all e-learning conferences, ICWL is a very broad conference, covering all kind of topics. This can be interesting but also resulted in me not being able to appreciate all of the talks. It left me wondering whether e-learning conferences need to further specialize?
In addition, some talks described systems/ideas without a real evaluation or without ever being used by real learners. Personally, I know how difficult it is to get a system evaluated but for a conference I would expect at least some data that gives an idea about whether the system is indeed useful or how it is used.

My personal highlights, with my cryptic notes:
Klaus Wannemacher. Articles as Assignments – Modalities and Experiences of Wikipedia Use in University Courses
good talk about how Wikipedia is used for teaching/learning and about potential problems.

Best paper award: Riina Vuorikari, Martin Sillaots, Silvia Panzavolta and Rob Koper. Are tags from Mars and descriptors from Venus? A study on the ecology of educational resource metadata
Users interact differently with tags: tag but not search, search but not tag, 1/3 never used tags. Power law distribution for tags.

Felix Mödritscher and Fridolin Wild. Sharing Good Practice through Mash-Up Personal Learning Environments
They describe how to model learning activities based on learning activity theory. Capture how students use the learning activities. Personally, I wonder for what kind of students this system is suited. My students would not be able to use it.

Benedikt Schmidt and Uwe Riss. Task Patterns as Means to Experience Sharing.
Quite interesting talk on work done at SAP. Integrated task patterns in the semantic desktop Nepomuk social semantic desktop.

Erik Duval gave a keynote talk entitled "Snowflake Effect"
Memorable quotes from the talk:
- Problem from the past: scarcity
- Standards/open source is not an excuse for lousy interfaces
- Solution to abundance: massive hyper-personalization
- Advice: remove unnecessary interactions and enable feedback from and between users

My talk about "Microblogging for Language Learning: Using Twitter to Train Communicative and Cultural Competence" was well received. About 20 people attended the session despite having a best paper candidate talk next door. Some good questions were asked afterward, for instance whether we saw any gender-related difference in our updates. Actually we did: the most updates were posted by two male students who engaged in some kind of most-updates-competition. We cannot say anything whether the content of the updates is different based on gender, sadly, we did not store the gender. During my talk I kept an eye on my Twitter stream, and indeed in the Q&A session one (fun) question came up, but also after my talk, some participants used Twitter to ask further questions.

Here are the slides:

Microblogging for Language Learning: Using Twitter to Train Communicative and Cultural Competence

View more presentations from Carsten Ullrich.

Ah, and the next ICWL will take place in Shanghai, probably end of July 2010.

Microblogging for Language Learning: Using Twitter to Train Communicative and Cultural Competence

2009-08-02T18:08:00.003+08:00

Next week Kerstin (my wife and colleague) and I will go back to Germany for a few weeks to visit family and friends, but also to attend the International Conference on Web-based Learning (ICWL) 2009 in Aachen. There, we will present our paper on "Microblogging for Language Learning: Using Twitter to Train Communicative and Cultural Competence".
Here is the abstract:

Our work analyzes the usefulness of microblogging in second language learning using the example of the social network Twitter. Most learners of English do not require even more passive input in form of texts, lectures or videos, etc. This input is readily available in numerous forms on the Internet. What learners of English need is the chance to actively produce language and the chance to use English as tool of communication. This calls for instructional methods and tools promoting ‘active’ learning that present opportunities for students to express themselves and interact in the target language. In this paper we describe how we used Twitter with students of English at the Distant College of Shanghai Jiao Tong University. We analyze the students’ messages and show how the usage of Twitter trained communicative and cultural competence.

To our knowledge, this is the first paper that analyzes in depth how Twitter can be used for language learning. Please read a preprint here.
Sadly, today we can no longer use Twitter in China. It is blocked. A pity how the government's fear of communication destroys learning opportunities. I hope that they will realize soon that by censorship they hurt themselves. But how can I as a German complain. Our government is following these tracks and currently also establishing a censorship infrastructure. Shame on them.

Workshop at the UDS-SJTU Joint Research Lab for Language Technology

2009-07-31T19:00:00.000+08:00

I was invited to give a presentation at a workshop at the UDS-SJTU Joint Research Lab for Language Technology, a joint research lab of Saarland University, Germany and Shanghai Jiao Tong University, China. I gave a brief overview on how we have been building Totuba’s research workspace based on existing services and data.
For the full report on the workshop, please head over to the Totuba Labs page!

Great short film about what it feels like growing up in the Chinese education system

2009-07-28T10:04:00.001+08:00

chinaSMACK posted this link to Water Brain, a short movie (animation) by Johann Poo. Some Chinese commenters remarked that while the quality of the animation and story line does not Miyazaki's work (obviously), the movie very well visualizes what it feels like to grow up in an education system that puts extreme pressure on the children.

Water Brain Complete Edition（16：9） from Johann.Poo on Vimeo.

Also illustrated by this quote (unrelated to the movie):

When she returns home, Guoguo begins writing her homework assignments: English, one hour; mathematics, one and a half hours; physics, two and a half hours. When she finishes, it is already one o’clock in the morning. Monthly examinations are coming soon, so Guoguo has arranged extra review material for herself: Listen for half an hour to English tapes; do the midterm review test in Lively Learning and Cleverly Designed Exercises [Huoxue qiaolian], do the study guide for chemistry; and do the exercises on the angle bisector theorem in Top Student [Jianzi sheng], her mathematics study guide. After she has finished this, Guoguo crawls wearily into bed at half past two in the morning.

Yang Guoguo is a student in the third year of junior high school (ninth grade) in Chongqing Municipality.

Source: Various Documents Related to Chinese Education, Attachment 5, Chinese Education and Society, vol. 39, no. 6, November/December 2006, pp. 45–66. No author given.

Totuba Labs

2009-07-21T11:22:00.000+08:00

Aside from my academic job at Shanghai Jiao Tong University, I am working as a manager on AI and Semantic Web Technologies for Totuba. Totuba is a Shanghai-based start-up company that supports education ventures in China via its corporate consulting services, the totuba.com course search and comparison website, the China Education Blog, and its research and development effort to improve educational tools and practices.
We are working on interesting and valuable project at Totuba and we decided to share our insights, research questions and in the future also our data on thousands of courses (e.g., Mandarin courses, MBA, etc.) with you. Therefore, we have set up Totuba Labs, the new home for discussion of Totuba's research and development activities, where I will keep you updated about our research activities at Totuba.
Looking forward to seeing you there!

Follow up to: VG WORT und Google

2009-07-17T07:11:00.002+08:00

Earlier this year, I expressed my anger about how VG Wort, an organization that represents authors in Germany is still stuck in the last millennium. Well, but surprisingly they answered my email. Since I'm quite sure that only few people there know what RSS and blogs are, and even less know how to be automatically informed when a post on VG Wort appears, and no one would bother commenting such new and senseless stuff, I wrote them a traditional email.
Surprisingly, I received an answer, where they asked for a personal discussion about my view. The discussion then turned into an exchange of two emails, but nevertheless. Sadly, the exchange was confirming what I suspected all along. Paraphrased and translated from German: "Oh no, it is so difficult for us to get in touch with our authors directly. We have committees for that." I was in a friendly mood, so I showed them how to set up a Web form on Google Docs at no cost and 5 minutes effort. That is an excellent example of how Web 2.0 profoundly changes the way you interact with your audience. But, hélas, after this demonstration they never bothered to contact me again. Well, I guess they have to discuss this in a committee first.

Going down now: PLEs and China

2009-06-10T15:30:00.002+08:00

Last week, we in China had to experience a rather bothersome Internet. Due to the anniversary of the Tiananmen demonstrations, the Chinese government went quite far and blocked access to a multitude of sites: Twitter, Flickr, Hotmail, and some more, in addition to YouTube and Blogspot that have been blocked since months now. In addition, thousands of Chinese sites went down for "maintenance reasons". While with some effort (proxies, VPN) the non-Chinese sites could still be accessed, but for the normal user they were gone.

I wonder what this means for PLEs and projects such as ROLE in particular. This is an impressive (I would even say the most extensive) example of the fragility of PLEs: you don't have control over the services. They can shut down or be made inaccessible.
On the other hand, it also shows the power of alternative access and APIs: Twitter itself was blocked but could be accessed indirectly via itweet.net, an independent Web interface. Desktop applications such as Tweetdeck did not work, since they access Twitter's API directly.

Does this suggest that a PLE environment should, like itweet, offer an alternative access to the services it integrates? Or that countries like China are not ready for PLEs?

JavaScript on Mobile Browsers

2009-04-16T11:11:00.003+08:00

Picture by sideshowbarker

I'm now in charge of a project in which we explore how to bring interactive exercises to mobile devices. For this task I first had to get an overview on the extent Javascript is supported on mobile devices. Our goal is to reach the largest audience possible, not only those with an iPhone.

Here are the results of my research done during a productive meeting (productive for me, since all talks were in Chinese and thus I could focus on my work...)

OperaMini
A great target "platform" is Opera Mini. It's free and runs on many many phones. However, its JavaScript support is a bit special. From a dev.opera article:
http://dev.opera.com/articles/view/javascript-support-in-opera-mini-4/
the Opera Mini servers are based on the brand new Opera 9.5 rendering engine, which means that all scripts executed before a page is fully loaded will work as expected when your page is loaded in Opera Mini.
After the page has been transferred to the client, things are a lot more limited - basically all events are processed on the server. The client does absolutely no JavaScript processing at all, and instead the page is kept in the server.

Opera Mini can execute JavaScript on the server if it's triggered by the JavaScript events onSubmit, onChange, onClick in forms and onClick in general. The XMLHttpRequest is supported.

WebKit
Wiki says that "WebKit is an application framework that provides a foundation upon which to build a web browser"
http://en.wikipedia.org/wiki/Webkit

It runs on Symbian S60, iPhone, Android, Palm Pre, and some set-top boxes.

In particular, the Web Browser for S60 has a wide support of industry standards including W3C's HTML 4.01, XHTML 1.0, CSS 1, 2, 3 (partially), DOM 1, 2, SVG-Tiny, and Web standards such as, ECMAScript.
http://opensource.nokia.com/projects/S60browser/

Opera Mobile
Opera Mobile uses the same core rendering engine as Opera Desktop 9.5 and seems to support the complete set of JavaScript features. It runs on Windows Mobile + S60. However, it costs 20$.
http://dev.opera.com/articles/view/opera-mobile-9-5-the-developer-angle/

In Practice
However, in practice, most browsers/frameworks seem to have quite some problems. The
Mobile compatibility tables at quirksmode.org give a good overview on the problems a developer has to expect.
http://quirksmode.org/m/table.html

Conclusion
OperaMini runs on most clients, but JavaScript is non-trivial due to the execution on the server. For less devices, but more powerful JavaScript, WebKit and the browser that support it looks like a good choice.

What is your experience with programming for mobile browsers? I only spend a few hours on this topic and I'm sure I didn't get the whole picture.

Spark09: Summary of an Inspiring Conference

2009-03-31T11:53:00.007+08:00

Last Saturday I attended and gave a talk at Spark09. It was an amazing conference, covering a wide range of topics, focused on igniting ideas. What I particularly liked was that interaction with the audience was enabled by giving the speakers a special place in the foyer (and forcing them to go there after their talk). I was kept busy for several hours after my talk.
I talked about Web technology and learning: ""Video killed the radio star, but will Web 3.0 kill the teacher?", here are my slides:

Video killed the radiostar, but will Web 3.0 kill the teacher?

View more presentations from Carsten Ullrich.

I attended the following talks:
Dr. Daniel Auriel - Bayer "How can nano technology be the biggest thing for our future?". A very interesting introduction on nanotubes and quantum dots. We will see these tools in our future, even though I wonder how to get rid of trash 1000x more resistant than steel.

Michael Kwok - Arup "How can buildings defy gravity?". Arup is the company behind landmarks such as Beijings Bird's Nest and the Watercube. Kwok talked about how the randomness of the Bird's Nest hides the supporting structure and showed us that the pattern on the Watercube is not random at all. I particularly liked how proud he was to have build objects that inspire a future generation of architects.

Tom Stader - Library Project"How can enthusiasm make children smarter? Tom is an amazing guy. Within a short amount of time and with a team of 5 people (including him) he build over 200 libraries in the West of China. Such work is only possible because he reuses structures where ever he can. They collaborate with existing networks, use their transportation and has no problem that other organizations take credit, too, even if it puts his library project in the background. I learned two things from his talk: 1) reuse allows you to set up large projects with little but focused investment, not only regarding Web technology; 2) if your task become more important than your ego, amazing things can happen. So, guys, if you want to donate money, please take a look at the Library Project.

