I googled for semantic web and all I found were powerpoints.

don't know who said this, but it is lovely anyhow

Back in Kaiserslautern, I finally got some time to blog the events of the last three weeks. We went to Argentine, where my cousin Robert married Luciana in Santiago del Estero. To use the time there, we also visited Buenos Aires for a few days.

Before the journey, I thought to myself: how to meet with SemWeb people in Argentine? and blogged this. Dan Brickley answered and mentioned someone called "Inkel" who might be in Argentine. I contacted Inkel (aka Leandro Mariano López) and yes - he lives in Buenos Aires and we agreed to meet on Monday, 15th August 2005 at 17:00 in our Hotel.

And thats what happened: Ingrid and me were a little late, because we spent so much time sightseeing, but when we came to the Hotel 5 minutes after 5 pm, two Semantic Web hackers were already waiting in the comfy hotel lobby chairs. Ingrid and I didn't know if all this would work, but it did. We chatted a little about how its in Buenos Aires and then went to a little restaurant to eat - here a pic while we are on the way.



In the cafe the talk was first about the work that Inkel and Daniel represent. Inkel came to the Semantic Web a few years ago and found the foaf idea, but he missed translations to spanish. So he translated a foaf primer. After a little chit-chat, the moment of truth came:

On this day I deliberatly dressed the SWAD-Europe t-shirt underneath my sweater, to impress the SW guys a little. I announced a surprise, stood up in the restaurant and showed the t-shirt. Inkel, not hesitating, pulled up his black pullover and also wore - underneath- the SWAD-Europe t-shirt. The moment was perfect - two semantic webbers trying to impress each other by wearing the same t-shirt.



Connected together by the bounds of foaf and the same cotton on our bodies, sympathies between us went up +10. Inkel works as programmer in a company at the moment and he explained us how he plans to connect the people in Argentie, Chile and other south-american countries. He also wants to motivate people to translate SemWeb literature to spanish, so that it can gain ground there. A problem is that most people don't speak english there, and scientists don't publish in spanish. Daniel, the friend of Inkel, also translated a few texts and published them.

I told them about the upcoming Semantic Desktop Workshop and showed them a few papers. I had all papers of the workshop with me in printed form to read them and to know the stuff deeply. They were interested to see that we have one spanish contribution to the workshop and perhaps this may connect people a little better. I explained them our upcoming Nepomuk project and the idea of the Semantic Desktop. Actually, I used my own paper for the workshop as a walkthrough to give a little exaplaination/mini talk. I gave my copy of it to inkel, so that he can use the links and references.

Daniel has to write diploma thesis in the next months and perhaps our Jena-based GUI thing from gnowsis could help him. We have to write some documentation for it, anyhow.

After a while we moved on to a coffe-house at the other side of town (south of Av 9 Julio), a coffee house that looks exactly the same as the coffee houses we used in vienna to meet with Michael Schuster or eaon or geri all the other SemWeb guys in vienna. Astonishing how the grass-roots culture is always the same.

We sat together and talked about possible business connections. Acutally, through the weak Peso (the currency in argentine), it is quite cheap to hire Argentine programmers for Semantic Web stuff.



Ingrid ordered a "submarino", hot milk where you put chocolote in to make hot chocolate.



The situation in Argentine is far from the ongoing work in Europe. They don't have the big budgets for Semantic Web and most people can only do this as a hobby. Also, university professors don't get paid very well and university research is not as good as in Europe. We are quite lucky to have the 4th, 5th and 6th framework programme of the EU with big amounts of SemWeb money. For example, the t-shirts that connected us were a deliverable of a EU project, SWAD.

Finally, after five hours of permanent chatting, I got a little achein my throat (a starting cold) and we decided to part ways again. It was a very fruitful discussion about deep topics and networking. Nice to meet Inkel and Daniel and best wishes to these guys!

according to a press release the new oracle server supports RDF. does this mean that the semantic web is mainstream now? hurray!

Ingrid and me are in Argentina right now. If you do not get emails by me or wonder why I am so silent - its permanent barbecue and sitting around drinking mate at the moment.

Btw: I have met Inkel! as organized through this weblog. The blogosphere worked: through the help of DanBri I got to know Inkel and we met and had, together with Daniel an evening of tough FOAF and Semantic Web Chat. We met in Buenos Aires, in our Hotel and then went eating together, it was like a Semantic Web Lounge Meeting in Vienna, cool.

Photos will follow - and be warned - the photos will definitly rock (surprise surprise...)

a question that is often faced by RDF developers: how to change those RDF graphs?

ususally, you would use jena or sesame's api to directly add triples, delete them, list classes using (x rdf:type rdfs:class) and so on...

Then, some developers make libraries that allow you to generate code that wraps the RDF in an object model.

This means, instead of adding triple
X.addProperty(FOAF.name, "Leo");
you say:
X.setFoafName("Leo");
where the setFoafName method was generated by such a code generator.

Now some code generators for you:

• rdf2java by Sven Schwarz, DFKI. Jena based, works on RDFS and OWL. used in some research projects.
• SWeDE by BBN. they mention Kazuki code generation.
• rdfreactor by some Karlsruhe guys.

from my perspective, all of them do the same. pick what you like.

