Some notes from day two of the XTech conference.
A more mixed selection of talks for me yesterday:
- Directions of the Mozilla RDF engine — Axel Hecht gave a low level look at how the RDF parser and APIs in Mozilla were going to be brought up to specification and optimised. Some new APIs will appear, but backwards compatibility seems to be a primary goal
- The convergence of structure and chaos — Paul Prescod gave a talk on the benefits of using wikis in concert with content management systems
- The Application of Weblike Design to Data – Designing Data for Reuse — nice talk from Matt Biddulph on the benefits of information modelling and good URL design when designing both web sites and web services. I predict that information architecture for web services is going to be a hot topic before long; IA’s have a lot to contribute.
- SKOS: A language to describe simple knowledge structures for the web — an good introduction to SKOS from Alistair Miles; I plan to take a closer look at the vocabulary, take a look at SWED for an example of how a facetted browsing interface can be built on a SKOS dataset
- Beagle: Free and Open Desktop Search — fun talk from Jon Trowbridge. Pleased to see that Beagle is being ported to Windows
- Topic Mapping The Restoration — Kal Ahmed discussed the process involved in creating a large topic map based on Samuel Pepys’s diary. Interesting on many different levels
But the two talks that really dominated yesterday and had the halls a-buzzin’ were Steven Pemberton’s on the design principles and new features of XHTML2, and Ian Hickson’s review of the work of the WHATWG to propose extensions to HTML4 and the DOM.
I’ve previously been concerned about the direction of XHTML2 as it seemed to be getting further and further away from the daily issues in web development.
Pemberton described the design goals behind XHTML2 which were to: use generic XML features where-ever possible; focus less presentation and more on structure, usability; accessibility; better I18N; better device independence; forms improvements; reduce need for scripting; and (most interesting to me) better semantics.
Pemberton observed that one of the difficulties in achieving movement on (X)HTML is in getting buy-in and consensus from all the differing communities that each believe XHTML is their standard. He then went on to summarise the pragmatic decisions they’d made in attempts to address the needs of each community.
I won’t summarise the details of all the new features, but the new support for metadata is very interesting. Micro-formats look like they’ll get a boost from XHTML, but with a standard (and clearly defined) interpretation of XHTML (via GRDDL) as RDF, these should now be fully integrated with the semantic web. Pemberton joked that with XHTML2 we don’t need RSS anymore.
Pemberton’s session proceeded Hickson’s and the room remained packed for both, although it got lively (in both the room and the IRC channel) during Hickson’s presentation which focussed primarily on the number of new HTML extensions (“HTML 5”) that the WHATG are proposing.
I was sympathetic to the general aims (making web development saner), but found myself inevitably drawn more to the structured approach of the HTML working group than the ad hoc HTML extensions that the WHATWG are specifying. I’m concerned with the profusion of new elements being added; I saw the shades of “marguee” and “blink” lurking in the corner of the room. Hickson was quick to point out that unlike previous extensions theirs were being openly specified and had backing from several big browser vendors: Apple, Opera and Mozilla. All of which are shortly shipping support for some/all of the features.
The main area of discomfort, and the focus of more than one corridor debate in the afternoon, was the relationship between the WHATWG and the W3C. Both sides (a number of W3C luminaries were in the audience) refused to be drawn on the topic. The W3C team comment on the Web Forms 2.0 submission contains some telling comments: “[we] strongly recommend that future work in this area should be in collaboration with the HTML and XForms Working Groups”
See the Q&A that I summarised in the Wiki for a flavour of the audience’s questions.
Oh, and in case you’re wondering, we ate Japanese but didn’t
find time for nun bowling.
xtech.
Lost Boy 
