headlines: can URIs be ambigous – democracy prevails!

There is a question related to the semantic web, and this question is – will it be a centralized, dictated system or open? Is there “one weird standard to rule us all?”

Update: Roy Fielding, who motivated to write me this post yesterday, answered in a comment and I reconsidered my post, rewriting it (29.2.2008). Updates are Italic, deleted text striked.

The fundamental question as such – is the semantic web a controllable system or a distributed (more chaotic) structure – shows up in different manifestations. I interpret the question of unambigous URIs – one URI for one concept, not multiple – as subtopic of this.

As you could guess already, the answer is no. The Semantic Web is as free, open, uncontrolled, unreliable as the web today, but with more features.

Roy Fielding said (actually cited) that one the Semantic Web’s goal is to unambigously identify resources. He also cited another quote by Tim Berners Lee:
I don’t want the Web to constrain what people do: the Web
is not there to constrain society. It’s there to model society
in its completeness, in its entirety. [Tim Berners-Lee, 1994]

What does this mean? Unambigous means when you talk about the Tesla Car, you must always use the same identifier (in our case, a URI) to refer to it. As could be expected, this idea is not a requirement of the semantic web and not practically required nor used much. Some people state it as a nice scientific goal, but deployers don’t have to care about it as the W3C recommendation has something else to say.

Instead, people continue to say things about the world in blogposts and wikipedia and elsewhere as always, minting new URIs for things as they want. In the Semantic Web, the standard tags “rdfs:seeAlso” and “owl:sameAs” are then used to link the different views about the same thing, or the Tesla. If you want to neen non-ambiguity, perhaps use sindice (or any other semantic web search engine). Horray, freedom of expression and scalability prevails.

And yes, the Semantic Web is already there, for example on openlinkeddata, or on GoPubMed. So maybe Roy’s statement “the semantic web will never happen” indicates Roy is living in the past? We will see in the future…

Sorry Mr Fielding, this sarcasm now rebounds to myself, I was wrong, you are right in citing both positions.

DFKI at CeBit 2008

I will be presenting NEPOMUK at the CeBit 2008 next week,
at Booth B37 Hall 9.

I will be there on 4th and 5th March, if you want an appointment, please phone me beforehand (I won’t read mail that much).
(germany) 0176 24548974

here is the full press release in German:

DFKI auf der CeBIT 2008

Das DFKI ist auf der CeBIT 2008 (04.03. – 09.03.2008) im Rahmen des CeBIT-Konzepts “future-parc” mit einem eigenen Messestand vertreten. Der Stand B37 in Halle 9 umfasst eine Fläche von 72qm und befindet sich in unmittelbarer Nachbarschaft zum Stand des Bundesministeriums für Bildung und Forschung, BMBF (Halle 9, Stand B40). Darüber hinaus präsentiert das DFKI Exponate auf einem weiteren DFKI-Stand, dem DFKI – John-Deere-Stand (Halle 9, Stand C07), auf dem Stand des BMBF und auf dem Gemeinschaftsstand der Universität des Saarlandes (Halle 9, Stand B35).

Exponate auf dem Haupt-Stand des DFKI (B37)

Exponate des FB Bildverstehen und Mustererkennung

  • OCRopus – Open Source Texterkennung
  • InViRe – Intelligentes Video Retrieval

Exponate des FB Intelligente Benutzerschnittstellen und FB Sprachtechnologie

  • BabbleTunes – Sprechen Sie mit Ihrem iPod
  • i2home – Mobiler multimodaler Zugang zum digitalen
    Zuhause für alle
  • Ideas for Games – A.I. Poker im Casino Virtuell

Exponate des FB Wissensmanagement

  • Nepomuk – The Social Semantic Desktop
  • ALOE – A Socially Aware Resource and Metadata Hub
  • iDocument – Intelligent document information extraction
  • Eye-Book – Multimediales Lesen

Exponate des FB Sichere Kognitive Systeme und FB Robotik

  • SAMS – Sicherungskomponente für Autonome Mobile Systeme
  • Robotik Videos

Expoante des FB Institut für Wirtschaftsinformatik im DFKI

  • Pipe – Hybride Wertschöpfung im Maschinen- und
  • R4eGov – Organisationsübergreifende Zusammenarbeit von öffentlichen Verwaltungen

Exponate des FB Deduktion und Multiagentensysteme

  • CASCOM – Intelligente Dienstagenten für medizinische Notfalleinsätze
  • MAS-Dispo XT – Multiagententechnologie in
    der Stahlproduktion
  • Scallops – Secure Agent-Based Pervasive Computing

DFKI – John Deere – Stand (C07)

Auf dem DFKI-Stand C07 präsentiert das DFKI in Kooperation mit John Deere das Projekt IVIP im Kontext “Green IT”.