Stephen Protz - Arc8x "How can you go 'Back to the Future' with dirt?" I did not attend this talk as it was parallel to Ton's talk, but a Chinese friend of mine was very impressed with it. It made him understand that if you are rich, then you carry responsibility.

A big thanks to ClarkMorgan for organizing this event.

In German: VG WORT und Google

2009-03-20T08:58:00.002+08:00

Sorry, this post is in German. It is a letter I sent to VG Wort, an organization that represents authors in Germany. The recent Google Book settlement is upsetting them . I know that this is a complicated issue and that the text of the settlement raises concerns, but at VG-Wort they are using the standard vocabulary: "a danger to the economic situation of the authors..." bla bla bla.

Well, I AM AN AUTHOR AND THEY DID NOT ASK FOR MY POSITION! And this makes me very upset. I don't understand why in such an important question they don't even ask for feedback!
Well, I gave them feedback, in German. I don't see Google Books as a threat. I want my stuff to be found easily and to be read easily. People who want to buy the book will buy it. Let's make this as easy as possible. Don't make it hard for my customers, otherwise they will go somewhere else and read/download my book for illegally.
But it looks like they will take the same way as the music industry...
----------------------
Sehr geehrte VG-Wort,

danke für die ausführliche Information zu den (geplanten) Vereinbarungen mit Google. Was mich allerdings unangenehm verwundert, ist die Art in der Sie diese Diskussion führen. Ich verstehe nicht, wieso Sie keine Rücksprache mit denen halten, die Sie vertreten.Es ist heute im Zeitalter des Internets ja nun wirklich kein Problem, eine echte Diskussion mit Ihren Autoren zu halten oder zumindest uns (und nicht nur ein paar Vertreter) nach unser Meinung zu fragen. Bei einer Frage mit einer solchen Reichweite halte ich dies für selbstverständlich!

Nun denn, hier halt ungefragt:

"Die Vergütungsansprüche für Digitalisierungen, die bis zum 5. Mai 2009 vorgenommen werden."
Ja.

"Das Recht, die Entfernung von sämtlichen vergriffenen Büchern zu verlangen. Gleichzeitig soll die VG Wort
das Recht eingeräumt bekommen, digitale Nutzungen von vergriffenen Büchern weltweit für Google (über das
Google-Partnerprogramm) oder Dritte zu lizenzieren, sofern nicht Autor oder Verlag dem widerspricht."
Der Sinn der Einstellung der vergriffenen Bücher ist es ja, diese wieder zugänglich zu machen. Verlage und Verwertungsgesellschaften haben sich de facto um eine große Anzahl von Werken eben nicht gekümmert. Die Zugänglichkeit sollte auch für meine Werke das Ziel sein.
Aber ich verstehe Ihr Ziel nicht. Google soll es zunächst verboten werden, aber dann wieder doch erlaubt?

"Das Recht, die Entfernung von sämtlichen lieferbaren Büchern zu verlangen. Gleichzeitig soll der VG WORT
möglicherweise das Recht eingeräumt werden, Suchmaschinen wie Google die Indexierung von Büchern
(Volltextsuche im Buchinhalt) zu lizenzieren, sofern dem Internetnutzer ausschließlich bibliographische Angaben
und keine Buchinhalte angezeigt werden."
Nein. Mit Verlaub, das ist genau die Denke die u.a. zu den massiven Problemen der Musikindustrie geführt hat. Information will gefunden und gelesen werden. Wird mein Buch nicht gefunden bzw. ist es nicht lesbar weil VG WORT dem widersprochen hat, dann wird der Benutzer ein anderes Buch nehmen oder es sich illegal besorgen. Wenn der Benutzer wirklich interessiert ist, wird er das Buch kaufen. Wenn er nicht interessiert, dann lässt er sich auch nicht durch Einschränkungen dazu zwingen,bzw. er geht dann zu Seiten die das gesamte Werk illegal gratis anbieten. Wir wollen doch den ehrlichen Benutzer nicht drangsalieren und ihm sein Leben schwermachen!
----------------
What do you think about this topic? Do you feel robbed because people can read your book on Google Books?

Some Interesting Conferences in China in 2009

2009-03-19T08:57:00.006+08:00

Here is a subjective list (in no particular order) of interesting conferences and workshops that take place in China in 2009 about Web technology, networks, e-learning, knowledge management.
For a full list of conferences in China, use my Yahoo! Pipe that takes the eventseer.net conference RSS feed and filters by keyword "China".

Two views on Personal Learning Environments: OpenLanguage and ROLE

2009-03-18T08:26:00.005+08:00

Over the last weeks, Hank Horkoff published a series post about a new project of his, OpenLanguage. Hank is one of the guys behind the Shanghai based PraxisLanguage, best known for ChinesePod. They are an amazing team and pushing a lot of innovation into the commercial e-learning world.
OpenLanguage is an ambitious project; Hank's mission statement says:

To create, as a community, the leading international language-learning platform for language classes that brings together the best lessons from publishers and the best study tools from developers.

I find his project particularly interesting since it mirrors some of the goals we identified in our recently started EU project ROLE.
Hanks criticizes the "rigid, stale platforms that typically represent" e-learning and proposes the "Mobile Learning Network" (MLN): through the MLN, "the user is constantly connected to the content, community, and the tools he needs to learn the target language. The result is a ubiquitous, immersive, learning environment over which he has full control".
Now compare this to one of the goals of ROLE:

"ROLE offers adaptivity and personalization in terms of content and navigation and the entire learning environment and its functionalities. This approach permits individualization of the components, tools, and functionalities of a learning environment, and their adjustment or replacement by existing web-based software tools."

Both project aim at giving control to the learners over their learning process and the tools they use.

Hank seems to have the successful model of ChinesePod in mind: mini-lessons are published regularly, a number of collaboration tools facilitate discussion, the MLN is accessible "channel agnostic" and can be accessed via mobile and desktop. An interesting feature, also available in ChinesePod, is called "teacher service". These are not service FOR teachers, such as information about their learning groups (those are subsumed under "administration"), but services BY teachers, such as individual or group support of learners.

In which sense is OpenLanauge "OPEN"? Hank plans to "allows publishers and developers to sell their lesson content and learning tools". Also, an installable client (unclear what this will be) site will be available via GPL. So, here is becomes really interesting and the differences to ROLE become visible. It looks like OpenLanguage will take the Apple iStore model: publishers and developers will go through OpenLanguage to sell their tools. That is not necessarily at bad thing at all, it can help to guarantee quality of both tools and lessons. In contrast, ROLE is designed to be an completely open framework. Third parties that follow the open standards to be developed in ROLE will be able to include their tools in any instantiation of a ROLE learning environment. Students will be able to include any of these tools in their own personal learning environment. Quality control (or in a broader scope, trust) is an interesting research problem in ROLE.

Now, I'm really happy about OpenLanguage and ROLE. Both were born out of the perception of similar needs, but tackle the problem from different directions. I'm sure we (spoken from my ROLE perspective) and the OpenLanguage team can learn a lot from each other.

Spark09 -- Ideas conference on March 28, 2009 in Shanghai

2009-03-17T17:15:00.004+08:00

In Shanghai on Saturday March 28th? Don't miss the Spark09 conference. This whole day event will see a whole lot of interesting talks, covering humanity, environment, business and science. I was given the opportunity to give a talk in the humanity section.
I know some of the other speakers of Spark09. Tom Stader for instance is an amazing guy who is building libraries in the Western parts of China with his library project. I cannot keep up with such work, but I will share my experience about Web and learning under the catchy title "Video killed the radio star, but will Web 3.0 kill the teacher?"
The similar catchy abstract reads:

Technology disrupts. The WWW makes established business models and institutions obsolete, often almost overnight. Nobody foresaw that the proud Encyclopaedia Britannica would be replaced by Wikipedia, a Web site to which everyone can contribute.
But what about schools and education? Will teachers by succeeded by avatars living in virtual worlds? Or will their role change, from an instructor to a moderator, from a know-it-all to a mediator of knowledge?
Web technology, from the collaborative and user-driven world of the Web 2.0 to the Web 3.0, the machine-understandable Semantic Web transforms education. Better be a part of it.

I hope that my audience gets an idea of the new possibilities for learning that the Web offers today and will offer tomorrow. If you can think of some great stuff I absolutely should talk about, please tell me so in the comments.

Literature on Twitter, especially on its usage for learning

2009-03-14T18:19:00.006+08:00

Kerstin and I are completing an article for the International Conference on Web-based Learning (ICWL) 2009 on how we used Twitter in Kerstin's English Listening & Speaking course.
Astonishingly few publications exist on Twitter and particularly Twitter in education. Sure, there are a lot of blog posts about ideas, but only few detailed experience reports with an analysis.
Whatever, I wanted to share with you the related work that I found, so I copied&pasted the section "Related Work". Please tell me what we missed!

UPDATE: I received links to two more publications, see below.

------------------------------------------------------
Related work on micro-blogging covers either general suggestions on how to use Twitter in the classroom or analyses properties of the network and conversations without a focus on learning.
In the former category, general hints and advice for academic usage have been published in several blogs (e.g., [1, 2, 10, 12]), Websites (e.g., http://twitterforteachers.wetpaint.com) and publications [4, 5, 13].
From an analytical viewpoint, the first published analysis of Twitter [8] reports on topological and geographical properties, and also describes the most frequent usage types (daily chatter, conversations, sharing information and reporting news). [9] ana-lyze a broader subset of Twitter users. In addition to information about its base in general and its geographical distribution, they identify different groups of Twitter us-ers which they label broadcasters, acquaintances, miscreants and evangelists based on the amount of followers, friends and update frequency.
The specific kind of communication that takes place on Twitter is analyzed in [6, 11]. [11] shows that communication on Twitter goes beyond updating one’s status messages. Most tweets address specific friends. This is confirmed by [6] who give a detailed analysis of the usage of Twitter as a conversation tool. They find that a high number of updates are indeed conversations, which often span several exchanges: (30% of the analyzed tweets use the @ sign, and 90% of these address a person).
In their Twitter analysis [7] distinguish between followers/followees and “friends”, where a friend is defined as a follower/followee who was personally addressed via the “@” syntax at least twice. They argue that only these contacts can be considered as being within the social network of a user that matters. However, they disregard private direct messages of Twitter, which cannot be accessed via the API.
Probably the only work that analyzes micro-blogging in the context of learning is [3]. They performed a frequency analysis over Twitter updates collected during an in-tensive one week summer school for PhD researchers. The students were instructed to use Twitter a back-channel for communication. In their data the most frequent key-words were directly related to the terms used in the summer school lectures, which indicates that the students were using Twitter as instructed.

UPDATE:
[14] describe how they used a Romanian microblogging service to deliver an online course.
Suggestions for analyzing microblogging are listed in [15].

References

[1 ] Hend S. Al-Khalifa. Twitter in academia: a case study from Saudi Arabia. eLearn, 2008(9):1–1, 2008. doi: http://doi.acm.org/10.1145/1454105.1454109.

[2 ] Carol Cooper-Taylor. 50 ideas on using Twitter for education, 2008. URL http://cooper-taylor.com/blog/2008/08/50-ideas-on-using-twitter-for-education/. This is an electronic document. Date of publication: August 22, 2008. Date retrieved: March 14, 2009.

[3 ] Cristina Costa, Guenter Beham, Wolfgang Reinhardt, and Martin Sillaots. Microblogging in technology enhanced learning: A use-case inspection of ppe summer school 2008. In Riina Vuorikari, Barbara Kieslinger, Ralf Klamma, and Erik Duval, editors, Proceedings of the 2nd SIRTEL'08 Workshop on Social Information Retrieval for Technology Enhanced Learning, Maastricht, Netherlands, September 2008.

[4 ] EDUCAUSE Learning Initiative. 7 things you should know about... twitter. Technical report, EDUCAUSE, 2007. URL http://connect.educause.edu/Library/ELI/7ThingsYouShouldKnowAbout/44762?time=1236935622.

[5 ] Gabriela Grosseck and Carmen Holotescu. Can we use Twitter for educational activities? In Proceedings of 4th International Scientific Conference eLSE "eLearning and Software for Education", Bucharest, April 2008. URL http://www.scribd.com/doc/2286799/Can-we-use-Twitter-for-educational-activities.

