Doing Guest Research at L3S

Update from last post, I am in Hannover engaged in Research. Julien Gaugaz from L3S and I are working on the topic of Current work Context of a user for Nepomuk. This is tricky, as first of all, there are about 600 definitions of “context” (there is a paper about analyzing 66 of them using text analytics, context buzzword word that everyone uses with its own interpretation!) and second, we have multiple existing ontologies and implementations to integrate.

btw, this is the room for the Hiwis and guest researchers where I sit at the moment:
Here I sit
I will blog about our extensive literature research on context later this week.

In the meantime, Cedric Mesnage from Lugano interviewed all DFKI and L3S members about what they are working on in the project, to improve the architecture of Nepomuk and communicate it better, a good task.

We drank beer and looked at the rainy (no, its not raining, its only dripping, Hannoverians have 100 words for rain) city of Hannover.
To Heiko Haller I said in a chat, that we will probably end up in a Fiasko, but actually we started there:
Es fängt im Fiasko an

Besides, Cedric can not only do great art, he can also do photographs (bummer that the autofocus didn’t get this one perfect):
Leobard Light Angel

Research visit at L3S

I am currently guest researching at the L3S lab in Hannover. L3S does research on Semantic Web, e-learning, search. They are partner in the NEPOMUK project, and I will be working on the ideas of the context a user currently works in, to guess what somebody is doing based on observations done. This is important to give support to people, showing them information that is related to the task you are doing at the moment. (for example: you are reading this blog-post, and related information is l3s, dfki, nepomuk, leo sauermann, context).

I will be at L3S for three weeks, from 24.9.2007 to probably 14.10.2007.

Here is the building where I am sitting at the moment: KBS building with L3S people

Where I sitIts the L3S offices at KBS, Appelstraße 4, 30167 Hannover. I am sitting in room 232. If you want to use this occassion to meet me, skype or jabber me first (check my foaf for how to).

Appartments from Gaststätte Kaiser I am living not far from that in an appartment room, where I have access to a kitchen and a bed, and free wifi for 90 minutes per day, great, what else do you need.

God and numbers blog

Mike Robb, my praying buddy in Kaiserslautern, and his lovely wife Katharina started a new blog to inform their friends about their trip and life in Aberdeen. They called the blog called “god and numbers” . Mike is a missionary, studying theology, Katharina is a mathematician, she worked for Fraunhofer before and is now searching for a Job in Aberdeen.

btw: Are you a company in Aberdeen, in need for a mathematician that can do stuff like this diploma thesis on “Spectral Analysis of Random Closed Sets. The Surface Measure Associated with a Random Closed Set”? Comment on their blog or write them.

The first blog post was about Mike moving his stuff. I showed up late for carrying, and then it became obvious that we had two problems. One was, that not everything will fit into the truck that Katharina’s dad has rented for the occasion. Using the great skills of re-structuring moving boxes I have learned with Ebo on my own move from Vienna to Kaiserslautern, we could solve that.

Here is the skills as used before:
Our stuff limited to a van
When we moved from vienna, in the van, the boxes reached up to the very ceiling and filled the whole thing. There was not place for one extra mouse in there. For Mike and Katharina, we produces a similar result, stuffing every item into the places of air that was left in the van (*proud of tetris skills*).

The second problem was, that we were not enough people. I came late, and besides me, the parents of Katharina were there. So, we were very lucky when two mormons biked by, saw the situation, looked, stopped, and asked if they could help. First we didn’t want to bother them, but looking at the situation, I asked them if they could help us carrying. They could, and then it was much easier. Love god, love your neighbour, love yourself, that is one of the many rules shared by our churches. Thanks god for giving these recommendations to make life better for this planet. At least for us, god works perfectly.

Got Schmapped

Schmap.com is a website providing dynamic travel guides. Instead of using boring stock photos, they also use flickr photos to illustrate their guides. I was contacted by them if they could use one of my photos for their guide (thanks to CC this all works).
Update (27.9.2007): As the photo wasn’t CC licensed (Thanks Maggi for pointing it out), I have to add that they asked me if they could do it. Hm, I wonder if they look for good pictures or for CC pictures.

Now, their guide is updated and here is the section where they use my photo (taken during ISWC 2005, of the conference dinner 🙂

They have a schmaplet to embed the post in blog-posts, but it seems that my blog-provider doesnt allow i-frames, so
click yourself.

This is the way to go: web 2.0 business!

Cyberspace “wonderland” by sun

SUN researcher Nicole Yankelovich presents an open-source 3-dimensional virtual environment, that can do many tricks, including application sharing in a like-VNC way.

Its called project wonderland, and builds on darkstar, an open source gaming server platform. Underneath, there is looking glass, an open source 3d platform.

Its a business-oriented, knowledge-work oriented environment, published open source. Unlike secondlife, which is closed-source and leisure-oriented.

With systems like this and all data represented as RDF on a Semantic Web, using Semantic Desktop applications, users could intuitively place their work in a virtual environment and share their ideas. Looking forward to see how this integrates with Semantic Web.

thanks to Henry Story for the link and his story on wonderland.

Microsoft Astoria on RDF: “no use” and pushing own format

Mike Bergman blogged about a new release of Microsoft Astoria, where they announce support for Atom, JSON (who-hoo), and Web3s.

I doubt that this can be taken as a signal from the whole Microsoft Corporation but is rather the view of project members of this one project. Nevertheless, its going to be interesting what Microsoft will do: bunker themselves in (terran style) and hope that their customers will never hear of open standards? I doubt, and therefore I do not fully agree with Mike:
This is yet another stunning and lame attempt by Microsoft to replace open standards with proprietary ones. Get a clue, Redmond!