Exponat des FB Wissensmanagement

  • IVIP – Intelligente Vernetzung verteilter Informationsquellen zur
    betriebs- und standortspezifischen Planung der Energiepflanzenerzeugung

Das BMBF stellt auf seinem Stand B40 das Förderprogramm der Bundesregierung “IKT 2020” vor.


Zentrum für Mensch-Maschine-Interaktion

  • Smartfactory – die
    intelligente Fabrik der Zukunft

FB Intelligente Benutzerschnittstellen

  • SoKNOS – Service-orientierte Architekturen zur Unterstützung
    von Netzwerken im Rahmen öfentlicher Sicherheit

FB Robotik

  • SentryBot – Ein Autonomes, Kooperatives Mehrrobotersystem für Sicherheit und Objektschutz

KWT-Stand (B35)

Forschungsbereich Deduktion und Multiagentensysteme

  • Verisoft_XT (Tom in der Rieden)
  • ActiveMath (Erika Melis)

Hallenplan der Halle 9


Einen detaillierten Hallenplan
im PDF-Format können Sie
hier herunterladen.

Publishing “The Sesame Lucene Sail: RDF Queries with Full-text Search”

We have written a Technical Report on our integration of Sesame2 with Lucene.

Enrico Minack, Leo Sauermann, Gunnar AAstrand Grimnes, Christiaan Fluit, Jeen Broekstra: The Sesame Lucene Sail: RDF Queries with Full-text Search.
download PDF (alternate link)

For short:
PREFIX search:
SELECT ?x ?score ?snippet WHERE {?x search:matches ?match.
?match search:query “person”;
search:score ?score;
search:snippet ?snippet. }

With the growth of the Semantic Web, the requirements on storing and querying RDF has become more sophisticated. When a larger amount of data has to be managed, queries in structured query languages, such as SPARQL, are not always powerful enough. Use of additional keywords for querying can further reduce the result set towards the actual relevant answers, however, SPARQL only provides complete string matching or filtering based on regular expressions, which is a very slow operation. In contrast, state of the art Information Retrieval (IR) techniques provide sophisticated features such as keyword search, lemmatisation, stemming and ranking. In this paper we present a combination of structured RDF queries and full-text search. It is implemented as an extension of an established RDF store (Sesame) with IR capabilities using the text search library Lucene, without requiring modifications to existing RDF query languages.

(in these files you find all my publications, including this one)
bibtex / rdf

The implementation lives here:

“Your Studio and You” – the movie

I found one of my favorite movies on google video: “Your Studio and You”.

You can read the wikipedia article to learn more, the film has three aspects which make it excellent:

  • It is funny and well done
  • The movie featured a bevy of celebrity appearances which includes Andrew Bergman, James Cameron (shown improving the Universal Studios landscaping), Shaun Cassidy, Robin Cook, Shelley Fabares, Michael J. Fox (sometimes mistaken in the film for Dian Bachar, a longtime Parker-Stone collaborator), Brian Grazer, Heavy D, Jeffrey Katzenberg, Barry Kemp, Angela Lansbury (shown painting the Psycho House), Mike Lobel, Traci Lords, Kevin Misher, Demi Moore, Darrin Pfeiffer, John Singleton, Steven Spielberg (shown as a Universal Studios Guide), Sylvester Stallone (in his Rocky Balboa character, and subtitled for comedic effect), Marty Stuart, David Zucker, and Jerry Zucker. (from Trivia about Seagrams)
  • It is historically interesting.

Let me focus on the last point. Read on Zap2it.com an interview with Trey Parker, who together with Matt Stone directed the movie.
On April 6 1995 Seagrams, back then one of the largest beverage company in the world, decided to sell its share in DuPont oil for $9 Billion to instead later buy Universal Studios, MCA, PolyGram, and Deutsche Grammophon.

Universal Studios PR employees contacted David and Jerry Zucker “to make a short film for Universal to be played at a big coming out party Seagrams was throwing for all of its employees.” link
The Zucker brothers (“Airplane”, “Naked Gun”, “Top Secret”) had no time and asked Trey Parker and Matt Stone (who were complete unknown punks) to do the movie, Zucker Bros. said to already have a “very funny script”.

But when Trey and Matt appeared on site for the takes with Steven Spielberg, James Cameron and Michael J Fox, they were surprised that the “very funny script” did not exist. Also, the PR people were aroused not to see the Zucker Bros on site, but punks.

“We suddenly just stopped everything and said, ‘let’s just do it all like a really stupid ’50s industrial movie. Because then at least that part of it will be funny. And so we just started trying to shoot it like those little ‘duck and cover’ nuclear things from the ’50s.”

