Wilbur-and-O http://www.lassila.org/blog/ Ora Lassila's blog about Common Lisp, Semantic Web, and Wilbur]]> 2012-04-10T20:24:49-05:00 Time to revisit Wilbur, part 2 http://www.lassila.org/blog/archive/2012/04/time_to_revisit_1.html I am happy to say that after I blogged about a Wilbur "redux", several people have already contacted me about the idea.

Now I have been thinking mostly what existing software I could use in the reimplementation. Last few years, many good quality libraries for Common Lisp have emerged. Here's my current thinking about the most critical bits:

  • HTTP client: Drakma
  • HTTP server: Hunchentoot (the Common Lisp version of OINK, on top of Wilbur, uses Portable Allegroserve, but I think we need to let go of that now)
  • XML parser: several choices are available (CL-XML, S-XML, CXML), I have to figure out which one would make sense; I just need a "SAX-like" API for the RDF parser (which will be recycled from the Wilbur2 implementation)
  • SQL connectivity: CL-SQL
  • HTML generation: possibly CL-WHO, but the functionality I implemented for OINK earlier is actually pretty good and could be reused

Many other libraries also under consideration (URIs/URLs, EXIF metadata extraction, JSON, etc.).

On a slightly different topic, someone suggested I should consider reimplementing Wilbur in Clojure. Certainly then the availability of libraries would not be a problem.

]]> Wilbur ora 2012-04-10T20:24:49-05:00 Time to revisit Wilbur http://www.lassila.org/blog/archive/2012/04/time_to_revisit.html I have been thinking about "revisiting" Wilbur, my Semantic Web toolkit implemented in Common Lisp. Wilbur has not been maintained for several years, and I have been doing most of my prototyping using Piglet, a reimplementation of Wilbur (in Python and C++).

If I were to redesign and reimplement Wilbur (probably as "Wilbur 3" this time), I would like to base it on

  • Piglet database design (including support for aspect-oriented data), and
  • best-of-breed CL libraries for various thing I might need (HTTP client, XML parser, etc.), rather than "rolling my own" as I did for the original Wilbur (because nothing existed at the time).

I also have some ideas about how to take the code written for OINK and use it as some kind of an "object-relational mapper" (or RDF equivalent thereof).

]]> Wilbur ora 2012-04-08T11:56:27-05:00 Another Interview http://www.lassila.org/blog/archive/2012/02/another_intervi.html I was interviewed last week by kimmicblog, mostly about innovation in general and my work on Semantic Web in particular.

]]> Semantic Web ora 2012-02-08T09:49:17-05:00 Linked Data Workshop @ W3C http://www.lassila.org/blog/archive/2011/12/linked_data_wor.html Last week, I attended the W3C workshop on Linked Enterprise Data Patterns. I gave a talk, together with my colleague Ryan McDonough (follow these links: position paper, presentation slides).

The workshop was very interesting, revealing a couple of things of particular interest (to me):

  • I am not the only one frustrated by the fact that there is a lot of confusion between URLs as queries and URIs as identities (of resources/objects). This may seem like a small thing, but actually it is a real obstacle in the adoption of Linked Data.
  • ROI of Linked Data in large organizations still unclear (as a perception).

If everything goes well, we will start some new work on clarifying what Linked Data applications will look like, and how to build them. Another small step towards my real goal, the adoption and deployment of Semantic Web technologies.

]]> Semantic Web ora 2011-12-11T21:44:08-05:00 Interview http://www.lassila.org/blog/archive/2011/02/interview.html Folks at the "Ideas Project" pulled me aside during the last Nokia World event and wanted to know if there was an idea I wanted to talk about. They videotaped this interview of me talking about data sharing between applications, and the "death of applications" (as we know them today). Those familiar with my old ideas about the Semantic Web will understand where this is coming from. Perhaps not a "new" idea per se, but certainly one we haven't seen implemented yet...

]]> Semantic Web ora 2011-02-22T10:25:57-05:00 Wilbur is Dead - Long Live Wilbur! http://www.lassila.org/blog/archive/2010/02/wilbur_is_dead.html As many of you have quite correctly concluded, we are no longer actively maintaining Wilbur at Nokia. Today I have released what tentatively is the "final" Nokia release of Wilbur, available here as a gzipped tarball. The release is merely the current development snapshot that I have been using for some experimentation, but (and this may be significant to some folks) we also changed the Wilbur license to LLGPL.

The Wilbur toolkit is now at least 10 years old, and during its lifetime the project yielded one Ph.D and almost a dozen conference papers. It also served as a foundational experimentation tool and helped us develop a better understanding of the problems and solutions of RDF processing and storage. The spirit of Wilbur will continue to live on in the Piglet toolkit, released as part of the Smart-M3 system.

I will continue to answer questions about Wilbur should somebody have any...

]]> Wilbur ora 2010-02-14T18:59:12-05:00 SPARQL revisited http://www.lassila.org/blog/archive/2010/02/sparql_revisite.html I was happy to see that complex property paths are being considered as an addition to SPARQL as part of the new SPARQL 1.1 work. Lack of transitive closure earlier made me kind of lose interest in SPARQL for a long time, now things might change.