[6 ] Courtenay Honeycutt and Susan C. Herring. Beyond microblogging: Conversation and collaboration via twitter. In HICSS '09: Proceedings of the 42nd Hawaii International Conference on System Sciences, pages 1–10, Washington, DC, USA, 2009. IEEE Computer Society. ISBN 978-0-7695-3450-3. doi: http://dx.doi.org/10.1109/HICSS.2009.89.

[7 ] Bernardo A. Huberman, Daniel M. Romero, and Fang Wu. Social networks that matter: Twitter under the microscope. First Monday, 14(1), 2008.

[8 ] Akshay Java, Xiaodan Song, Tim Finin, and Belle Tseng. Why we twitter: understanding microblogging usage and communities. In WebKDD/SNA-KDD '07: Proceedings of the 9th WebKDD and 1st SNA-KDD 2007 workshop on Web mining and social network analysis, pages 56–65, New York, NY, USA, 2007. ACM. ISBN 978-1-59593-848-0. doi: http://doi.acm.org/10.1145/1348549.1348556.

[9 ] Balachander Krishnamurthy, Phillipa Gill, and Martin Arlitt. A few chirps about Twitter. In WOSP '08: Proceedings of the first workshop on Online social networks, pages 19–24, New York, NY, USA, 2008. ACM. ISBN 978-1-60558-182-8. doi: http://doi.acm.org/10.1145/1397735.1397741.

[10 ] Alan A. Lew. Twitter tweets for higher education, 2007. URL http://web20teach.blogspot.com/2007/08/twitter-tweets-for-higher-education.html. This is an electronic document. Date of publication: August 19, 2007. Date retrieved: March 14, 2009.

[11 ] Edward Mischaud. Twitter: Expressions of the whole self - an investigation into user appropriation of a web-based communications platform. Master's thesis, LSE - MEDIA@LSE, 2007. URL http://www.lse.ac.uk/collections/media@lse/mediaWorkingPapers/MScDissertationSeries/Mishaud_Final.pdf.

[12 ] Dave Parry. Twitter for academia, 2008. URL http://academhack.outsidethetext.com/home/2008/twitter-for-academia/. This is an electronic document. Date of publication: January 23, 2008. Date retrieved: March 14, 2009.

[13 ] Vance Stevens. Trial by Twitter: The rise and slide of the year's most viral microblogging platform. Teaching English as a Second or Foreign Language, 12(1):1–14, 2008. URL http://tesl-ej.org/ej45/int.html.

[14] G. Grosseck and C. Holotescu, “Indicators for the analysis of learning and practice commu-nities from the perspective of microblogging as a provocative sociolect in virtual space,” in 5th International Scientific Conference eLSE - eLearning and Software for Education, 2009. http://www.scribd.com/doc/13273385
[15] C. Holotescu and G. Grosseck, “Using microblogging in education. case study: Cirip.ro,” in Proceedings of 6th International Conference on e-Learning Application, 2009. http://www.scribd.com/doc/8551345

Workshop on "Megacities & Learning", Summary

2009-01-20T10:54:00.005+08:00

Last week I've attended a workshop on "Megacities & Learning". I'm posting the other way round, with a previous post discussing some of the results. In this post, I talk a bit about bits and parts of the workshop itself.
The workshop is part of a series on Distributed Learning and Cognition (led by Eric Hamilton); and this most recent instantiation was organized by Anthony Kelly as PI (with much input from Eric Hamilton and David Shaffer). A big thanks to the George Mason University for making this workshop possible.

The temporary workshop Web site is found here. Previous workshops on megacities are found here, together with slides and recordings of the presentations.

Megacities raise problems that require interdisciplinary collaboration. Traffic, pollution, poverty, etc., multi-dimensional problems that interact. Solving these problems requires understanding them and being able to make predictions about the effects of taken measures. One way to do this is via modeling, that is, identifying sub-systems and describe mathematically how these sub-systems work. The subsystems are highly complex, connected, with feedback loops.

Understanding and mastering such models require skills that are not taught in school today. But how can we train people, from city planners to the general audience to cope with the problems raised by megacities?

Richard Lesh talked about his experiences with designing problems that contain the model without telling the model. Let learners find the model on their own. Evaluating the solutions gives feedback to the instructor what his students know. Now for megacities, we need to understand what are the problems, what are the models?

To allow us to get deeper into the topic, Jean-Pierre Bardet gave an talk about megacities in general. You will find his slides in the proceedings Web pages of the previous workshop. I can't do justice to his (nor any of the other talks) and I'll just talk about a few of his points. These are mostly copied from my Twitter stream.

Since 2008, half of world's population lives in urban areas. 2050: 2/3 lives in cities. 1950: about 8 megacities. 2015: about 35. Megacities have more than 10 million inhabitants. Depend on mega-infrastructures so complicated that they are no longer understood. High concentration of people,values,infrastructure. high interconnectivity with region/country/continent/world. For instance, port of Los Angeles is gateway to USA for ports of China. If LA has problems, than will the whole USA. Main Challenges: Pollution, Disasters, Energy/Water, Transportation, Crime.

The complexity of the problems is impressively illustrated by the following example. In the West, we experienced a significant reduction of visible pollution due to particle filters. But finer, invisible pollution particles might be even worse. In some studies even linked to autism. See these slides Constantinos Sioutas for details. So, we might have traded bad visible pollution with worse invisible pollution.

On the second day, Larry Shuman gave a talk about modeling megacities. The interactions in megacities are not well understood, but its components are tightly coupled. Thus, low probability event can result in very serious consequences. There are significant differences between megacities in developed and less developed world, but challenges in both. Megacities are an ideal area for non-linear modeling, where the modeling can happen on various level of detail. Larry then reminded us of the goal of workshop: how to educate people to cope with the challenges of megacities. Today's school textbook do not contain these types of problems, but students must learn to solve non-linear problems with feedback and uncertainty. The public as well must become more informed, especially since the lack of knowledge in the public about complex systems creates a void which can be exploited, e.g., by media.

In the second part of his talk Larry presented his research about disaster planning. Given the fact that man-made and natural events are stochastic and hard to control, those in charge when a disaster occurs require decision-support. Larry's research focuses on simulation systems for decision makers. One system he developed is D4S2 (dynamic simulation model generation). It uses an agent based simulation model to model real-life intelligent entities.

The next talk was by Tschangho John Kim on "Technology, Sustainability and Cities" Tschangho show how to plot socio-economic Systems in 2 dimensions: Centralize vs. Decentralized decision making process in one dimension and Groups vs. Individual Make Decisions on the other dimension. He then talk about his research on the relationship between technology and city development. US Cities in the 19th century unpleasant places to live due to pollution, diseases, overcrowding. In the 20th century, technology made cities more desirable place to live, due to better access and usage of electricity, water, etc. Tschangho shows that the urban growth in the US is result of the land economizing process termed as the "technology-land substitution", a claim that is supported by various tests on the available data about cities. Tschangho then presented his research that makes use of sensory networks that again hopefully make today's cities a better place to live. He show a location-based concierge service that helps to find shops,services taking into account constraints such as budget, shortest path.

The next talk was given by Cherri Pancake on "Cyberinfracstructure for Networks of Scientists". Cherri started by given an example of lack of cyberinfracstructure, in the domain of tsunami research. Until now, the research typically consists of individual effort on a subproblem. The result is that tsunami research had little impact on guidelines and codes, since small experients do not convince policy makers. Cherri's then presented the results of her research in setting up such a cyberinfrastructure for tsunami research. She stressed several times that the key was to identify needs of users, still too often overlooked in software design. I found the experience on data sharing particularly interesting. For my, data sharing is something so obvious that I forgot that there might be other points of view. Cherri clearly showed that it is crucial to motivate need for data sharing. In their project, they established data-sharing policies: participants had to agree to provide data in standard way, with the data becoming public after 12 months. Lessons learned: even people who have clear things to gain from Cyberinfrastructure are reluctant to adopt new technologies. There is a need to incentivize data sharing, biggest barrier is trust.

This concludes my report. Take a look at my earlier post for some of my thoughts on the matter. Just a few sentences on the general problem of which skill we need for understanding megacities and other complex system, and how to develop them in the most beneficial way. Modeling skills in themselves will not result in a better world. They will help us to better understand the world, but such knowledge can be used for purposes that only profit a few and have negative effect on others. The current financial crisis is a striking example. It's old news, but we also need to teach and learn skills like compassion and awareness. These are even more critical, in my view.

I welcome feedback - feedback given now might end up in the workshop report.

New FP7 Projects

2009-01-17T13:21:00.002+08:00

This is just a repost of the TeLearn newsletter (TeLearn - NEWS from the European Commission, technology-enhanced learning research).
My group at the Shanghai Jiao Tong university is a proud member of the ROLE project.

The following technology-enhanced learning projects, resulting from ICT Call 3 and supported with 27,2 mio euro EU-funding, have just started or will be launched in February and March 2009:

COSPATIAL will create collaborative environments to enhance interaction and learning, addressing in particular the needs of children with Autistic Spectrum Disorders.
DynaLearn will build engaging tools for acquiring conceptual knowledge, with the potential to increase students' interest in science studies.
ROLE addresses theoretical models and associated technologies that allow learners to tailor learning environments according to their needs. The results of this research will contribute to improving adaptive and responsive learning environments for the individual learner in different contexts.
The Network of Excellence STELLAR sets out to strengthen the capacity in technology-enhanced learning research across Europe. It will set a new and critical foresight agenda and address the various challenges and degrees of fragmentation that remain in the area.
TARGET will develop a responsive learning system with serious games at its core, that presents the learner with complex situations and results in experiences that are gradually honed into knowledge.
xDELIA will explore novel technology-supported approaches to training and support for non-formal and informal learning tested in the domain of financial decision making. This research will increase the understanding of emotional states in decision making processes and help handling biases.

More information will soon be available in the projects section of the TeLearn website: http://cordis.europa.eu/fp7/ict/telearn-digicult/telearn-projects_en.html

Sedona Workshop on Megacities and Learning

2009-01-16T13:33:00.005+08:00

This week I take part in the Workshop on Megacities and Learning in Sedona, Arizona. The goal of the workshop is to identify research topics and agendas on how to teach the modeling skills that people need to possess in order to cope with the challenges raised by megacities.
Megacites can be defined as as an urban area exceeding 20 million, an urban area whose complexity exceeds management capacity, etc. Megacities are characterized by an enormous complexity: different sub-systems interact, in a non-linear manner, with feedback loops.
How can city planners but also the inhabitants of such cities obtain the skills necessary to make these place a nice place to live?
In particular, this workshop focuses on modelling skills. How to model these enormous entities such that they can be understood and consequences of measures taken be forseen?
The agenda and hopefully soon the participant list can be found here. Basically, there were a few talks and then brain-storming and discussions took place.
Today, each of us was asked to submit his thoughts and ideas. While I was sitting on the hotel veranda, enjoying this beautiful view I came up with the following. Please be kind, it is more a thought collection (especially near the end!) than a revised document. I'm sharing it as I hope to get some feedback and to hear your ideas. During the workshop, I was twittering and got already some feedback from Twitter users gregorylent and rjhinz. I will post a summary of some of the talks a later day.

From my point of view as a Web researcher, being familiar with complex systems not from a modeling point of view, but a more data-oriented view, I see the following areas of research:

Data acquisition
Data sharing
Modeling & Sharing
Visualization

In brief, modeling megacities and learning how to model megacities requires collection of data, be it sensory networks or hard to get sociological data. This data needs to be shared to enable the complex and cross-disciplinary system thinking required by problems of megacities. The data is not a goal in its own, but servers as input to the models. Just like the data, the models themselves should be exchangeable so to increase synergy between researcher groups. Finally, the outputs of the models need to be visualized in a way that the target groups can understand and possibly interact.

In detail
1. Modeling complex systems such as megacities requires massive amounts of diverse data. Data sources include sensor networks (ranging from sensors measuring water pressure to traffic data, cellphone information), but also localized Web data, such as Google searches from a specific region or localized news and blog posts, etc).
Links: Reality Mining ("the collection of machine-sensed environmental data pertaining to human social behavior")
http://reality.media.mit.edu/
http://radar.oreilly.com/2008/05/the-results-of-reality-mining.html
Google search terms are good indicators of flu activity:
http://www.google.org/flutrends/

Data will also come from more "traditional" sources, such as interviews, surveys, etc. Walter (a workshop participant) pointed out that these methods are far more trivial, especially if the interviewed party is afraid that sharing information will result in negative consequences, e.g, illegal immigrants (or whatever the PC term is).

