Tuesday, October 18, 2005

SnipSnap (Weblog and Wiki) and K-Logs

While checking SnipSnap, the free "easy Weblog and Wiki Software", I encountered the concept of K-Logs (knowledge management weblogs, in short: klogs), and hence klogging. Cool!
http://www.snipsnap.org/space/k-logs.

In consequence, this might lead us to "semklogging" instead of semblogging, of course :-)

Monday, October 17, 2005

Sent Lutz Maicher some comments on his German Terminology of TMDM Topic Map terms

http://www.informatik.uni-leipzig.de/~maicher/tmt/latest_tmt.pdf

I should probably one day check the finalized version against my thesis.

Friday, October 14, 2005

Semantic Wikis (SemWikis), Topic Map Wikis and Semblogging

How can I couple (Topic Map-based) semblogging with SemWikis?

Would one write LTM on the wiki pages and/or use a topic map editor?

Karsten Böhm (boehm@informatik.uni-leipzig.de) kindly pointed me to some references in SemWiki land at University of Leipzig, Germany:

In addition I found:

Semantic wikimedia is a project just started (pre-alpha) coupling MediaWiki (the Wikipedia-Software) with semantics (http://meta.wikimedia.org/wiki/Semantic_MediaWiki). AIFB Karlsruhe people have a project on a semantic MediaWiki (http://www.aifb.uni-karlsruhe.de/Projekte/viewProjektenglish?id_db=67)
http://lists.w3.org/Archives/Public/www-html/2005Sep/0002.html talks inter alia about Topic Maps. I have not found more co-citing of Topic Maps and SemWikis.

They want to employ Redland as RDF backend.

There exist topic map-based wikis, z.B. Topiki (http://www.shelter.nu/blog-070.html) or TMWiki by Hendrik Thomas (http://www.topic-maps.org/)

Addition 2005-12-08:
================
IkeWiki is a RDF/OWL-based rewrite of MediaWiki, based on Jena
<http://ikewiki.salzburgresearch.at/>
(Sourceforge: <https://sourceforge.net/projects/ikewiki/> )
<http://jena.sourceforge.net/>

Thursday, October 13, 2005

Topic Properties, PSIs and more (Email by Jack Park)

As a follow-up of informal discussions at TMRA05 in Leipzig with Jack Park on a PSI registry, how difficult this might be with notions in a flux, and what this has to do with subject identification/identity, he today sent an email. I do not fully agree and want to come back soon to this (in a comment). We probably should discuss this at Bernhard Vatant's blog.

In my understanding Jack's main points are:
* Topic Maps can help humans to mediate among heterogeneous populations of ontologies
* we need a working ontology to map most other ontologies
* a lone PSI is not sufficient to unambigously identify a subject among a universe of subjects
* TMRM acknowledges this and allows specifying the properties of subjects
* subject properties means key-value pairs (key-value-valueType or propertyType-value-valueType triples)
* we need PSIs for keys (propertyTypes) and valueTypes (derived from XML dataTypes)
* AI feature vectors might be related to subject properties
* we need a TMA (topic map application) to identify all the ontologies to define the notion of "subject properties"
* the most important aspect of future mapping of subjects with topic maps will be to establish subject identity by way of subject property declarations
* in order to not reinvent terminology, such a TMA should attempt to include all relevant metadata/ontology standards.

Wiki on Topic Maps in Libraries (Suellen Stringer-Hye)

As announced on topicmapmail, Suellen Stringer-Hye of Vanderbilt University is putting together a Topic Maps for Libraries wiki (with pmwiki) at
<http://tm4lib.library.vanderbilt.edu/wiki/>
and has established a Topic Maps Interest Group within LITA.

Website: http://staffweb.library.vanderbilt.edu/libtech/stringer/
Email: suellen.stringer-hye@Vanderbilt.Edu

Monday, October 10, 2005

Semblogging use case: 40.000 digital fotos of destroyed paintings online

Reading the article [1] about the 40.000 digital fotos now online from http://www.zi.fotothek.org/, it occurred to me that this might be an interesting user community for semblogging (semantic blogging). I hope to be able to look into providing a first semblogging facility for such users in some weeks.

Unfortunately, the application uses popup windows and not REST style!

[1] "Jetzt online: Hitlers Dia-Sammlung im Netz". Süddeutsche Zeitung, Feuilleton, Freitag 7. Oktober 2005, S. 14

For Semblogging, see my open space presentation slide at TMRA05, included below:

Semblogging with Topic Maps

Motivation
----------
* smart content aggregation of blog entries needs more semantics than just tag clouds
* see e.g. Jack Park‘s elaboration on tagging in his „just for me“ paper
* Semblogging as a special case of semantic annotation in line with DKM (Distributed Knowledge Management)

Prior work
----------
* seminal work by Cayzer (RDF semblogging concept and protoype), citing the XTM book
* redo it, and even better with Topic Maps!
* ideas by Jack Park (on semblogging as an example for Augmented Storytelling)
* idea of Dmitry Bogachev. Prototypical OKS semblogging application by Lars Marius Garshol

Current work
------------
* Developing some Use Cases
* One student implementing in his diploma thesis a prototype coupling blojsom with tmapi and TM engine
* Further work under way looking more into distributed aspects and semantic web services

What to Semblog about?
----------------------
* bibMap or the Topic Map Research and Applications Landscape?
* All TMRA05 participants and community semblogging on Topic Maps? SemWikiBlogging?
* Learning process in a teaching course in information and knowledge management?
[And now added: the collection of 40.000 digital fotos, as discussed above)

Your ideas? Who is interested in what? Who will contribute what?

