Older blog entries for Uche (starting at number 9)

I added full-text search to my Python/XML and 4Suite "Akara" sites (my document collation projects). As I predicted, it took less than an hour.

Dave Carlson pointed me to his article: Modeling XML Applications. I like his thinking in general, though I do have some specific qualms. For one thing, I don't think XML schemata are really solid information modeling bases for the model in the MDA. They're text tree modeling systems, with primitive type inferencing, and we need something more powerful. I, of course, advocate RDF, but even Topic Maps are more suitable. I notice that Dave credits WXS for being the message content model system built into SOAP, but it's actually not WXS but a specialized SOAP serialization, which has some superficial similarity to WXS but is fundamentally quite different.

All business today, pretty much, except for a break to see The Bourne Identity. I highly recommend it, especially for old-school Robert Ludlum fans. That is, for people, like me, who think Tom Clancy basically writes cowboys and injuns books and that John le Carré can't seem to put down the foie gras. The film actually does rather pull it off: I didn't expect even the most virtuoso film making to approach the richness and nuance of the book. While I'm on film, I recently went to see Minority Report. I should have known anything with Tom Cruise in it would be too slick to do justice to Philip K Dick (after all, look what they did to Abre los Ojos).

I contributed Versa examples to RDF Query and Rule languages Use Cases and Examples survey, announced just today, though only the first 3 examples I posted seem to have shown up so far. I think this site could be a good idea with some tweaks (especially giving better definitions of the RDF models being queried).

My Recipe: Merging XBEL Bookmark files has appeared in the Python Cookbook, though the note at the end with sample XBEL files for merging seems to have been corrupted in production. See the comment I added which addresses this.

Craeg Strong pointed me to Metacrap: Putting the torch to seven straw-men of the meta-utopia. It's a good hazing for folks in semantic/intelligent Web communities. Unfortunate that the author chooses the term "straw men" and thus makes it all too easy for detractors to knock him off the log, but I think some of these are points I've been trying to make for a while now. Sometimes an off-hand irreverend statement is worth loads of earnest advocacy. Points 2.1, 2.2 and 2.3 are the motivations of my article The Languages of the Semantic Web. Point 2.4, I think, is a dud: Any Intelligent Web must be a marketplace of ideas. Stupid people will simply lose, just as they do in today's Web. I've put points 2.5 and 2.7 into practice in projects involving ontologies. My motto in these has been: support diversity and disagreement, or fail. I didn't really follow point 2.6.

Speaking of my Sem Web article, looks like it got very kind words in a recent SearchDay issue.

I found my Intel Developer article The XML Menagerie. It discusses a variety of XML technologies, including WXS, RELAX NG and XLink/XPointer.

I also started on an Akara site for 4Suite, focusing on the RDF and respository features.

The Akara stuff is fun to put together, and I hope shaping out to be useful docs. I need to implement sophisticated ordering of some sort in the RDF query to organize the indices a bit more. I also have to add search (should be a snap) and some better style. I'm also contemplating the annotation features of crit as a possible addition. Thanks to vdv for pointing it out.

My document collation system has a name now: "Akara". WikiWiki became MoinMoin in its Python incarnation. Moin Moin is, in a large swathe of West Africa, a meal made from soaked and ground black-eyed peas boiled in water. "Akara" is a similar food, but made from the bean paste fried in groundnut oil. Impeccable nomenclature, right?

Anyway, more importantly, my Akara site for PyXML now has quite a bit more content. For instance, I cover all the relevant entries in the ASPN Python Cookbook. Much more example code to have. Pick it up!

Posted my first Advogato article. Warning: it is not on a technical topic :-)

Yetsrday I resurrected another old article SVG: What comes after XML? from the graveyard of ITWorld. Interesting to see the progress SVG has made these two years later.

A query from someone looking for code for merging XBEL files caused me to dig up an old proposal I made for XBEL merging rules. I think I'll try to build something with this into the BookerT demo that comes with 4Suite. The correspondent also found an old bookmark merging script of mine as well, but this looks too byzantine not to just rewrite from scratch :-).