The collected data needs to be stored. Considering the amount of data that megacities will produce, this is far from trivial, but will pose new challenges to the field of Very Large Databases (VLDB) and Cloud Computing.

2. Sharing: In order to produce knowledge, the data must be made available. This again raises several technical as well as societal challenges.
In what format will the data be available? Different data formats, storage mechanisms and semantics make a naive approach unworkable. Life Sciences faced a similar challenge beginning of this millennium. Genome and protein information was available, but advancing the field required integrating information from various sources. These integration challenges were overcome thanks to the development of the Semantic Web, the next generation of the Web, that offers standards for knowledge representation that inherently enable sharing and reuse. Applying the Semantic Web techniques will require developing shared representations of the terms used in the domain of study, that is ontologies for megacities and the different components that make up the concept of a megacity. It will also require mappings between ontologies representing different aspects of megacities.
Equally important as these technological issues enabling sharing, are societal issues. How and to what extend can privacy be enabled and guaranteed? Will the concept just fade away and if so what are the means to avoid that the balance of power completely disadvantages the individual citizen (e.g., a Freedom of Information Act for Data?).
How can sharing be enabled? Cherrie (a workshop participant) pointed out that sharing needs to be motivated. Researchers do not automatically wish to give their data away. What mechanisms can be build to facilitate sharing of data most efficiently? E.g., do we require Creative Common Licenses for data? What would these licenses look like?

3. Modeling. The data will serve as input into modeling tools. I'm not an expert in this area at all, but I think we should pay attention to the following. The Semantic Web research also investigates how to perform reasoning on the semantic data available in the Web. The reasoning is performed by ontological reasoning (OWL), but also but executing rules (SWRL, Semantic Web Rule Language). These rules themselves can be exchanged and re-used. Similar mechanisms should be investigated for the mathematical knowledge underlying the modeling and simulation tools.

4. Visualization. Appealing visualization and intuitive usage is a must if the target audience is different from researchers and society as a whole is to profit from this research program. Investigate to what extent existing software can be reused. Can game engines be used for this purpose? E.g., using the graphics engine to plot the megacity and let the student use the game to explore the model. Requires research in software as a service but in significantly more demanding environments as today, that is a business environment (more static) vs. a gaming environment (very dynamic).

Related to point 3&4: mash-ups. What are the requirements to enable user-driven mash-up of the data and services to build new services.

Additional points to keep in mind:
- Mobile learning: especially in less developed regions, users use mobile phones to access the Web. They do not own a desktop or laptop computer but their mobile device.
http://www.itu.int/ITU-D/ict/statistics/ict/index.html
- a new related research area: Web Science, created by Tim Berners-Lee and others. http://webscience.org/
From the Web site:
"The Web is the largest human information construct in history. The Web is transforming society. In order to...

understand what the Web is
engineer its future
ensure its social benefit

...we need a new interdisciplinary field that we call Web Science.
The Web Science Research Initiative brings together academics, scientists, sociologists, entrepreneurs and decision makers from around the world. These people will create the first multidisciplinary research body to examine the World Wide Web and offer the practical solutions needed to help guide its future use and design."
Face similar challenges (complex system, etc).
-modeling skills in themselves will not make the world a better place. System knowledge can be exploited for personal and group advantage (corruption)

How to detect a XHTML-Voice enabled browser

2008-11-20T21:01:00.004+08:00

I'm working on a small project using XHTML-Voice(X+V), a standard for multimodal applications.
Opera is the only common browser that supports X+V, once you install the voice extensions (Windows only). But then, it's quite impressive. You can let the browser read a page, part of a page, and, if the Web site is voice-enabled, even fill out form and have complete dialogs. X+V is easy to program. If you know HTML, then you can build your own X+V pages.
The problem is that it doesn't make too much sense, since I guess almost no-one uses a voice-enabled browser, YET (more on that below). Thus, I faced the problem that I wanted to present speech buttons only if the user has a X+V enabled browser. I didn't the solution anywhere else (the Wikipedia one is not as detailed), so here it is.
Check whether the X+V mime type is enabled:

var voiceEnabled = navigator.mimeTypes["application/xhtml+voice+xml"]

If the variable is null, then the browser has no X+V support.

X+V does not run on any mobile browser I know of. I guess it is due to the limited resources on these devices, but mobiles are the ideal platform for speech input and navigation. How often did I not Twitter something because it simply would have taken too long. Google thinks similarly and just released a voice search for the iPhone. They do the voice processing on their server. Even though this is only half the story (no speech output), that's the way to go. Symbian anyone?

Pic by Xanboozled.

Open APIs as a threat? And other thoughts on Data Portability

2008-10-19T12:35:00.003+08:00

Data portability is based on the idea that you own the data that you produce in your social networks, in the same way you own the data that your produce using your word processor. Thus, you should be able to move it around as you like, for instance taking your Facebook contacts and use them in Twitter. The video below explains it very nicely:

While listening to the ReadWriteTalk podcast on data portability Marshall Kirkpatrick made some interesting comments that got me thinking.

Open APIs as a means to restrict access (at 17 minutes into the podcast): this sounds strange at first. Having an Open API is one the principles that differentiates Web 1.0 and 2.0, meaning that other sites can access the data on the site in an easy and convenient manner. However, Marschall points at the music industry as an example for offering something while restricting it at the same time. They made music available digitally, in DRM format, that is, digitally protected, and were able to make it illegal to rip your own CD (if the CD has a copy protection). The same danger lies in Open APIs. A site might say: "yes, of course you can download your contact data -- ah, but only the last name and email address" and make other means of accessing the data illegal. The standard way of getting data from a site is screen-scraping, where your program takes username and password, logs in and retrieves the data by reading through the html code, just like you as a human do. I wonder when we will see trials to make screen-scraping illegal, not only if it violates the terms of service of a Web site, but generally.

A pity that this thought was not taken up by the other participants.

Later, Marshall pointed to Facebook Beacon as an example of how easy data portability should be. Facebook Beacon automatically added a users actions on other sites into his activity feed, without any configuration, setup of RSS, whatever. The privacy nightmare of Beacon itself is not the point here, but the extremely easy way how it works from the user site. The user hasn't to do anything! Cannot be simpler.

Two interesting questions were sadly not asked during the podcast. First of all, Angus Logan described how long Microsoft is working in this area and that they have offered solutions for data portability since long. So, why wasn't it more successful? What made it fail and what is the current data portability approach doing better?
And second, data portability rests on the assumption that the users own their social networking data. But do the big players agree? They invest lots of money in the development of their social network site and in keeping them running. If I buy a software and produce something, then definitely, what is produce is mine. But if I use a Web site for free? Is it mine, too? Or only if I pay my monthly membership?

My Big Fat Chinese Wedding

2008-10-12T17:42:00.001+08:00

I wanted to use this blog only for professional blogging, but John from ChinesePod persuaded me to make an exception for this entry. Friday night at the really nice ChinesePod party I mentioned the Chinese wedding I would attend the next day. Nothing so extraordinary, if it weren't for so special circumstances with John found so hilariously unbelievable that he made me promise to blog about it.

How did I get invited to this wedding? We have a Chinese friend, let's call her C who, for some reasons I won't explain here, told her colleagues (all Chinese) that she has a German boyfriend. Then, one colleague invited her to be her maid of honor for her wedding and to bring along her boyfriend. Me being the only German she knows, guess who her choice was. What made things complicated though, was that her colleagues have already seen picture of C together Kerstin and I from a trip to the country side, and she had told them about her happy German couple friends. But no despair. She invented Stefan, the sibling of Carsten, who happens to be live in Shanghai, too.

Thus, so C told me proudly, all I had to do at the wedding was to pretend to be my non-existing sibling, Stefan.

Sounded like the perfect setup for a 1930ies screwball comedy for me. Luckily nothing bad happened. To make a long story short, nobody got suspicious and all I received were some broad hints about how important the wedding is for Chinese girls. Continue reading if you like to learn about what happens at today's Chinese weddings.

Saturday 9:30 in the morning we entered the apartment of the bride's parents. Next to a blinking plastic Christmas tree left over from last year was an aquarium with a gold fish who looked at the outside world with a slightly confused expression in his eyes. Exactly how I felt the whole day. But one of things I learned quickly in China was that sometimes the only sensible thing to do was to relax, go with the flow and everything will turn out all right.

At 10:00 the bridegroom tried to enter the apartment. Seems that in China that is not so easy. First, he had to bribe everyone with hongbao (red envelopes containing money). Then he had to convince his wife to open the door. She made him say "I love you" in more Chinese dialects I thought existed. Finally, after he shouted some poem from the balcony, he could at least enter her room. Then he had to do some other performance that sadly I couldn't see as the room was as crowded as at People Square's station at 18:30. He convinced finally, and the ceremony continued. He and his future wife served tea to her parents and fed each other sweet soup. Afterwards, everyone went to his parents and again the tea and sweet soup ceremony took place. Everything went fine, mostly helped by the fact the almost nobody spoke English and most spoke Shanghainese.

After lunch we went to a hotel south of Shanghai, almost at the seaside. The ceremony (who had to walk where, what to say, when to kiss) was rehearsed once and then started officially, on a nice place outside at a lake shore. I could feel my cultural heritage when the ceremony reached the most important point, the vow. It really felt strange that it wasn't a priest how blessed the couple, but some guy from the hotel in a suit.

Walking back to the hotel I suddenly became encircled by C's colleagues who were on a mission to find out whether I was a guy worthy of C. Where do you work? Oh, a small German computer company, building Web sites. Citing some made up, complicated German name, they were satisfied. Then I was told how important the wedding day is for Chinese women. After they were sure I understood, they wanted to continue their investigation, but C came to my rescue.

Then, the banquet started, accompanied by performance by the bridal couple and some friends. Dancing, singing, speeches, and delicious food. A series of games made sure that every guest did win at least a small price. Really cute. I was awaiting bottles of baijiu (absolutely incorrect translated as "white whine", rather hard liquor), but were lucky, nobody drank this stuff at our table. At 20:00, everything was over. We went back to Shanghai. Since everybody was sleepy, no one was in the mood for another in-depth interview.

Mission accomplished. Let's see whether there will be a sequel for twin brother Stefan .

"Pedagogically Founded Courseware Generation for Web-Based Learning"

2008-09-25T10:14:00.006+08:00

Springer Verlag now published my book "Pedagogically Founded Courseware Generation for Web-Based Learning" in their series Lecture Notes in Artificial Intelligence. Here is the short summary:

Automatic course generation is a very important area of research with numerous practical applications in e-learning. It has been studied since the 1980s within the fields of intelligent tutoring, AI and education, adaptive hypermedia and web-based educational systems. Many approaches have been proposed, but hardly any have resulted in generic and practically applied systems. A number of problems have remained unresolved. These problems are addressed by this work. This book focuses on course generation based on Hierarchical Task Network planning (HTN planning). This course generation framework enables the formalization and application of complex and realistic pedagogical knowledge. The volume describes basic techniques for course generation, which are used to formalize seven different types of courses (for instance, introducing the learner to previously unknown concepts and supporting the learner during rehearsal) and several elementary learning goals (e.g., selecting an appropriate example or exercise). This framework has been implemented and evaluated with good results in several domains, with users from different countries and universities, in the context of an EU project. Course generation based on HTN planning is implemented in PAIGOS and has been evaluated by technical, formative and summative evaluations.

The cover art: the diagram illustrates the basic mechanism of hierarchical planning as used in the book. A complex task is repeatedly broken down into easier subtasks, until primitive tasks are reached than can be solved directly. A very natural way of problem solving.

I'm deeply indebted to Erica Melis and Jörg Siekmann, my former group leaders, for enabling me to do this research, and to the ActiveMath group for providing such a stimulating research environment. And of course to Julita Vassileva, whose work inspired me from my earliest research days onward and who made me so proud when she accepted to contribute a foreword to this book.

Good Introductions into the Semantic Web

2008-09-10T13:07:00.002+08:00

On Monday I gave a lecture at Shanghai Jiao Tong University about the Semantic Web. I replaced a colleague and had 90 minutes to introduce the undergraduate students into this new topic. I based my talk on the slides by Ivan Hermann "Tutorial on Semantic Web". His tutorial covers almost everything and is a good set of slides to select from.
Yesterday I found another set of slides that are also very good introductions and visually extremely well done. I wish I could author such slides. They were made by
Fabien Gandon, researcher at INRIA.
Here is his set about ontologies:

Ontologies in computer science and on the web

View SlideShare presentation or Upload your own. (tags: introduction folksonomy)

And here the one on RDF:

Rdf In A Nutshell V1

View SlideShare presentation or Upload your own. (tags: v1 web)

Take a look at Fabien's other slides on slideshare.