Sunday, October 09, 2005

Towards a P2P(?) PSI registry

Yes, I advocated a PSI registry in the first open space session of TMRA05.

Lars Marius Garshol had blogged on this:
"Alexander Sigel held a passionate talk about the need for a PSI registry, and really wanted people to help him out. There was vigourous discussion on the subject, and general agreement that this was needed, without anyone really stepping up to pay for it. The subject was eventually deferred to the evening social session."
http://www.garshol.priv.no/blog/8.html

I do not intend to do this myself, but I want to see this happen.

A server from Hetzner starts from 39 EUR/month. http://www.hetzner.de/
Maybe we could use some PSI work from Morpheus? http://www.mssm.nl/

I agree with Jack Park that not all subjects have stable notions ("are in a flux"), and that a sophisticated PSI registry concept would have to deal with this. Upper categories are more stable, and the notions are stable at the time of describing the PSI. How can we include the (knowledge organizational) characteristics of a concept from its definition in the PSI? How are they related to topic characteristics in establishing identity?

Maybe we could have the domain psi-registry.net registered on the Norwegian Topic Map association?
(Sorry, last night I had mistyped topicmap-registry.net here. Of course this should be RDF-friendly!)

Here is my slide:
http://www.informatik.uni-leipzig.de/~tmra05/PRES/ASa.pdf

> Motivation
* Better semantic interoperability with PSI identity. Merging. SLUO (collocation objective)
* Reuse. Best Practice. Distributed Knowledge Management (DKM). Federated Seamless Knowledge
* Infrastructure for emerging collaborative distributed lightweight ontology engineering
* No long-term, public registry („PSIpedia“) existing. Wikipedia not the full solution for PSIs!
* No working group on this, not much progress seen since the inception of XTM
> Some Use Cases
* Has someone published a PSI for the topic I want to make an assertion about?
* Has someone published a set of PSIs for the type of topic I want to make an assertion about?
* (How) could I reuse an existing, even 3rd party Knowledge Organization Schema (upconvert, proxify)
* How to handle: deletion? Update/newer versions?
* Assess not only identity, but thesaurus-like relations
* Provide a pointer to a information or service offer (like a sample chapter of a book)
> Architectural and Technical Issues
* How to avoid the „arbitrary proliferation of PSI entries, or even PSI registries“?
* P2P or central? TMRAP?
* Replication?
* TM engine backend? Querying? Search engine?
> Implementation and Organization
* Sourceforge project? Language? Hosting of the application? Responsibility?
> Cooperation with content owners
* Creative Commons Licencing vs. What‘s the incentive for publishers/content producers?
> Trust Networks
* Scope? Values? like in P2P Trust Networks?
> Who is interested in what? Who will contribute what?

---
Addition 2005-11-28:
Today, Michael Chapman has hinted me to his PSI registry:
http://psi.mchapman.com/vl/

Probabilistic Datalog and tolog

The TMRA05 presentation of a TM application framework by Naito-san made me mention "probabilistic datalog" because I think this might be relevant to fuzzy Topic Maps retrieval with tolog-like languages.

Lars Marius Garshol blogged my remark right away:
"In the questions Alex Sigel brought up something I think was called "probabilistic Datalog", which is used in information retrieval. That sounded quite relevant, and interesting. I may need to do some digging on that. "
http://www.garshol.priv.no/blog/8.html

A decade ago, Prof. Norbert Fuhr had extended Datalog to Probabilistic Datalog.
In turn it had been extended for usage with DAML+OIL, OWL's predecessor.

Here some references:
http://doi.acm.org/10.1145/215206.215372
http://citeseer.ist.psu.edu/17554.html
For implementing the logical model of databases, a variant of Horn logic called Datalogis widely used. Regarding IR as generalization of database retrieval, it seems quite naturalto develop a probabilistic version of Datalog.

Norbert Fuhr: Probabilistic Datalog: Implementing Logical Information Retrieval for Advanced Applications Journal of the American Society for Information Science JASIS 51:2 (Jan 2000), 95

@incollection{Roelleke/Fuhr:98a,
author={Thomas R{\"o}lleke and Norbert Fuhr},
title={Information Retrieval with Probabilistic {Datalog}},
chapter=9,
pages={221--245},
crossref={LOGICIR:98},
entrydate=19971028,
key={Roelleke/Fuhr:98a},
year=1998,
}

http://www.is.informatik.uni-duisburg.de/bib/pdf/ir/Nottelmann_Fuhr:04a.pdf