Leaving Crete

September 21st, 2007 by Martin

DSC01402

It started to rain (well, sort of) outside, and I’m sitting in the lobby of the Kalimera Kriti enjoying the free WiFi :-) It was a great time here in Crete, especially because of the inspiring discussions and good times with interesting people. And not to forget the awesome minigolf session last night (freaky!) I will surely miss the people, the surroundings … and the food ;-)

A lot of material is available about EC-TEL 2007: The wiki with a lot of additional material and presentations, the conference blog, and also photos on flickr. Check it out!

20070919 kreta 002
EC-TEL Science Fiction Session with Hermann Maurer and Bruce Sterling

PS: Join the EC-TEL07 group on facebook!

Evaluating Usability in Technology-Enhanced Learning - 100 Ways Do It Right?

September 20th, 2007 by Eric

I am currently attending several paper presentation during the EC-TEL 2007. Many papers present new tools for authoring, connecting people, aggregating content etc. Most have evaluated their recently developed tools. That is the most interesting part, isn’t it? I have seen evaluating the efficiency, effectivity, usability etc. Especially in the domain of usability evaluation “standards” exist. At least everybody is stating that he is using “the standard”. Even if I thought that I know the standards for measuring usuability, I have to say that in technology-enhanced learning new instruments are said to be a standard instrument. Cool! So I spent some time in page turning the proceedings. My first impression was that usability is often evaluated by using self-developed measuring instruments. Why? Are TEL tools so different from other software? The reason for using standards measure instruments is to compare the usability of TEL tools. Using proprietary things makes it impossible to compare tools.
I am working at a software engineering institute, we mostly use the “technology acceptance model” also known as TAM or the Unified Theory of Acceptance and Use of Technology model also called UTAUT. These scales have been evolved over the years and work fine - at least for us. If I ask people ask the people whether they know TAM or UTAUT, I get the answer “well in fact - no”. One of the speakers uses “SUS” to evaluate the usability. SUS stands for quite and dirty scale for usability evaluation. I had a look at it: the webpage not available anymore. I only find a paper about it as PDF. The scale look ok - put creating to a scale value looks very suspicious because very different items were just summed up with some weights …

Could it be that usability evaluation is dependent on the context of the tools to be evaluated or is it just another case of “not knowing and looking at other domains”? One thing is for sure. If we continue to use different “standards” or self-developed instruments, we will never be able to compare things….(ok it is always a risk to compare your stuff with other stuff in the right way - you risk to loose your face ;-))

Tell me what you mean!

September 19th, 2007 by Martin

Why do people so often give talks about “context” without explaining how they understand it? There’s surely no common understanding of this concept… but that’s not only concerning the term “context” - terminology is a big mess both in the field of KM andTEL. And I’m a mathematician, maybe that’s the reason why this seems to disturb me more than others: You cannot have a meaningful discussion or collaboration (or presentation) without a common understanding of the key concepts. I WANT DEFINITIONS! ;-)

Irony

September 19th, 2007 by Martin

I’m attending the “Web2.0 and Social Software” session. On the presenter’s slides, I always read “Making Semantic Web Real”. I wonder if a lot of the things now claimed to be “Semantic Web” projects really belong to the field. I have the impression that it’s often just a desperate effort to keep the label after the rather slow uptake in the last years. E.g., just using RDF somewhere in an application doesn’t make it a Semantic Web application…

Tim O’Reilly gives a very nice explanation about Semantic Web and Web2.0 in this posting:

And so, it seems to me that Pagerank illustrates the fundamental difference between the approaches of the Semantic Web and Web 2.0. The Semantic Web sees meaning as something that needs to be added to documents so that computers can act intelligently about them. Web 2.0 seeks to discover the ways that meaning has already been implicitly encoded by the way people use documents and digital objects, and then to extract that meaning, often by statistical means by studying large aggregates of related documents.

Don’t get me wrong: Using Semantic Web Technologies is fine when trying to add some more semantics, e.g., to provide recommendations. But we may not rely just on these technologies, we have to develop hybrid approaches in the future. Welcome to Web3.0!

EC-TEL day 1 and 2

September 19th, 2007 by Martin

The only way to blog about the things going on here is to do it during a session. Wireless is not available in my room, and outside my room, I prefer talking to others. Or just enjoying the surroundings ;-)

Anyway, Monday started with an interview between Teemu Arina and me about Web2.0 and social technologies. It was fun and I’m really curious about how it will look like - a link will follow as soon as it’s online! In the afternoon, I took part in the workshop WWWrong: What Went Wrong? What Went Right? Exchanging Experiences in Technology Enhanced Learning. It’s difficult to give a summary of the contributions and discussions - a lot of the problems are very similar to the ones we face in the area of Knowledge Management. In my opinion, the technical aspects are not the main issues, although there is still too less focus on interoperability and the ability to be adaptable to a changing environment. The main point is: We have to think about the interests of all stakeholders involved when introducing some new technology! Why should somebody use it, especially when this means changing habits? Will users have a benefit? What about the (middle) management? There are lots of cultural issues and personal interests one has to take care of. As computer scientists, we usually tend to neglect this and focus far too much on technology.
Another interesting thing to talk about is the meaning of “success” - when do we consider a project to be successful? There are many dimensions, such as attracting a significant number of users, educating people (also the ones involved in a project!), doing some basic research, building great prototypes, getting funding…and you surely can’t have all of them. It will be interesting to follow the discussions on the Wiki!