For the course, I also read the following books:

Semantic Web for the Working Ontologist: Effective Modeling in RDFS and OWL
by Dean Allemang and James Hendler. This book is really great to read. I learned a lot about the power of inferencing in RDFS and OWL. It is well written and features good examples. Must-by for everyone interested/working with SW.
A Semantic Web Primer by Grigoris Antoniou and Frank van Harmelen. I felt a bit mixed about this one. While it provides a more throughout introduction into the basics of XML, RDF, etc, not all details were explained and some explanations raised more questions than they answered. I read the second edition and would have expected such issues to be resolved. Still, it is a good and comprehensive introduction into SW and the book's web page has a lot to offer (slides and additional materials).

Chrome's Biggest Feature: "Create Application Shortcuts"

2008-09-03T15:18:00.003+08:00

Since I need to prepare a lecture for next week and I don't have any slides yet, I will waste some time to blog about my 2 cents about Google Chrome. Yesterday I wondered why the comic is targeted at Web developers.

Today I played around with Chrome and suddenly got it. I used one particular feature several times: "Create application shortcut". This creates a link on the desktop (and start menu), which opens the site just like an application. It is opened in the Chrome browser, but has no navigation bar etc. It looks similar to any other application, and links are opened in your browser window, not the application window. While this feature does not make too much sense for a "normal" Web page, it makes perfect sense for Web applications, like GMail. GMail suddenly becomes an application, no longer a Web site. This not a new concept, Mozilla's version is called Prism.

So, what does this mean? Well, actually that users will not download Chrome. They will download an application that uses Chrome. Example: Imagine a button in your GMail Web site that says "Download the GMail application". What users are actually installing is Chrome. A similar link will be in your Google Reader page, but since you already installed Chrome previously, there is no need to download it again, just the desktop link will be created. You as a user won't notice anything. You just have a few nice applications, and you know what, isn't that cool, they download immediately and don't waste any space on your hard disk. Ah, and by the way, you have a nice new browser.

This is one step towards the Web as an operating system. Your Google OS.

And that is why Google needs to target developers: they need to be convinced of using and developing for Chrome and making use of the advanced features that it offers. Since they are already developing HTML/JavaScript applications, there is no effort in coding for Chrome, only if you want to offer things like offline access.

I'm sure these thoughts have been published much clearer somewhere else already, I'm sorry, I had no time to read blogs today and to do the linking, I need to prepare slides for my lecture!

Course on "Connectivism and Connective Knowledge: The Future of Education and Training"

2008-09-02T10:08:00.004+08:00

George Siemens and Stephen Downes will hold a MOOC (Massive Open Online Course) about "Connectivism and Connective Knowledge". The official course description reads:
What will education and training look like in the future? How must we equip ourselves to serve learners in the next decade? Some educators suggest that the "connections that enable us to learn more are more important than our current state of knowing". The networks we create to acquire and manage information are the foundations of learning. This course will explore the concepts of networked learning and suggest ways for educators and administrators to respond. The course will be delivered fully online through live class sessions and discussion forums. Learners will use a variety of different technologies representative of the inherent socially connective nature of learning today. Technologies used during the course will be freely available as open source or open access.

Participation is for free, about 1200 people already registered. Well, that is about the amount of students in several of our courses here at the distant university of Shanghai Jiao Tong University.
Here is the schedule (copied from the course's Wiki page, also available in Chinese):

Weekly Topics & Schedule

Week 1: (September 7-13) What is Connectivism?
Week 2: (September 14-20) Rethinking epistemology: Connective knowledge
Week 3: (September 21-27) Properties of Networks
Week 4: (September 28-October 4) History of networked learning
Week 5: (October 5-11) Connectives and Collectives: Distinctions between networks and groups
Week 6: (October 12-18) Complexity, Chaos and Research
Week 7: (October 18-25) Instructional design and connectivism
Week 8: (October 26-November 1) Power, control, validity, and authority in distributed environments
Week 9: (November 2-8) What becomes of the teacher? New roles for educators
Week 10: (November 9-15) Openness: social change and future directions
Week 11: (November 16-22) Systemic change: How do institutions respond?
Week 12: (November 23-29) The Future of Connectivism

Until it starts, read the official course description and then register by entering your email.
Thanks to Ignatia for her post that made realize that the course is going to start soon.

If you are interested in these topics, then don't miss this opportunity!

Google's Browser "Chrome": First Target Techies?

2008-09-02T09:43:00.004+08:00

Google Chrome By the Google Chrome team, comics adaptation by Scott McCloud
Today, Google will launch "Chrome", a new Web browser (thanks to frankyu for the link). Techcrunch has a detailed description of its features that I won't repeat here (faster, safer, more efficient).
What I find quite interesting is how Google introduces the browser. They hired Scott McCloud, the author of Understanding Comics: The Invisible Art, a must read for anyone interested in literature. McCloud wrote a comic that explains the concepts behind Chrome. This document is clearly not targeted at the main user. It has all the nittygritty details about why every tab should run in its own process, etc. My dad wouldn't care less about processes and garbage collection, actually, I don't really know whether I care. Seems like Google has choosen the strategy to persuade the techies first.
Well, but on the other hand, the browser hasn't officially launched yet. Who knows what other nice comics they prepared.

Mobile Internet Users in China: Some Figures

2008-08-25T16:21:00.004+08:00

For an article on mobile learning I compiled some figures from the CNNIC Statistical Survey Report on the Internet Development in China (published 2008/8/15).

The July 2008 survey reports 84.7 million computers connected to the Internet (including desktop and laptop computers) compared to 592 million mobile phone numbers (growing at a rate of 18%). The mobile access to the Internet is explored by a growing number of users: of the 253 million Internet users in China, about a third (84.7 million) surf the Web with their mobile phone, 22.65 million more than in the first half of 2008. The proportion of desktop Internet users is actually dropping compared to the proportion of mobile netizens.

Babbage & Lovelace: The designer of the analytical engine and its programmer

2008-08-25T09:59:00.004+08:00

Years ago, I took part at an excursion to London about the topic "Victorian London". I gave a talk on Charles Babbage and Augusta Ada King, Countess of Lovelace, the designer of the analytical engine (the first computer) and its programmer.

I just found the slides again and since they are quite nice, I share them:

Babbage & Lovelace: The designer of the analytical engine and its programmer

View SlideShare presentation or Upload your own. (tags: babbage lovelace.)

In the 1840s, Babbage, at that time a famous British mathematician, designed what we would call today the first computer. He met Augusta Ada King, Countess of Lovelace, the Daughter of Lord Byron and Anne Isabelle Milbanke. She understood the potential of his invention and wrote the first published computer program: Instructions on how to calculate the Bernoulli numbers.

Some quotes from her "notes (the most important work describing the Analytical Engine) clearly show that she understood the potential of Babbage's invention:

"The operating mechanism of the Analytical Engine ... might act upon other things ... whose mutual fundamental relations could be expressed by … the abstract science of operations ... . Supposing … that the fundamental relations of pitched sounds in the science of harmony … were susceptible of such expression …, the engine might compose elaborate and scientific pieces of music..."

"... the Analytical Engine weaves algebraical patterns just as the Jacquard-loom weaves flowers and leaves."

"The Analytical Engine has no pretensions whatever to originate anything. It can do whatever we know how to order it to perform."

Due to the limitations of manufacturing techniques at that time and Babbage's "creeping featuritis" (he kept changing the design of the machine), neither the difference engine nor the analytical engine were build.

For more details, please take a look at the slides.

I also took a cool picture of the London Science Museum's replica of the difference engine:

Given a search term, how to find the corresponding entity in DBPedia

2008-08-12T09:25:00.003+08:00

DBPedia represents the content of Wikipedia using RDF and thus makes some of Wikipedia's content understandable by machines. Since Wikipedia covers a vast amount of topics, you can use the URI of DBPedia resources to unambiguously identify entities. http://dbpedia.org/resource/Steve_Ballmer for instance stands for a guy who works for Microsoft.

Sometimes you start with a term and want to find the most adequate resource in DBPedia. For instance, given "ballmer", you want the Microsoft guy. How do you do that?

One way is to use free text search queries using Virtuoso's bif:contains(). But this is a free text search and does not necessarily give you the result you are looking for.

Another way is to build the URI by yourself and check whether it exists in DBPedia. For instance, we concatenate http://dbpedia.org/resource/ with our search term (and replacing any blanks by underscores). The works fine for the term "Steve Ballmer" since the concatenated term is exactly how the entity is identified in DBPedia. For other terms we would need to do more work, e.g., checking for redirects (http://dbpedia.org/resource/Ballmer), manually changing the case (http://dbpedia.org/resource/steve_ballmer does not exist). Especially uppercase/lowercase is difficult to handle since there is no throughout standard for names in Wikipedia and thus DBPedia. For instance, it is http://dbpedia.org/resource/Learning_management_system but http://dbpedia.org/resource/Technology-Enhanced_Learning.

So, my preferred solution is to use the Wikipedia search API. Interestingly, the API offers two search functions:
http://en.wikipedia.org/w/api.php?action=opensearch&format=xmlfm&search=ballmer yields:

<SearchSuggestion version="2.0" xmlns="http://opensearch.org/searchsuggest2">
 <Query>ballmer</Query>
 <Section>
   <Item>
     <Text>Steve Ballmer</Text>
     <Description>Steve Anthony Ballmer (born March 24, 1956) is an American businessman. </Description>
     <Url>http://en.wikipedia.org/wiki/Steve_Ballmer</Url>
     <Image source="http://upload.wikimedia.org/wikipedia/commons/thumb/b/b6/Steve_ballmer_2007_outdoors2.jpg/35px-Steve_ballmer_2007_outdoors2.jpg" width="35" height="49" />
   </Item>
   <Item>
     <Text>Balmer series</Text>
     <Description>The Balmer series or Balmer lines in atomic physics, is the designation of one of a set of six different named series describing the spectral line emissions of the hydrogen atom.
     </Description>
     <Url>http://en.wikipedia.org/wiki/Balmer_series</Url>
     <Image source="http://upload.wikimedia.org/wikipedia/en/thumb/f/f3/Deuterium_lamp_1.png/50px-Deuterium_lamp_1.png" width="50" height="33" />
   </Item>
 </Section>
</SearchSuggestion>

The other query http://en.wikipedia.org/w/api.php?action=query&list=search&format=xmlfm&srsearch=ballmer results in

<api>
 <query-continue>
  <search sroffset="10"/>
 </query-continue>
 <query>
  <search>
   <p ns="0" title="Steve Ballmer"/>
   <p ns="0" title="Mark Lucovsky"/>
   <p ns="0" title="Microsoft Bob"/>
   <p ns="0" title="Lausen"/>
   <p ns="0" title="Fredric Alan Maxwell"/>
   <p ns="0" title="Developer"/>
   <p ns="0" title="Get on Your Feet"/>
   <p ns="0" title="Marc McDonald"/>
   <p ns="0" title="Mark Zbikowski"/>
   <p ns="0" title="Titans of Tech"/>
  </search>
 </query>
</api>

I did not find an explanation of the differences of these two ways to search Wikipedia, but we can spot two main differences: The opensearch one gives more details about the found result (these details sadly disappear if you use another output format than xml, I don't understand why). The srsearch one gives you more results, with less details. So take whichever result you like, grab the title, replace blanks with underscores, add the title to the DBPedia prefix http://dbpedia.org/resource/, et voila, your URI. You might want to check in DBPedia whether there is indeed such a resource, since DBPedia is not updated daily.

