Introduction
While hacking here at www.dfki.de we came to some "knee deep in the dirt" problems of Semantic Web querying and triple transmission.

We have the opinion that a Semantic Web server does not have "models" or you query by "passing a model uri". That is not feasible in a world that goes towards a global triplespace. So what we do instead is have one big virtual model that is inside build out of many different models that contain data. These models can be made out of adatpers (like gnowsis adapter or think of something like D2RQ ).

So from the outside you have a "Semantic Web Server" that answers your queries. The queries are in three different forms:

1. find (s p o) patterns
2. RDQL queries
3. Chatty Bounded Descriptions

Find(s p o) is easy to understand, every hacker has called one of those before. RDQL is also well known, you pass a few patterns and get a result as RDF subgraph or variable binding. The third, Chatty Bounded Descriptions are the gnowsis way of handling "Concise Bounded Descriptions". In short, you ask for data about a resource (by passing the url of the resource) and get back a subgraph of RDF around the resource, mostly literals and links to other resources.

The problem is: When you have anonymous nodes in your result, what do you do?

In case (3) it is no problem, as ChBD return a subgraph that has a closure around anonymous resources.

But in (2) and (1) you have a problem. Consider yourself querying a remote store and the store returns an anonymous resource as part of the result. like
"find (?person ?foaf:name "Leo")
and the result is an anon identifier
?person = "234234:234234:243234"

Ok, if the server has just one big model then no problem - but what if the server is an aggregation engine, embedded in an enterprise integration environment?

We have this problem right now: we implemented above search and return an anonymour resource, but just by looking at the anonymous resource it is not possible to guess where to look for more information about it. Sven Schwarz and I thought about writing a buffering system, that holds the triples with anon resources but implementing a buffering system of outgoing triples would be the source of much bug.

So we decided to create the principles of boundaries

Principles of Boundaries
In the Semantic Web we always talk about models or chunks of RDF to do something with them.

The principle of boundary is, that a Semantic Server only returns closed boundaries, with no anonymous resources at the end. If you ask a "find (s p o)" question and get an anonymous resource at the end (o=anonid), that is your problem. The server does not have to answer to "find (anonid p1 o1)". You have instead to ask the question again in RDQL, with "SELECT (s p o) (o p1 o2)".

So no anonymous resources are part in communication between servers. They may still be passed in models, but only in chunks of rdf.

This approach of Boundaries does help us very much here to implement our Semantic Web Service. If you understand what I mean, you are a real hacker.

Semantic Web is alive!

I heard of the IBM Semantics Toolkit and had to d/l it to give it a try.
https://www.alphaworks.ibm.com/tech/semanticstk

Roughly, it consists of three parts (see above site):

• Orient - Visual editor
• EODM - like Jena, but less
• RStar - storage like kowari or Sesame

I focused on Orient, as it is Eclipse based and a visual editor. The other two may be interesting, but -
1. I am Jena based
2. they are not open source
3. RStar is only IBM databases (needs either IBM DB2 or IBM CloudScape which I don't have and are too lazy to install. Cloudscape seems to be an embedded database, whatever, I stick to Orient first)

Orient

Installation is easy, read their readme.txt and unzip a few files. One mistake I made was to add Orient to my existing Java Development Eclipse - that does not work. You need the Eclipse platform, 3.0.

Then I changed to the Window-Open Perspective-Other-Orient. Voila. easy. Good looking. Reminds me of Protege. only neater. Nice typos "Please select a Konwledge I/O". First try: new project. A lot of stunning options comes up, I keep pressing "next". Ah, next experiment - import FOAF. Like Danny Ayers' FOAF driven Development. WORKS. I see foaf:Person and other stuff.

Then i tried some editing. Disturbing things happen: when I create a new resource, to add its rdf:type and therefore class, I have to enter the URI of the class. At least I can select the namespace, but the existing classes are not listed. Drag-Drop doesn't work either. Can this be true... lets see. Yes, this is low tech. I am not able to even add properties to this resource, or maybe I have to try harder. Ok, it is not a resource editor, it is an ontology editor. On the ontology side, it is better. I can add properties to classes - again by entering the URI of the property by hand.
Undo does not work also. Not always.

And then the obvious RDFS limit. The thing I often do is have one property with two classes as domain. I do this and add "foaf:depicts" as a property to "foaf:Project" (ok, not very wise). boom - it has two rdfs:domain entries, which has some disadvantages (the property is only valid for resources that are of type of both classes). This is only expressable in OWL, so RDFS crashes here anyway, so maybe this is a desired behaviour.

Export to RDF is good. The ontology comes out as clean, nice RDFS.

my bit: cool looking UI, eclipse rules. Good stable IBM ground, the EODM framework looks like it can handle this. Very easy to install, no exceptions, everything works - end user compatible. But no features, we have to wait and see until they implemented all of Protege :-)

Danny Ayers blogged something about avatars that should read your blog.

Ok, my unshaved picture on the top may be a start, please imagine now myself reading loud....