DSC01353

On Tuesday, Eric Ras, Martin Wolpers and me organized the 3rd LOKMOL workshop, where we had some very inspiring talks and discussions. The main question was how the use of context can bridge the gap between Knowledge Management and E-Learning. Especially: How can information about people and context be represented and used later? In my opinion, we have to be very careful about what what we are trying to formalize (e.g., competencies) and in which way we intend to exploit it. There are some things that should be left up to be interpreted by humans! Especially when it’s about defining profiles of people!

DSC01358

Last but not least: The poster session in the evening. Nice setting, isn’t it? The ALOE poster fitted well into the surroudings, and it looks as if there are several possibilities to integrate it into other approaches - or the other way round.

LOKMOL - Schedule is available

September 13th, 2007 by Martin

We’ve set up an agenda for the LOKMOL workshop taking place next Tuesday in Crete at EC-TEL. You can find it here.

I’m looking very much forward to the presentations, and many thanks especially to our invited speakers: Luk Vervenne, Teemu Arina and Margit Hofer!

Aggregate displays and collective intelligence

September 9th, 2007 by Martin

Picture by
(Photo by Azlii Jamil)

In his posting “The Danger of Aggregate Displays in Social Software“, Joshua Porter writes about the influence that aggregate displays like “most popular” or “most viewed” can have on what users consider as good or popular. Aggregate displays are also an important factor concerning the creation and exploitation of what is ususally denoted as “collective intelligence“.

A I’ve presented on some of my slides of my talk “Sharing Digital Resources and Metadata for Open and Flexible Knowledge Management Systems” about ALOE at I-KNOW, it’s not sufficient “just” to attract enough users - left alone that this is difficult enough. In his brillant book “The Wisdom of Crowds“, James Surowiecki has identified four key factors that characterize a “wise crowd”:

  1. Diversity of opinion (each person should have some private information)
  2. Independence (people’s opinions are not determined by the opinions of those around them)
  3. Decentralization (people are able to specialize and draw on local knowledge)
  4. Aggregation (some mechanism exists for turning private judgements into a collective decision)

What does this mean for an application working based on “collective intelligence”? I think it should

  • encourage users to participate (e.g., by using reward mechanisms)
  • ease the flow of information (e.g., offering users the possibility to import and export data such as their profile or their tags)
  • aggregate information (e.g., offering access to all comments and tags concerning a resource in an easy way and showing average ratings)
  • offer access to the application’s data and functionalities in the user’s usual contexts and applications (i.e., providing a proper API to allow other applications to use the data and functionalities, providing widgets, a drop box on the desktop, …)

But as aggregate displays have an influence on user’s opinions, does this mean that we should avoid to show them to users that will contribute data? Or can there also be a positive impact to a certain extent? Any feedback is very much appreciated!

ALOE in Graz and Crete

September 3rd, 2007 by Martin

Busy times. Tomorrow I’ll fly to Graz for the I-KNOW ‘07. I’ll give a talk on Wednesday at 2pm about “Sharing Digital Resources and Metadata for Open and Flexible Knowledge Management Systems”, presenting some of the work that my colleague Rafael Schirru and me did in the last months in the ALOE project.

2007IKnow Memmel

Here’s the abstract:

This paper discusses the requirements of a framework for sharing digital resources and metadata to meet the needs of open, flexible Knowledge Management solutions. The changing nature of the Web and its users as observed in recent years clearly establishes the need for new approaches and technologies to fully exploit the potential for working with existing digital resources. Formal metadata about the resources can be combined with information created in lightweight and user-centric approaches in order to significantly enhance resource descriptions and enable more efficient access to existing knowledge. The ALOE system, currently in development at DFKI, is one such solution and it is used here as the basis for a sample realization of an appropriate framework.

On Friday, I’ll also present some of this work in the Pro-LC Workshop “Process-Oriented Learning and Knowledge Work”, with a focus on “Sharing Learning Resources and Metadata”.

And one week later, I’ll present a poster about ALOE during EC-TEL 2007 in Crete, where Martin Wolpers, Eric Ras and me will also organize LOKMOL 2007 on Tuesday, September 18th.