Mobile/Pervasive/Ubiquitous Learning

2008-08-06T11:19:00.004+08:00

Just read a well-written and interesting article on mobile learning: "Computer Supported Ubiquitous Learning Environment for Vocabulary Learning Using RFID Tags" by Hiroaki Ogata, Ryo Akamatsu and Yoneo Yano.
In the article, the authors describe TANGO (Tag Added learNinG Objects), a system which uses a PDA and RFID-tagged common objects to practive language learning. The system asks the learner questions ("Where is the cup?") and gives small tasks ("Put the pencil on the microwave"). Since the system knows about the environment, it can judge whether the student gave the correct answer.
It is a good example of what becomes possible if we are able to integrate information about the environment into language learning. Obviously, in the described incarnation, it is not realistically applicable, but still it started to get me thinking about the potential.
What makes this paper also very interesting is the definition of "ubiquitous learning" and the distinction between mobile, pervasive and ubiquitous learning. Based on Lyytinen et. al., 2002, the authors define the different type of learning as shown in the figure:

Traditional Desktop Computer Assisted Learning: Low mobility, not embedded in or using information about the environment.
Mobile Learning: High mobility, but also not using information about the environment.
Pervasive Learning: Uses information about the environment (the devices, objects, etc nearby), but requires heavy setup, thus not mobile.
Ubiquitous Learning: highly mobile. System dynamically knows about the environment.

I wasn't aware about these different characterizations until know. In a more general sense (ubiquitous computing), this view is discussed here: Lyytinen, K. & Yoo, Y. Issues and Challenges in Ubiquitous Computing Communications of the ACM, ACM, 2002, 45, 62-65.

Then, what defines "ubiquitous learning"?

Permanency: all work/learning is permanently and continuously recorded.
Accessibility: access to documents/data/information everywhere.
Immediacy: access to information is instantaneous.
Interactivity: communication with peers/teachers in asynchronous/synchronous manner.
Situated: the environment is taken into account for the learning.

What I like about this definition is that it triggers my thinking. I could explore all these different aspects and think about what possibilities they offer.

Please Help! How to get rid of "Fatal error: Maximum execution time of 60 seconds exceeded"

2008-08-04T15:05:00.008+08:00

I'm using RAP (Rdf API for PHP) for my SPARQL queries. I'm using RAP due to its an excellent documentation, which is much better than related frameworks.
But still, I'm having trouble, unrelated to RAP I think, but due to PHP. Way too often I receive the error message "Fatal error: Maximum execution time of 60 seconds exceeded". How the heck can I get rid of it?
I tried thousands of things:

I edited php5.ini and and increased every time-out value
I did the same with php.ini (why are there two files anyway?).
I edited the HTTP get request of RAP and increased the time-out value (stream_set_timeout($fp, 20000)). See the full code below.
I replaced "fsockopen" with curl.

Here is the current code:

$fp = fsockopen($url['host'], $port);
stream_set_blocking($fp, TRUE );
// changed the timeout
stream_set_timeout($fp, 20000);
$replace = $url['path'];
fputs($fp, "GET ".$replace."?".$url['query']." HTTP/1.0\n");
fputs($fp, "Host:". $url['host']." \r\n");
//fputs($fp, "Accept: application/sparql-results+xml, application/rdf+xml\r\n");
fputs($fp, "Connection: close\n\n");
$buffer = "";
while ($tmp = fread($fp, 1024))
// here comes the timeout error:
{
$buffer .= $tmp;
}
$pos1 = strpos($buffer,"\r\n\r\n");
$pos2 = strpos($buffer,"\n\n");
if ($pos1 === false) {
$pos = $pos2;
} else {
$pos = $pos1;
}

No luck. I still get the error message. Dear reader, can you help? I'm using xampp under Windows XP. Ah, yes, I restarted xampp.
Got it! There are not only two php.ini files but a whole lot of them! I did a grep and edited all of them:
apache/bin/php.ini
php/php.ini
php/php4/php.ini
php/php4/php4.ini
php/php5.ini
Now it works!
Second Update: it seems that editing apache/bin/php.ini is enough!

My problems with the L3S DBLP Bibliography Database

2008-08-04T12:02:00.003+08:00

I'm currently exploring open scholarly data. One big source is the DBLP Bibliography Database hosted at L3S Research Center, which contains a huge set of information about authors and articles in the domain of Computer Science. They offer a SPARQL endpoint, so that you can pose interesting and intelligent queries, in theory.
Why in theory? Well, the big problem is that, in order to save resources, results are truncated. Thus, the following queries yield different results:

SELECT DISTINCT ?title ?uri WHERE {
  ?publication rdf:type foaf:Document.
  ?publication dc:creator ?author.
  ?author rdfs:label "Carsten Ullrich".
  ?publication dc:title ?title.
  ?publication foaf:homepage ?uri.
FILTER (regex(str(?title), "Why"))}

This yields the expected result, an article of mine.

SELECT DISTINCT ?title ?uri WHERE  {
?publication rdf:type foaf:Document.
?publication dc:title ?title.
?publication dc:creator ?author.
?author rdfs:label "Carsten Ullrich".
?publication foaf:homepage ?uri.
FILTER (regex(str(?title), "Why"))}

No results.

Did you spot the difference? The position of "?publication dc:title ?title" is different.

In my view, not showing all results is a serious problem. In some sense, these databases become useless. I can't rely on them and use them only for demo purposes.
Of course, there is a dilemma. Setting up and maintaining these databases costs time and money. I'm really grateful that l3s is doing this task. But still, if not all results are returned, then it is useless for real applications.
I guess the solution would be to download the dataset and setup my own database. But do we really want 300 different instantiations of the DBLP dataset? Hm, what do you think?

By the way, the DBLP RKB Explorer does not truncate results.

Accepted EU proposal "ROLE"

2008-07-24T14:35:00.004+08:00

During the last months, I contributed to a EU FP7 proposal for my lab at SJTU. Recently, we received the good news that our project got accepted (a total of 238 proposals were submitted).
I wrote a press release for our lab:

European Project Proposal of SJTU E-Learning Lab Granted

The E-Learning Lab of Shanghai Jiao Tong University under the leadership of Prof. Ruimin Shen has successfully applied for a proposal in the FP7 ICT EU Research Programme. The project called "ROLE" (Responsive Open Learning Environments) focuses on highly responsive TEL environments, offering breakthrough levels of effectiveness, flexibility, user-control and mass-individualisation. The consortium is led by the Fraunhofer Institute for Applied Information Technology (FHG). Other project partners include RWTH Aachen University, Technical University of Graz, Katholieke Universiteit Leuven, Uppsala University, École Polytechnique Fédérale de Lausanne, University of Leicester, Open University UK, and others. The project, starting in January 2009, will run for 4 years with a budget of 6.6 Mio Euros.

The project will advance the state-of-the-art in human resource management, self-regulated and social learning, psycho-pedagogical theories of adaptive education and educational psychology, service composition and orchestration, and the use of ICT in lifelong learning. Significant benefits arise for learners, their communities, employers, TEL developers and society.

I'm looking forward to contributing to this exciting work!

Google's 3D Chat Lively

2008-07-09T08:37:00.002+08:00

Google has release the 3D Chat/ Second Life rival Lively. Lively is not a single connected world, but consists of rooms created by the users. I played with it a bit, set up an avatar and a room. Explore it here:

15 years ago, me and my classmates passed the Abitur. Time for a class reunion. Since we are dispersed all over the world, why not meet in Lively or Second Life? Lively offers text-chat only, I don't know about Second Life.
Now, what I really would like to see is the combination of 3D images derived from my photos with a chat. This would allow meeting my friends at a beach, in my house, in the Mongolian desert. Chatting in Google Earth would be a nice start.

Two bits from a talk by Tim Berners-Lee

2008-07-08T11:32:00.003+08:00

I finally found the time to watch the event "Washington, Wikipedia, and Web 3.0: What is the Future of the Web?". Organized at the Rensselaer Polytechnic Institute by the Tetherless World Research Constellation (yes, such a thing exists, it's no Star Trek techno babble), it features a talk by Tim Berners-Lee and a podium discussion with Tim Berners-Lee, Jim Hendler, Deborah McGuinness, Wendy Hall, Nigel Shadbolt, and Nova Spivack.

Until now, I only managed to watch the talk by Tim Berners-Lee. He talks about the Web (of course) and how difficult it was for people to understand the potential of the Web in the early days of the Web. He argues that it is the same today with the Semantic Web.
Two quotes from the question&answer session are especially noteworthy: when asked about education & computer science, Tim Berners-Lee talks about the connection between research and education, and comments "as kids become older they should start to connect to [research], so producing an education infrastructure should just blend completely with the research infrastructure at the edges" (1 hour, 3 minutes in the talk). This added a facet to a question I'm thinking about since quite some time, namely whether learning should primarily aim at making the students members of the research community (that how I understand, e.g, Stephen Downes), or whether it isn't sufficient in some cases to teach some facts ("these are some laws of physics, learn how to apply them"). I know people are debating over this since ages, but well, I wanna have my own thoughts.
Coming back to the talk, two minutes later Tim Berners-Lee points out again the fundamental feature that made the Web successful: freedom, especially "freedom to connect", and openness. I sincerely believe that freedom and openness are virtues worth realizing and protecting, so it is always a pleasure to hear examples that show their actual relevance. Here, Tim Berners-Lee takes the mobile Internet as an example. Nobody used it when access was at the service provider's mercy. It only took off when mobile devices became free Internet platforms.

Start Pages for Learning

2008-06-24T07:31:00.002+08:00

Malinka Ivanova has published an interesting paper on using configurable start pages such as iGoogle and Netvibes for learning. She reviews a dozen of these homepages and gives examples of the widgets to use for different types of activities (my personal term for widgets used for learning is studgets). Find Malinka's blog and the slides here.
From my work with start pages, I agree with Malinka's rating of Netvibes among the best. They also offer a quite powerful framework (Netvibes Universal Widget API, UWA) for building one's own widgets and studget. They also have a good mobile interface, which looks nicer than the Google one and, despite the iPhone prefix, works fine with Opera Mini and the native browser of my Nokia N95.

Praxis Language's Personalized Learning System

2008-05-30T14:19:00.005+08:00

Congrats to the team of Praxis Language for the release of the new version of ChinesePod and SpanishPod and the newcomer FrenchPod. For those whose don't know, the Pod family of Praxis Language are among the best sites for language learning. They offer free daily podcasts on various difficulty levels and complementary learning materials on the sites. I'm listening to Ken and Jenny almost daily and their podcast is a significant part of my daily learning dosage.

Ken Carroll used the occasion to blog about the vision behind the platform used for the Praxis Language pods. He calls the new platform "Personalized Learning System, or PLS". According to Ken, the major objective of PLS is "to fit the learning around her own needs (rather than forcing her to conform to some outside requirements)".

Brief description

The lesson view:

I spend the morning with taking a close look at the ChinesePod. Previously, I only used the free podcasts, but intrigued by the "PLS", an area that I'm also doing research about, I wanted to explore its features. And indeed, the platform is impressive. Structured in four main sections ("me", "lessons", "community" and "resources"), it offers a vast amount of utilities for language learning.

The "me" section makes it obvious that the focus of PLS lies on the learner. I can browse lessons, add them to my archive, collect my vocabulary (the feature that I find most interesting, see below), receive counseling (with a "guided" subscription), and post messages.

The "lesson" section displays lessons and comments, the "community" section allows to communicate with peers via conversations, and the "resource" section contains a glossary, information about grammar and pronunciation, and a test. The test helps me to determine my current level, which is such an important feature that I wonder why it didn't make it to the start page. There is quite a lot of functionality to explore, and I can't do justice to it all, so go and take a look.

This set of features makes ChinesePod and it sisters quite unique and a valuable resource for learning. I find the price tag a bit high, $29 for the (in my view) minimum sensible feature set. This is about half of what a book costs.

The current state of the system seems still a bit in development. While there are no real bad bugs, several minor issues need to be improved. For instance, I was not able to save vocabulary via mouse clicks. All terms I collected were wrongly encoded.

Why the PLS is interesting

The PLS offers a wealth of services that I have not yet seen in a non-research Web-based learning environment. I especially appreciate the notion of being able to explicitly collect my vocabulary. Building up a vocabulary is one part of language learning, and the PLS allows me to represent my cognitive state in the system (in a learner model). With one click, I can save the vocabulary of a lesson in my vocabulary set.

In my view, this vocabulary is the most crucial part of the PLS and the one with the most potential. In my view, everything else (the carefully constructed lessons, the grammar explanations, etc) supports my learning of these concepts. The vocabulary terms come closest to what I construct in my mind during learning. Of course, I do not "construct" individual items, but entities connected with each other, forming relationships to other items, to situations where they are used, to dialogues, etc. Representing this information in a computer, in a learner model, is quite hard. The standard technique is to represent concepts, often interlinked (so called overlay user models). Even though this representation is a pale imitation of whatever happens in my mind, it can still be exploited in interesting ways.