The movie features “Seagrams Wine Coolers”, and good advice how to improve Universial City Studios (UCS) – all employees agree “It’s UCS for me”. So, Seagram beverages decision to get more fun in entertainment business led to this movie a year later.

Really weird: Trey Parker did not see the movie for four years after the take, because it was so internal that they did not give a copy to the directors (!?) but it showed up on the web (!!hurray!!) in 2001, and then… haha.

The result is above, and the guys made many great things after and before it, such as Southpark. True Art.

call for tutorial proposals at KI 2008

Submit your proposal, this conference is in my hometown and if you come, we can drink a beer. Submit your proposal!

KI 2008 — 31st German Conference on Artificial Intelligence
23 – 26 September 2008, Kaiserslautern, Germany



KI-2008 will include a small number of tutorials, to be held the day
before the technical program starts. Tutorials will be free of charge
for conference participants. We aim at a small number of high-quality
tutorials suitable for a large percentage of conference participants,
including graduate students as well as experienced researchers and


Tutorials should give a comprehensive, in-depth perspective on
innovative AI methods or technologies that have an obvious potential for
research and/or application and are not covered by typical AI textbooks.
We especially encourage tutorial proposals covering cognitive aspects of
AI. Do not hesitate to contact the conference or tutorial Chairs if you
are in doubt about the suitability of a particular topic for the
purpose. Both full-day (6 hours) and half-day (3 hours) tutorials are of


KI-2008 tutorials will be held on September 23, 2008. They should
preferably be given in English. The revenue for tutor(s) consists of
free participation at the KI-2008 conference, plus reasonable travel
support if necessary. Teaching material will be printed and distributed
by the conference organization.

How to Propose a Tutorial

Proposals must contain the necessary information to judge the
importance, quality and community interest in the proposed tutorial as
well as the expertise of its tutor(s). Proposals must include
descriptions of the tutorial topic, goals, the intended audience, an
outline of the contents, and brief CVs of the tutor(s), including their
expertise and teaching experience in the field, and the intended length
(half- or full- day). Proposers are encouraged to include excerpts of
material from recent teaching about the proposed topic as an annex of
their submission, if available.
Proposal texts should be submitted by e-mail to the Tutorial Chair in
plain text format. Annexes may be sent as .pdf, .ps, .ppt, or .doc

Important Dates

Proposal deadline: Mar 28, 2008
Acceptance Notification: May 30, 2008
Camera-ready teaching material due: July 2, 2008
Tutorial: Sep 23, 2008

KI-2008 Tutorial Chair

Prof. Dr. Frank Bomarius
frank.bomarius (at) iese.fraunhofer.de

never installing a laptop os

Your computer doesn’t start? Black screen, a cursor, nothing else? Well, you may be tempted to wipe & reinstall. Back at impact we had the company directive not to fumble around installing operating systems again and again, but to stick with the os the vendor has preinstalled and fix it. The reason behind this is that most vendors tune the installation and also that re-installing all software and data costs too much time = money.

So yesterday Ingrid’s compaq evo laptop got a black screen of nothingness after the bios logo and instead of windows, a cursor flashing once. Then reboot and infinite repeat. I used knoppix to check what happened, harddisk still there, files looked as usual. Next thought: master boot record (MBR) so I read there are two ways: using gentoo startup cd and ms-sys or windows xp install cd. I had only knoppix so I choose windows xp boot cd. But my boot CD didn’t have the “recovery console” menu when installing. Pressing F10 continously while it bootet finally brought up console (how intuitive). Later I found the windows xp cd shipped with the laptop, that had the “r” for recovery, so there is a difference between windows and windows.

fixing mbr and boot in the recovery console is easy, type fixmbr and fixboot and it should work. Not for me, I also thought that maybe all this is wrong and my floppy drive is broken (ha, that is the first in boot order), so I moved it down in bios list of boot devices. I also checked HP’s guide for fixing boot problems, and blindly entered commands like “fixmbr /device/harddisk0” in recovery console.

Ok, it bootet now. But then:
Windows XP could not start because the following file is missing or corrupt: \WINDOWS\SYSTEM32\CONFIG\SYSTEM

A thing I often heard from nightmare stories, it happened the registry was missing. Lucky me, the Microsoft Guide to recover the registry is on the first page when googling for the problem. The boot loader was right, half of the files of the registry were gone. I copied them from an automated backup which is done by windows every day (ha, you never know what useful things this os does when “being busy”).

Alas, after 3 hours the system was back. So, it took me three hours to get my system back, compared to the 2hours of installing an operating system and then endless days of updatefest and installing software, it pays off. I am quite slow with computers, others may have done this faster. Ingrid was quite happy to have her machine back, though. I don’t pick any OS better than the other, I have the same problems with all of them. But keeping one alive is always better than wiping, at least for me.