Ingrid and I are going to argentine on August 13, to be guests at the wedding of my cousin Robert Sauermann. We'll be first in Buenos Aires from 14th to about 17th August and then in Santiago del Estero, at the 20th is the wedding there.

Later on we'll move through the country and see a few places. I don't know anything about Argentine except the Evita movie and that che was born there. Hm, steaks and Churrascaria.

So how will I find out whom i foaf:know in Argentine to visit a few Swhack and rdfig people?

In the last two months we shifted the gnowsis search services to SPARQL. Our problem was, that the common ARQ implementation is slow and does not work in a named graph scenario.

A fine twist in history brought Richard Cyganiak to work at HP Labs for a while and he hacked sparql2sql there, a mapping of sparql to the jena database scheme with support for named graphs.

I asked him if I can use it in gnowsis and he was quite happy to have this as a test-case. Result is, that Richard added a special framework for fulltext search and I hacked some stuff regarding MySQL's fine FULLTEXT index functions (if you don't know it: its similiar to lucene, can do relevance and query expansion and so on)

outcome: an hour ago I transformed the SQL index to the new fulltext format and got hit in the face with the new answer times:

SPARQL of this kind works practically instantly (10-20msec):
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
SELECT ?x ?label WHERE
{ GRAPH ?source {
?x rdfs:label ?label.
FILTER REGEX(?label, "test", "i")}}

But the real astonishing thing is that SPARQL of this kind is also fast as hell (10-500ms):
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
SELECT ?x ?p ?label WHERE
{ GRAPH ?source {
?x ?p ?label.
FILTER REGEX(?label, "test", "i")}}

In simple words: this is a fulltext scan over all properties of all statements. don't get bothered by the warmup time, the thing will need about 10-20 sec warmup, but then its great.

triplecount:
371994

for the freaks we will probably pack a little executable jar that packs all into a nice demo. for the real freaks:
http://gnowsis.opendfki.de/repos/gnowsis/trunk/sparql2sql/

demo app

I am putting together more about smushing, which will be a key factor in the global semantic web: to connect annotations that were made by different people.

A typical smushing algorithm would be:

• take a large datastore DS that contains a set of triples Tset = {Ta, Tb, Tc, ... }
• iterate through known InverseFunctionalProperties IFPset = {Ia, Ib, Ic, ....}
• for each InverseFunctionalProperty Iy that is represented in the Tset as predicate, do a check for smushing.
• find all triples TxIy so that Tx has Property Iy
• find one triple Txc of TxIy that points to a grounding resource / canonical resource (see below)
• Use the subject Sx from the triple Txc and aggregate all other triples of subjects of TxIy to Sx. This means, change the subject in the triples to Sx.
• add owl:sameAs triples to connect all Subjects(TxIy) to Sx

The problem is, when you have a set of triples TxIy that have several subjects that should be the same - as defined by IFP - to choose which subject is the "canonical" subject and should now be filled with the triples.

There are different approaches to find the canonical resource:

• take by random
• prefer the resource that is annotated in special ontology (ie prefer SKOS concepts over foaf:Persons)
• prefer the more public resource (googlefight, public urls wins over private uris)
• prefer the best annotated resource (the resource with the most triples - attention, this is self-amplification of single resources)
• prefer the resource with the shortest / the longest uri
• prefer named resources over anonymous resources (this is very important, you must not smush to anonyms)

Another question is what to do with the smushing. Different approaches

1. store the smushing in an extra graph
2. delete the old triples, add the smushing
3. add the smushing additional to the old triples (tricky)

Each has obvious advantages and disadvantages. For gnowsis I would prefer (1)to smush into an extra graph, which is similiar to (3) but seperates the data.

In gnowsis we have the problem of incremental smushing, which means that we crawl thousands of emails per day and then would like to smush the persons in the addresses, but only of the new messages.


I have posted this algorithm also in the ESW wiki, where you can comment on it.

I updated my diploma thesis website, where I offer topics related to the Semantic Desktop as Master Thesis.

http://www.dfki.uni-kl.de/~sauermann/projekte/index.html

this always takes so much time, to write down my ideas and things that will be needed. The next student starts in October with a Thesis, so there are still places open.

here are the current topics. Well, I should add another one for GUI!

• your own idea - "I have an idea for the Semantic Web, its X and Leo, you will like it!" mail me.
  
• business card annotation- annotating business events
  
• managing concepts using SKOS - concepts for the masses
• semantic tools - evaluate semantic tools for our future
  
• semantic search - search inside the semantic desktop (taken by student)
  
• semantic email - how to annotate emails and how to classify emails. Eat Spam!
  
• text annotation - using the semantic desktop to annotate documents or write notes about sport events
• metadata integration - integrate information using RDF
• cyberspace in ultima
  online - Are you a mage or a hacker? or both?
  

http://futurezone.orf.at/futurezone.orf?read=detail&id=270140&tmp=17566

there is a orf futurezone entry about the i-know conference, that goes more into detail about the work of me and Matteo Bonifacio.

Some of the gnowsis and Semantic Desktop features are mentioned there, great thing to be linked on the fuzo.

And also Matteo, I am always happy to see this guy, he is full of energy and good science.

the conference was a nice place to be!