The vocabulary list:

In my ideal learning environment, I would be able to click on a vocabulary term and see in which lessons it is trained. PLS does this do a minor extent: not all lessons in which the term occurs are shown (which could be difficult for frequently occurring terms) and it does not work for terms I add manually. In the ideal learning environment, I would be able to train my vocabulary. Again, the PLS does this to some extent with two games.

I want also be able to add any term in my vocabulary. Again, PLS does this for some pages within the site, but not for all. Actually, why restrict that to the site? I might be reading Chinese texts on other Web sites and want to be able to add terms from there. There, a bookmarklet could come handy.

Could we visualize the relationships between terms? I learned term "a" and "b" in lesson "x" (actually in situation "y" that is topic of lesson "x"). Sounds like something that could be shown in a graph. Whether it is useful, I don't know.

This brings up another point: the data collected in the PLS is valuable. Therefore, it should be available. For download for the users (which is implemented, even though the CSV export scrambled the Chinese characters). But also via API to other applications. This would allow third-parties to build applications on top of it. Twitter's success (and it's current downtime problem) is due to the numerous clients and mash-ups that others have buit. Twitter never built a desktop client or a valuable search function. Others did. Imagine what could be done with the PLS data. For instance, cool games (a vocabulary trainer that look-ups images in the Web, comparison of my vocab set with yours, etc).

It the user data were extended a bit, e.g., with attributes such as "mastered", "new", etc, these tools would become even more interesting. APML might also come handy.

Is it a PLE? Or Personalized?

Let's stop daydreaming a bit and come back to the name: Personalized Learning System. Is the name well-chosen? The similarity to "Personal Learning Environment" (PLE) is of course intended. According to Ken, the PLS is consistent with PLE in the sense "to allow the user in every way possible to fit the learning around her own needs". However, a "real" PLE would allow me to include other content sources and services. Currently, PraxisLanguage PLS is similar to any other LMS: its functionality is restricted to whatever a central authority deems important. So, it the PLS is not a PLE. But is it "personalized"?

Wikipedia defines personalization as the "tailoring a consumer product, electronic or written medium to a user based on personal details or characteristics they provide". In the context of the Web, the technical term is "Adaptive Hypermedia". Until now, the PLS is not really personalized, at least in the above sense. The lesson overview does not take my vocabulary or my educational level into account, neither does the community. But I guess these features will come. They offer so much possibilities, for instance using the zone of proximal development to suggest the next lessons I could learn.

Summary: Great Platform for Language Learning

Congrats again to the team at Praxis Language. Their system is one of the most advanced tools for language learning I've seen. And, even better, it has a lot of potential. Just by spending half a day with the system, I could smell dozens of opportunities to facilitate learning even more.

Some time I ago I had the opportunity of visiting the Praxis Langage team (finding them was actually quite a challenge -- I'm glad for them that they moved in a new office). I was deeply impressed. It's not often that you can find so many talented people in one place.

Making Surfing the Web in China a bit Easier

2008-05-25T09:27:00.007+08:00

Surfing the Web in China can sometimes be a pain. Over the last year, inspired by various sources, I compiled my own little no-cost solution of making it a bit easier to access Web sites that certain agencies "harmonized". The following instructions will circumvent the Great Firewall and its Web censorphip. They work for other countries than China as well, of course. They are for Windows XP but should work with Linux, Vista and Mac OS (I did not test it, though. Feel free to send me a Macbook Air and I will).

You need the following:

The Web browser Firefox
The TOR Software. This site is no longer accessible from within China, so use this proxy access. Your browser will complain about a certificate exception, but that is ok for this special case.

Download and install both. Start Firefox. You should now have a little icon in the bottom right. If you click on it, TOR starts and will allow you to access any Web site. Since surfing via TOR is quite slow, let's improve our method and let's tell the browser to use TOR automatically for specific Web sites. Switch off TOR by clicking one more time on its icon.
Download the file proxy.pac and put it in c:\. This file will tell Firefox to automatically use TOR (and other proxies) for some sites. Let's tell Firefox where to find it: Open "Tools" -> "Options". Click on "Advanced" -> "Network". Now click on "Settings" in the row "Configure how Firefox connects to the Internet". In the window that opens, select the last option "Automatic proxy configuration URL and type "file:///C:/proxy.pac" (assuming that you saved the file in that directory). Click on "reload" and "OK".
Firefox will now access most sites directly. However, for some sites it will use proxies and thus circumvent "harmonizers". Let's take a look at the file itself:


function FindProxyForURL(url,host){
 // adapted from http://ya.iyee.cn/labels/proxy.html and
 // http://behindgfw.blogspot.com/2007/03/how-to-use-tor.html

 // use Tor for some sites by using the following proxy:
 proxySite = "PROXY 127.0.0.1:8118";


 if(dnsDomainIs(host, ".blogspot.com")) {return proxySite;}
 if(dnsDomainIs(host, ".wordpress.com")){return "PROXY 72.232.101.41:80";}
 if(dnsDomainIs(host, ".livejournal.com")){return "PROXY 204.9.177.19:80";}


 // be aware that passwords are not safe unless you connect
 // to encrypted sites (using https)!
 // anyone owning the proxy or the outgoing tor server will
 // be able to see your password unless you use https!!!

 if(dnsDomainIs(host, "chinadigitaltimes.net")){return proxySite;}
... [removed some stuff here]
 if(dnsDomainIs(host, "feeds.wired.com")){return proxySite;}
 if(dnsDomainIs(host, "typepad.com")){return proxySite;}

 // otherwise access the site directly
 return "DIRECT";
}

You can add sites that you visit often but which are not accessible. Simply add a line above "//otherwise ..." with your site URL in it. For instance, if you want to add "www.baidu.cn", add the following line:

if(dnsDomainIs(host, "www.baidu.cn")){return proxySite;}

This way, you can add all sites you need and access without clicking any button.

One important point: Never use an important site that requires login and password via TOR or other proxies. This data is not safe and can easily be intercepted by others. If it is just a course blog, that might not be a problem, but don't use it for online banking. The only exception is when the site that you access uses an encrypted connection. In that case, the address starts with "https". Thus: login+password is ok for "https", but not for "http".

Fascinating Mobile Learning Scenario

2008-05-24T10:48:00.004+08:00

A fascinating mobile learing scenario from "Sharples, M. Big Issues in Mobile Learning: Report of a Workshop by the Kaleidoscope Network of Excellence Mobile Learning Initiative. University of Nottingham, 2006."

The history sheet
The history sheet is a foldaway device, able to be easily carried in a rucksack or a shoulder bag. Light and semi-translucent, it can be opened out to form a large ‘window’ through which a large group can view the world. Its aim is to make the intangible tangible, and is deliberately large to support shared group awareness, and to be inclusive. Consider that you’re standing on the shore, overlooking the Solent. The sun shines on the relatively calm sea, light reflecting off the small waves in a shimmering pattern. It was on a day just like this in 1545 that the Mary Rose, a purpose-built warship, rolled over and sank whilst trying to engage the French navy. But on a day like today, it’s just a seascape, and hard to bring to life. Roll out the history sheet, spread it out, and watch the Mary Rose sail into view – see and hear the battle, investigate the theories of how it came to sink – understand how the geography of the Solent contributed to the tactics, understand how small waves could sink such a large warship, and so on. The history sheet is a massive, computationally capable display, which can provide a real-time, interactive, exploratory overlay of information, images, re-enactments and so on over the real world.

I like this scenario. Such a device facilitates making history and other information graspable. Although, on the other hand, I should be able to visualize these events by myself, using my imagination. Just one page earlier, the report discusses potential drawbacks of technology in general: "Technology can deskill users, with old and new skills interfering". Would the extensive usage of such a history sheet contribute to de-training my imagination? I guess, the answer is the same as always: technology can offer tools, it is up to you to use them wisely.

My Homepage is back

2008-05-21T08:36:00.003+08:00

My homepage is now back online at http://www.carstenullrich.net. If you have any links pointing to the prior location, please update (for one last time).

Barcamp Shanghai 2008

2008-05-17T20:03:00.004+08:00

Yesterday I attended the Barcamp Shanghai 2008. Due to some organizational issues, it was announced on rather short notice and a similar event took place at the same time. As a consequence, the Barcamp was less well visited than last year's and mostly only people from Shanghai took part. But still, a big thanks to the organizers, Robert Scales and others whose names I forgot and sadly don't find on the Barcamp Web page.

The organizers took a more structured approach this year, with a set of slots with given themes (Demo camp, Start-up camp) and two free slots.
The demo camp was for start-ups to present their company and collect some feedback. One start-up was Reci-P, a website for recipes. What I find interesting is that they use AI techniques to scrape recipes from other Websites. Is this is legal? Scott, the guy behind the site, adopted a "let's see whether someone complains" attitude. In my view, as long as they put a link back to the original site (and they do), it's ok. It's a bit like Google News.
Another nice start-up is italki (pronounced "I-talk-I"), a collaborative site for language learning. You can find language learning partners, upload and download language learning resources, etc. Some weeks ago, the guys behind italki asked me to provide feedback about their site and I took a close look at it. It is an excellent site if you want to find someone to practice your language skills with and they offer a multitude of services to facilitate communication. If you want to learn a language, that is one of the sites to use. However, you will mostly find other language learners, who most probably are not teachers (although teachers can open their own group on the site). For efficient learning, you still need learning resources (e.g, books) , and what you learned you can pratice it with the language friends at italki.

In the last session I gave a presentation about the Semantic Web. Motivated by Ivan Herman's comment at WWW 2008 that there is a lack of education about the Semantic Web, I mashed-up two slide sets and tried to convey the basic ideas behind the Semantic Web. At one point, I heard one member of the audience exclaiming "aaaaaahhh!!!", and I knew that at least with that guy I was successful. I used the following two slide sets:

Ivan Herman: What is the Semantic Web?
Jamie Taylor, Colin Evans, Toby Segaran: Creating Semantic Mashups: Bridging Web 2.0 and the Semantic Web

The session was quite well visited, 20-30 people, and the audience was really interested. People really want to learn about the Semantic Web. And they are quite astonished to see that the ideas behind the Semantic Web are so straightforward that the Semantic Web concepts are easy to understand.
Update: Attila Gárdos has posted a detailed description of the example by Ivan Herman that I used for the talk. Check it out.

It was an interesting experience to give a mash-up talk, based on slides I did not write myself. The quality of the talk would have been higher if I had used hand-made slides, not because my slides would have been better, but because it would have been my slides. For instance, I didn't understand the purpose of each and every sentence in the slides I used, so I skipped parts or couldn't properly explain everything. But on the other hand, without these pre-existing slides, I wouldn't have given the presentation at all, simply because I did not have enough time. Lesson for the mash-up culture in general? High-quality content may not be achievable by mashed-up data and still requires manual work. But in situations with limited time, reuse can yield sufficiently good stuff to work with.

Stuff I read

2008-05-12T10:19:00.005+08:00

Interesting stuff I read recently:

Benjamins, V. R.; Davies, J.; Baeza-Yates, R.; Mika, P.; Zaragoza, H.; Greaves, M.; Gómez-Pérez, J. M.; Contreras, J.; Domingue, J. & Fensel, D.
Near-Term Prospects for Semantic Technologies
IEEE Intelligent Systems, IEEE Computer Society, 2008, 23, 76-88
A collection of short statements about impact of Semantic Web technologies by various authors, sadly imprisoned behind payed access. My take-away:

Semantics can add value to search by knowing that '“Mr. Brown,” “Gordon Brown,” “the Prime Minster,” or “the PM”' all refer to the same entity. I did not yet see much of that, though. This might also help to bridge gaps between groups (e.g., administration) that use their specialized jargons and ordinary people.
Yahoo is doing quite a lot in SW. They offer tools to developers to add semantics easily. See also Paul Miller's podcast with Peter Mika.
Two types of customers: corporations and consumers. Problem with corporations: "inherently bureaucratic and risk-averse". Advantage with consumers: "thousands of individual consumers are willing to be early adopters".
Context: context as a necessary requirement to better understand queries and needs of the user. Dear reader, if you know any starting point for literature about context, please let me know in the comments.
2008 will be quite an interesting year for SW. Current state comparable with early days of the Web. Now, growth is becoming exponential.
Services: the next big questions will be about services. Once the data is there, how do we make the services available such that they can be accessed easily. My guess is that similar to how the Web 2.0 community is starting to understand the importance of semantic data, they will realize the importance of semantic services. Currently, the mash-ups that are available are nice and they are easy to use for humans. In the future, we will combine the pragmatic Web 2.0 approach and the advanced techniques of the research community and come up with some nice solution to make such services easy to use for machines.