But on the other hand, it is true that "being online" can also mean that your "avatar" is moving around and talking on the "cyberspace".
I have to define the three things again:

• Being online - to have your brain think about some resources on the web (or your own computer). Your attention is on the resource in the information world - aka cyberspace
• Avatar - a representation of yourself in the information world - cyberspace
• Cyberspace - the semantic web of tomorrow, where we can hang around and consume and produce information. Also the place where we see other people, when they hang around a certain resource.

These three things are part of building "cyberspace now". I would go the easy way of looking at the datastructures that are used in online games like Ultima Online or Everquest. Then take these structures, RDF them and build a server that is capable of hosting some people (f.e. our DFKI workgroup).
Then go around and fill the empty space with data. And "going around" means that all websites you visit show other visitors that you are "there".

oh cool. If anyone wants to code this, I am in.

A big YES for the Semantic Desktop comes from the Desktop Developers Conference.

I wasn't there but Edd Dumbll blogged very good.

My bit is: Edd talks the same language as we do. RDF.

The gnowsis Semantic Desktop project we do right now at the DFKI.de is nearly what he means. We represent all "user-interesting objects" in a BIG rdf graph. And that is built dynamically, by extracting the data from the different applications.

Ontologies like FOAF, DC and VCARD are spread all over our system and make it easy to annotate photos and so. We are working night and day now to make a release soon, as I see that our thing is definitely interesting for the rest of the world.

Expect a gnowsis release in September.

in the meantime, read my thesis

hm, a hurray for the heuristics of MS-Word and the similarity of names.



(btw, <foaf:Person foaf:mbox="mailto:leo@gnowsis.com" foaf:surname="Sauermann" />)

Will Word ever be the same in the Semantic Web?

I forgot to post this link, at least I think
wwwdc4 Video

Spotlight is in the middle of it (thanks to Michael Zeltner)

although this project is a little silent (at least to my knowledge) it is surely interesting as a storage backend for Semantic Web Resources on a local desktop.

gnome-Storage managed by Seth Nickell.

In fenfire by Benja Fallenstein and in Stefan Decker's distributed P2P Semantic Desktop we will need a local storage for Resources, that is able to be accessed from normal Desktop Applications and that can be used to add RDF metadata to objects. And storage does more: it has global unique identifiers, and even better: it has URLs that are bound to hosts. great, i love it.

Best part is, that "storage" is done by some core gnome developers and so it will probably have good acceptance in the community.

The bad message is that the developer page looks dusty.

Tiger Search Technology.

Looks like a hit into a Longhorn's bullface. Right time to do it :-)

Like
Andrew Newman did already blog, this is good stuff.

I just mention it here as we are going to look at it for gnowsis. love that.

Dieter Fensel gave a talk at Vienna, on June 21, which I attended.
Talk: https://wit.tuwien.ac.at/events/fensel/index.html

Michael Zeltner joined me and we wore funny t-shirts (have a look).

Dieter spoke about his vision of Semantic Web Services, adapted to the quite mixed audience, it was also a Semantic Web introduction. Beeing the chef of www.deri.org he gave much information about what projects they do and what goals they have.

I liked this List for "Semantic Web is an enabler for":

• eCommerce
• Enterprise Application Integration
• Knowledge Management

btw, Download his slides

A big thing at the moment is the wsmo -www.wsmo.org. Big in funding is also lion.deri.ie. He also mentioned an older thing from www.ontoknowledge.org, the OntoWrapper, this could be interesting for gnowsis. And also the sekt project, which I was interested in before.

Hm, knowledgeweb.semanticweb.org is open for new members, perhaps interesting.

A nice part of the talk was about visions. Dieter sketchs two ways to a Semantic Web Service world: one through the Web Service approach, enhancing Web Services with Semantic Web tech (RDF and so on, UDDI enhancing etc) and the other by having Semantic Web applications like FOAF or so and add web services. I like the second thing.
The future may be a kind of global triple space. This leads me back to my ideas of uri-based routing and using URIQA for worldwide distributed database. We WILL have something like it, don't know yet how.
- whoa, just this moment two links popped up for Michi: deri paper, blogged in #rdfig.
I asked, "when will it be" - 5 years or 10 years? Nice answer: "5 years, we have funding for 5 years now". Or was it "I plan to retire in 5 years" - forgive me, I didn't remember.
But the good news is that Dieter thinks that smaller services will be available quite soon, surely in the next 3 years and that we will have a decent growth, not an instant revolution.

Michi and me had our moment when during the talk Dieter did stop, looked at us and asked: "Where have you got this t-shirt from?". We did answer correctly, "L....M.....".

So it was a meet and greet of Vienniese Semantic Web guys, I met also Gerald Reif and Hannes Lischka and Elke Michlmayr.

The happiest day in my life: My girl and me got married! Ingrid and I are deeply in honeymoon right now and enjoy every day of our life like never before.

We had a marvellous day and god blessed us on every way, good weather, good spirit, we felt great and were Queen and King for a day.

Interesting for the RDF community: a friend wrote a RDF representation of the wedding and dedicated it to us as a present.

ingrid_und_leo1 (rdf, 1 KB) - RDF
ingrid_und_leo (emf, 24 KB) - Visual Rendering

We are now away to honeymoon on kohsamui, expect more about mozilla&rdf after our return.

In between, watch some photos of our wedding and wish us best.