Eric van der Vlist has announced XVIF, his Python implementation of RELAX NG with extensions for processing pipelines as part of the validation process. Some people have expressed concerns about this combination, but I have long argued that what I call a processing plan should be one of the outputs of XML Schema. I think Eric's design feels just right.

Slow day. I put up this Paper on the use and architecture of 4XSLT (a PDF in Dutch) by Davy Friedrich, Joe Achten, and Wim Deprez.

It seems my brother, Chime, just got an article published in XML.com: Editing XML Data Using XUpdate and HTML Forms. Great stuff, though I'm biased. Micah Dubinko did lament that he didn't cover XForms, but you can't get it all in one article.

I also found my recent column in the Application Development Trends Web page: The many heads of XML modeling.

I also began a document collation instance for 4Suite, though it's not ready for consumption yet.

My article Tip: Namespaces and versioning went live on developerWorks.

Jürgen Hermann (jhermann) set up an XML-SIG Wiki. This is excellent, as the XML-SIG desperately needs a conveneient place to stitch together some resources and documentation in order to make its products less opaque to users.

Unfortunately, I've had to go a bit lightly on my MoinMoin explorations. The wiki parsing code is very tightly coupled with the wiki web architecture code, so it's not suitable for my documentation collation project.

Speaking of this documentation project, I did end up hacking up some wiki parsing code for 4XSLT as a quick way to move forward, so it's much more respectably formatted now. I also added some new content. I'm also looking for a name for the RDF/Wiki framework on which it's based. "Bazaar" is a recent thought. Or perhaps "Afor", the Igbo market day non-Igbos are most likely to pronounce correctly :-).

The simple little 4XSLT extension element I wrote for the wiki conversion is implemented in this module. It takes wiki text in a "select" attribute and emits HTML to the processor output stream.

The latest installment of my Thinking XML column posted today: XML meets semantics - Will XML live up to its promise?.

Today I completed editing version 0.4 of the Versa spec. According to the revision notes:

Add string functions. Add filter expressions. Add find-regex function. Add quote escaping in strings. Remove all()[] form. More clean-up.

I also updated the versa-by-example tutorial, adding lessons 9 and 27, making corrections and style changes to other lessons. Finally, I implemented an optimization in 4Suite's Versa implemention for metadata search engine use cases.

Added the 4Suite project page (and a dead-end "4SUite" one through a fat-finger error)

Took a look at the MoinMoin Wiki clone, which Jürgen Hermann pointed out after reading yesterdays's mention of a Wiki parser. I need to be smacked. Jürgen is a long time user of 4Suite, and has mentioned MoinMoin before. The MoinMoin parser code looks especially useful. I hope to mod it to emit docbook or XHTML as well. I think I might also set up a 4Suite MoinMoin.

I'm still torn on whether to update my Red Hat on my laptop, or to go to Debian.

etc:

Watched Germany basically beat Cameroon through sheer composure. That ref was the most officious bastard since Savonarola. Senegal/Uruguay was priceless entertainment. No poem memorized today: slacker that I am.

Currently I'm working on more flexible server config features for 4Suite. I just added a 4XML command line last night. I also added a "Rule" server config option. I refactored my personal Web site to use this option to avoid double-indirects, which were choking MSIE and Opera because of bugs in those browsers.

I'm also working on a little documentation collation project for Python and XML tools. It's sort of a mix between a FAQ and a HOWTO, covering many common processing tasks with references to external sources of info, and local brief examples. All powered by RDF, of course. I need to write a Wiki Markup parser in order to make it usable. amk pointed out to me that The ActiveState Python Cookbook has 13 XML-related items

I'm also working on the next iteration of Versa, the RDF query language we've been working on at Fourthought.

Finally, I'm working on a 4Suite XML/RDF repository tutorial for IBM developerWorks.