Slides: Creating Semantic mashups: Bridging Web 2.0 and the Semantic Web

Jamie Taylor (Metaweb), Colin Evans (Metaweb), Toby Segaran (Metaweb)
Really cool slides from the Web 2.0 Expo that show the power of Semantic Web data and make the audience understand the step from Web 2.0 to the Semantic Web.

Featured Slide on Slideshare

2008-05-09T19:19:00.003+08:00

That is nice: the slides (see two posts earlier) of my talk at the 17th International World Wide Web Conference have been featured on the SlideShare homepage by their editorial team.
In the slides, I present joint work with Kerstin Borau, Heng Luo, Xiaohong Tan, Liping Shen and Ruimin Shen on "Why Web 2.0 is Good for Learning and for Research: Principles and Prototypes".

My homepage gone...

2008-05-08T08:46:00.002+08:00

In case you are wondering: my homepage is gone. The free Webspace provider Gigacities got a new management and closed down everything without notice. They claim they send a notification, but I didn't receive anything. Well, you get what you pay for... I learned my lesson and will soon set up a new page, not hosted at a free provider.

Trip Report: 17th International World Wide Web Conference, Beijing

2008-05-08T07:47:00.004+08:00

Finally I found some time to put up my notes about WWW2008. All in all, a great trip. I gave a total of three presentations, met interesting people, and the hotel was just next to the Olympic site with an amazing sight on the National Stadium and the Watercube (see below for pics).

Before the conference started, I visited the Chinese Academy of Science. Prof Lu Ruqian invited me to give a talk at the Academy of Mathematics and Systems Science about my previous work (intelligent courseware generation) at the DFKI in the ActiveMath group. Afterwords, I visited the group of
Prof. Shi Zhongzhi at the Intelligence Science Lab of the Institute of Computing Technology. There, I presented a slightly adapted version of my WWW talk. I was deeply impressed by the hard working and bright PhD students at the Academy. I had a chance to spend quite some time with Changlin Wan and Rui Huang and it was really inspiring to see the amount of creativity coming from these students.

Then, on Friday, I had my talk. TI has highly astonished by the high interest in the talk. The room was full, people sitting on the floor. About 150 in total. It seems that our paper was the only one about e-learning that was accepted for the conference. Some good questions ("excellent questions" to quote on of the keynote speakers) were asked about why our usage of Twitter was successful and the problems on depending on third-party services for mash-ups. The slides are uploaded at Slideshare.

| View | Upload your own

The proceedings are available online.

My notes from the conference (mostly unrevised notes taking during the talks):

Major topics:

user generated tags and what to do with them
Semantic Web and Linked Data
social network analysis
rich media: search and usage of media other than text
mash-ups

Context will become a highly relevant topic in the future (eg how to establish and use information about the context of the user to give better search results and recommendations)

First keynote:
Kai-Fu Lee (Google): Title: Cloud Computing
Describes how Google is moving towards and enabling Cloud Computing (which is a fancier word for "grid computing")
"Cloud computing is about opening your browser and staying there."

Panel on "The Future of Online Social Interactions: What to Expect in 2020"
Marc Davis (Yahoo): Communication in the future will be based on a broadcast model and direct communication will be the exception.

W3C Track: Open your data

Linked Data: present data in RDF and use existing URI to link your data to existing data
Tim Berners-Lee: "Linked Data is how the Semantic Web should be. Linked Data is how the Web should be!"

Semantic Web research in China:

China Southeast university is strong in Semantic Web: SW search engine Falcons
APEX group from Shanghai Jiao Tong University also strong.
KEG from Tsinghua University
Dartgrid from Zhejiang University

Track on "Social Networks: Discovery and Evolution of Communities"
Xin Li (Yahoo): Top 2000 popular tags can cover 90% URLs and 95% users (case study in del.icio.us)
Yu-Ru Lin: FacetNet: A Framework for Analyzing Communities and Their Evolutions in Dynamic Networks
Good collection of references on Social Networks Discovery
Jure Leskovec, CMU: Statistical Properties of Community Structure in Large Social and Information Networks

Large scale networks look different from small scale ones.
Large networks have a core-periphery structure. The core consist of 60% node and 80% edges of the network
standard measures to determine communities might lead to communities which actually aren't communities.
Forest Fire model can model large-scale community networks.
dunbar number: a person can maintain social relationship to at most 150 people (for Twitter it's way less, judging from my experience...).

Keynote. Dr. Harry Shum (Microsoft) talks about "Taking Search to New Frontiers"
Not so much new stuff, doing quite intelligent search in videos (spoken text recognition)
Shum: "Semantics is the right way to go in search"
However: here, semantics doesn't mean RDF and friends, but "only" recognizing the entities that the documents are about
His main topics: Move from

content to concepts
users to people
actions to intent

Anton V. Riabov (IBM) Wishful Search: Interactive Composition of Data Mashups
Users compose flows by iteratively selecting tags. Planner than composes the flows.users add tags iteratively

William Chang (Chief Scientist Baidu): Three Generations of Intelligent Search
Third generation search:

personalized search results
integration of search and recommendation
predictive recommendation with feedback

When I talked with him after his presentation about research at Baidu & Google, he complained about the low creativity of the Chinese students who start working at Baidu.

Keynote by Tim Berners-Lee: "The Future of Web Applications"

the importance of Linked Data
Social machines: machines that facilitate social interaction
The Web as an operation system: a packaging system for Javascript that allows for powerful Web applications

These three points don't really give justice to his talk :-) Better visit Paul Miller's blog for more detail.

Third keynote: David G. Belanger (AT&T Labs)
Title: "Three Screens: Evolution of Devices, Services, & Networks"

Small sensors will become more relevant in the future
seamless: change from one device to another while retaining the experience
shared: sharing experiences, interacting with friends that use other devices
interactive: two devices changes the way of interaction: iPhone and Wii
AT&T working on virtual worlds
1 billion devices connected in the AT&T network
in the future: multiple personas per user
convergence will continue forever: existing devices will converge, then a new device will come up, and convergence will happen again
devices other than computers will be much more numerous
10^7 (computer), 10^8 vehicles &handhelds ... to 10^12 consumer items
Bandwith isn't everything. Latency is king (gaming)
most of today's research in next-generation seamless services

Qi Su (IBM China Research): Hidden Sentiment Association in Chinese Web Opinion Mining

hard to relate opinions words to product features
they construct an association set offline and use it for determining what the users are talking about

Xiao Ling, Shanghai Jiao Tong University (APEX): Classifying Chinese Web Pages based on English Training Data

Text classification on Chinese Web pages
learn from labeled English Web pages to classify unlabeled Chinese Web pages
they translate the Chinese pages into English ones and then classify

Ivan Herman (W3C): there is a lack of education in the Semantic Web. The industry is interested, but there is no expertise available.

Panel "Commercializing the Semantic Web"

organized by Talis. They offer a platform for Semantic Web (Triple Store, etc). Free for research and as long as you don't make any money with their platform (then they want a little share). Contact them in case we need such services. They were very much interested in our work.
huge VC interest in SW start-ups

Some additional information about WWW2008 from Graham Webster at sinobyte.

One of the best outcomes was that I finally met the guys from Talis. They showed me an impressive demo of a social network for research. Yawn, I hear? Yet another social network, requiring me to add all my friends/colleagues? Ha, not at all. They mine and compile available information about your research. Your network is there before you even log in. What an impressive illustration about the relevance of open data!!

As promised, some pics from the Olympic site:

Easy Posting from Google Reader to Twine

2008-03-15T10:22:00.004+08:00

Despite what some critics say, Twine might well be the social bookmarking "killer application". At least for me :-) I never got comfortable with del.icio.us and the like, it was simply to hard for me to come up with tags.
Twine makes making sense from bookmarked articles much easier. It parses the documents and recognizes entities, such as name, places, organization. This functionality is similar to Open Calais (see my latest post). But while Open Calais offers a basic service, Twine builds a whole social network around it.
I don't want to review Twine in details since I'm still exploring it. What I have seen until is enough to make me continue to use it. A pity that I do not have any invites to share!
I couldn't find any description of how to submit items from Google Reader with minimal effort, and some may find it interesting:

Most of my Web time I spend in Google Reader, and so I looked for the easiest possible way to push the most interesting feed items into Twine. Twine accepts post send my email, but using the "Mail to" feature from Google Reader requires to manually input my Twine email address, and takes to much time.
The "one-click" solution: use the "Share" feature of Google Reader, whichposts the current item to a RSS feed) and use RSSFWD (no registration required) to convert the RSS feed into emails.
In Google Reader, click on "Your shared items" to get the URL of your Google Shared Feed. Then, input the URL at RSSFWD.
Note that you must tell Twine to accepts emails from RSSFWD. Go to yourTwine settings, and enter "rssfwd@rssfwd.com" under secondary email addresses. The first email you receive for RSSFWD will contain a "confirm" link. Once you click the link, everything is fine (in theory): any item you share in Google Reader will be send to Twine.
In practice, sadly, it worked for a few days, then it stopped. I had to add the feed again.

Do you have similar or better tricks to send content to Twine?

Update: This nice trick works only for the first one who registers the generic email address from RSSFWD. Lucky me!

Extracting Semantics: The OpenCalais Web Service

2008-03-12T12:54:00.003+08:00

Open Calais is an extremely promising Web service by Reuters. Via API, you can submit a document and receive extracted metadata as a result. The metadata captures the topics, persons, places etc the document is about.

Previously, it was accessible only over the API, but now they released a Web interface, which allows to experiment with Open Calais using copy&paste.

As an example I submitted the first few paragraphs of the ReadWriteWeb article discussing Open Calais. The result is shown in the screen shot. Open Calais recognized entities, industry
terms, persons, technology, but also events "Reuters launcing an
API...". Quite impressive.

I tried it with some snippets of my blog, too. Since Open Calais is using existing content for supporting the recognition of, e.g., persons, unknown names are not recognized. From the Web interface I cannot tell whether the specific instance is recognized, e.g., the Carsten Ullrich who is a computer scientist, not the one from social sciences.

License: "basic service is free, ... paid upgrades for advanced features such as guaranteed service levels". "The service stores the provided and derived metadata and makes it available to others". You can only retrieve your content and content you have access to: "you will not attempt to retrieve content by the submission of random keys."

Would you want to build your own service on it? Open Calais sounds like a perfect enabler for semantic applications. It offers a high-level service with great functionality. But it raises a question that we will see more and more often as more of such services will see the light. Will they be here to stay? The software I bought and installed will typically run as long as I like. But a Web service? If I make Open Calais a central part of my (commercial) application, then I have to be sure that it is here to stay. The license agreement states that "Reuters may terminate your access to all or any part of the Calais service at any time, with or without cause, with or without notice, effective immediately." Doesn't sound like something I can completely rely on.

Why Web 2.0 is Good for Learning and for Research: Principles and Prototypes

2008-03-07T11:50:00.000+08:00

I uploaded the final version of my paper "Why Web 2.0 is Good for Learning and for Research: Principles and Prototypes", co-authored by Kerstin Borau; Heng Luo; Xiaohong Tan; Liping Shen & Ruimin Shen, on my Web site. It was accepted for publication at the International World Wide Web Conference in Beijing.

The abstract:
The term "Web 2.0" is used to describe applications that distinguish themselves from previous generations of software by a number of principles. Existing work shows that Web 2.0 applications can be successfully exploited for technology-enhance learning. However, in-depth analyses of the relationship between Web 2.0 technology on the one hand and teaching and learning on the other hand are still rare. In this article, we will analyze the technological principles of the Web 2.0 and describe their pedagogical implications on learning. We will furthermore show that Web 2.0 is not only well suited for learning but also for research on learning: the wealth of services that is available and their openness regarding API and data allow to assemble prototypes of technology-supported learning applications in amazingly small amount of time. These prototypes can be used to evaluate research hypotheses quickly. We will present two example prototypes and discuss the lessons we learned from building and using these prototypes.

If you are interested, read the html or pdf. Looking forward to feedback!

Phun - 2D physics sandbox

2008-02-26T09:08:00.004+08:00

Hu, that is an amazing tool. When I was a kid I tried to build such machines with Lego and Fischertechnik. Doing it in software is nice, but I miss the "touch & feel".

Let's see what tools they come up with "augmented reality".

This (Crayon Physics Deluxe) one looks fun, too: