- From: James Graham <jg307@cam.ac.uk>
- Date: Tue, 04 Jan 2005 15:29:16 +0000
Bill McCoy wrote: >I find it odd that some WHATWG members argue that transmitting arbitrary XML >data over the wire to clients is a bad idea ( >http://ln.hixie.ch/?start=1064828134&count=1 )given that the solutions >already widely employed in Street HTML - encoding data in JavaScript arrays >and other tricks - are so much less "Semantic Web" friendly. To pick an >extreme example, Google folks admit that search-indexing a Gmail screen >would be extremely challenging (they themselves index the raw data of >course, but that's the whole point - it's much easier to process structured >data before it's been compiled into a presentational form that is partially >programmatic code). The WHATWG proposals to promote scripting, rather than >declarative markup, as the basis of forms data binding and validation, and >to dispense with an XML-based data model separate from presentation, would >seem likely to worsen this situation by leading to more Gmail-like >JavaScript-centric web applications. > > It's not only an extreme example, it's a terrible example. Google have repeatedly shown that they have no interest in using the semantics available in HTML, let alone whatever additional complexity is provided by XForms - just see any Google search results page. This is clearly not an inability to understand the benefit to the client of good markup and so must be a conscious decision on the part of Google. So the fact that GMail is hard to index is just as likely evidence of malice as it is evidence that better data/presentation separation is needed. Even allowing that your example is misguided, I'm not sure I believe the rest of the argument either. In order to believe that HTML should be starved in order that XForms should flourish, you have to take the position that there will be a migration to XForms (and XHTML) from HTML. In order to believe that this will offer a significant advantage to the semantics of the web, you have to believe that authors will tend to use the new features offered by XForms in the way that they are designed to be used. In practice, I'm thoroughly unconvinced of the first and skeptical of the second. At the moment, there is simply _no_ way of migrating away from "street" HTML. XHTML is fundamentally incompatible with "street" HTML because the processing requirements of HTML are satisfied by a negligible proportion of all HTML documents. XHTML is not supported in any meaningful way by the world's most commonly used web browser. There are few existing CMS solutions that are able to produce XHTML reliably, let alone arbitrary XML, and few users who understand the technical requirements of XML. The combination of all these factors makes it very hard to migrate away from HTML to anything that is not explicitly compatible with it. Even if people could migrate away from HTML, it is doubtful that they would suddenly learn the benefits of sending semantic documents over the wire, nor the ability to understand complex specifications well enough that they would produce high-quality documents. I would expect that in XForms, as in HTML, if there is a way to hack a solution that looks/works as expected then authors will use that at least as often as they use the neat solution provided by the language. The major advantages of the WHATWG specs over XHTML/XForms are 1) Backward compatibility. They can be used in situations where there is existing content that must be integrated and a limited migration budget (i.e. most real world situations) 2) Limited additional complexity. Because Web Forms 2 is close to HTML 4 and, in particular, because it does not switch to a model in which everything is done declaratively, it is familiar to existing authors and, as a consequence, is more likely to be understood, adopted, and used in a way compatible with the specification.
Received on Tuesday, 4 January 2005 07:29:16 UTC