]]> Semantic Web ora 2010-02-02T04:43:32-05:00 Semantics for the Rest of Us http://www.lassila.org/blog/archive/2009/11/semantics_for_t.html We had a successful workshop at ISWC 2009 last week, titled "Semantics for the Rest of Us -- Variants of Semantic Web Languages in the Real World" (proceedings). Sandro Hawke from W3C gave the keynote talk, and we also had a panel discussion, chaired by Jim Hendler (my position statement slides are here).

Many thanks to everyone who participated (it was standing room only!). My particular thanks go to Lalana Kagal of MIT who did most of the work to get this workshop together.

]]> Semantic Web ora 2009-11-02T11:25:02-05:00 Status Report http://www.lassila.org/blog/archive/2009/03/status_report.html A couple of months ago I ended my 13-year tenure at Nokia Research and moved on to Nokia Services to work on technology strategy. Nokia Services is an exciting new unit of Nokia, responsible for the Ovi portal and some other services aimed at enhancing your mobile phone experience.

]]> ora 2009-03-27T07:53:29-05:00 PICKME'08 keynote http://www.lassila.org/blog/archive/2008/10/pickme08_keynot.html This may sound a bit like repetition, but this afternoon, I gave the keynote address at the PICKME'08 workshop, titled "Does Your Mobile Device Need to be Socially Aware? A Semantic Web Perspective". The workshop is co-located with ISWC 2008 in Karlsruhe, Germany.

]]> Semantic Web ora 2008-10-27T13:56:42-05:00 URSW 2008 keynote http://www.lassila.org/blog/archive/2008/10/ursw_2008_keyno.html This morning, I gave the keynote address at the URSW 2008 workshop, titled "Some Personal Thoughts on Semantic Web and 'Non-Symbolic' AI". The workshop is co-located with ISWC 2008 in Karlsruhe, Germany.

]]> Semantic Web ora 2008-10-26T05:03:12-05:00 Fulbright no more... http://www.lassila.org/blog/archive/2008/07/fulbright_no_mo.html Over the past year, we have had Mr. Mikko Perttunen from the University of Oulu visiting my group. Mikko is being funded by the Fulbright Program. He is now returning to Finland to continue his studies.

It has been an interesting year, with many interesting discussions on context-awareness, Semantic Web, pervasive computing, etc. Mikko's work on "audio contexts" is quite cool. We have an upcoming paper in UBICOMM 2008.

]]> Semantic Web ora 2008-07-25T09:46:20-05:00 Name Conflict http://www.lassila.org/blog/archive/2008/05/name_conflict.html After blogging about Piglet and mentioning my Python "wrapper" for Piglet, dubbed "pyglet", many people noted that the name is already taken. Thank you everyone for letting me know.

I hate picking names...

For now, I am renaming the Python wrapper "piglet" as well. We'll see how confusing that will be.

Some people also wrote to me about other RDF triple stores and toolkits written in Python. I should note that I am not implementing yet-another-Python-RDF-toolkit. Instead, I wrote a Python interface to the Piglet library (libpiglet) to allow us to start using it with our existing Python software. I think of it as an interim solution.

]]> Wilbur ora 2008-05-10T09:57:19-05:00 Piglet http://www.lassila.org/blog/archive/2008/05/piglet_1.html For some time now I have been working on "Piglet", a small RDF triple store written in C++, incorporating all those things we learned from building Wilbur first in Common Lisp and subsequently in Python. So far, I have a version ready that replicates functionality of the lowest layers of Wilbur. Piglet makes heavy use of SQLite3.

Piglet can be used as the back end for programs built on Wilbur. It merely replaces the parser (Piglet makes use of Raptor for this purpose) and the layers under the RDF++ reasoner and WilburQL path query engine. Eventually Piglet will have its own implementation of the path query language.

Right now, I am working on adapting Piglet for use as a Python extension module (working title: "Pyglet").

]]> Wilbur ora 2008-05-02T09:47:00-05:00 Switching to SBCL http://www.lassila.org/blog/archive/2008/04/switching_to_sb_1.html After developing on OpenMCL (and also Allegro) for a long time, I recently decided to switch to SBCL. My development machine (MacBook Pro) is still running MacOS 10.4.xx, and I just got tired of having to deal with the "32-bit or 64-bit" question. So I built a version of SBCL with multithreading enabled. It seems to work nicely, and I can now run all those things I like (MySQL client libraries, SQLite3, and Portable Allegroserve -- this is the one that needs the threads) without the aforementioned problem.

I am starting to like SBCL. I use Slime with Emacs (I switched from ILisp), and debugging etc. works reasonably well. SBCL is very fussy about the "purity" of the language, though (and I thought I was a perfectionist when it comes to how to write CL code -- little did I know...).

]]> Common Lisp ora 2008-04-29T14:27:25-05:00 Wilbur-and-O feed