They probably got their clue already, but for big companies it always takes time… HP (Jena), IBM (SLRP) and SUN are faster this time.

Triple-I 2007 Wrapup

As blogged before, I attended the Triple-I conference 2007.

triple-i logo

Summing it up:

  • There are pictures online from me, bblfish (Henry Story), and others.
  • The four Keynotes were interesting, and gave insights both from scientific view (innovation, trends)
  • as from the corporate view (buzzwords, trends, where to invest)
  • DFKI contribute quite much in the field of Knowledge Management, we had several papers published on: the Nepomuk Social Semantic Desktop (with many co-authors), the PIMO personal information model, the ConTag tagging system, Philosophy and Cognitive aspects of Semantic Desktops, and E-Learning
  • Martin Memmel also blogged about his trip to Graz

Papers were I contributed were:

Keynotes:
 Keynote Marc Smith demoing Vista "Tagging Game"

The Triple-I is balanced between science, technology fair with customer contacts, and social networking event. For example, enjoy the social event:

Social Event

Klaus Tochtermann and the Know-Center are the main organizers, the newly re-branded Semantic Web Company assisted, as did Salzburg Research. Hermann Maurer was omnipresent, as scientific father of many innovations.

Goodbye Handshake

Here is a short summary of the conference, extending my previous post.
Triple-I was this year not-so-organized during the paper submission phase (they changed paper length once) but on-site, its very well organized. The keynotes are at 09:30, leaving enough time to get your morning coffee. Between the sessions are long 1h breaks to knit your social network. I met many old and new aquaintances and friends. Fine conference.

At the evening of the first day, there was a welcome reception at Minoritensaal.

Second day I presented a haptic personal semantic network, a birthday present from my university friends:

In the afternoon of Thursday, the second day, the FFG and the Innovation Relay Centre Austria organised an International Cooperation Event. I represented DFKI at this Event and revived contacts with old partners and found new possible partners, alltogether a speed dating with 10 people, 20 minutes each. Exhausting, but worth it.

Friday Evening, Andreas Blumauer organized a dinner I was very happy to attend.

Andreas Blumauer:
Andreas Blumauer

Here is what they say about themselves:
The TRIPLE-I Conference series is a joint venture of the conferences I-KNOW, I-MEDIA and I-SEMANTICS. TRIPLE-I reflects the increasing importance and convergence of knowledge management, new media technologies and semantic systems. This unique concept aims at bridging the gaps between the various communities and their technology fields.

Building upon the reputation of the premium conference series I-KNOW and SEMANTICS the upcoming event addresses representatives from academia, industry and public administrations. TRIPLE-I will offer its participants a unique platform either to present latest and leading edge developments or to catch up with the developments of most innovative IT technologies, content applications, business models and emerging market opportunities.

Aperture – what it can, what we do next

What is the current status of Aperture, where are we heading, what needs to be done?

Aperture is a Java framework for extracting full-text content and metadata from various information systems (e.g. file systems, web sites, mail boxes) and the file formats (e.g. documents, images) occurring in these systems. At the moment, many people use Aperture and we try to compile a list of users on this wiki page:

http://aperture.wiki.sourceforge.net/ProjectsUsingAperture

If you are using Aperture, please give a link there.

We want to make a release in the next months, probably sooner than later, if you have free time and want to fix bugs, be our guest to quicken this 😉

The next urgent things I would like to do are:

  • fix bugs in aperture – there is one for everyone
  • A better Lucene handler, allowing Lucene developers to use Aperture as framework
  • Make e-mails openable in Thunderbird (in Outlook it works, but Thunderbird’s programming model is an open secret only readable in C code). I already tried to understand Thunderbird’s API, any new clues welcome.
  • fix the IMAP uris, they were ok in gnowsis but are broken in Aperture, thats an old one.

Danny Ayers blogged about using a mork parser to extract the browsing history of Firefox.

I replied that we may want to have this also in Aperture.

The demork code is here: http://gnowsis.opendfki.de/browser/trunk/demork

Its also packed as JAR with aperture, and used in the ThunderbirdCrawler.

Anyone wants to take this code and improve it to crawl the browsing history? Write a note to Aperture Developers (have to register first)

Attending Triple-I Conference 2007

At the moment I attend the Triple-I conference 2007 in Graz, Austria.

triple-i logo

Triple-I was this year not-so-organized during the paper submission phase (they changed paper length once) but on-site, its very well organized. The keynotes are at 09:30, leaving enough time to get your morning coffee. Between the sessions are long 1h breaks to knit your social network. I met many old and new aquaintances and friends. Fine conference.

At the evening of the first day, there was a welcome reception at Minoritensaal.

Second day I presented a haptic personal semantic network, a birthday present from my university friends:

On the first day, I presented Danish Nadeems paper on Cognitive Semantic Desktop. There was a welcome reception at the evening.

The TRIPLE-I Conference series is a joint venture of the conferences I-KNOW, I-MEDIA and I-SEMANTICS. TRIPLE-I reflects the increasing importance and convergence of knowledge management, new media technologies and semantic systems. This unique concept aims at bridging the gaps between the various communities and their technology fields.

Building upon the reputation of the premium conference series I-KNOW and SEMANTICS the upcoming event addresses representatives from academia, industry and public administrations. TRIPLE-I will offer its participants a unique platform either to present latest and leading edge developments or to catch up with the developments of most innovative IT technologies, content applications, business models and emerging market opportunities.