- From: Bijan Parsia <bparsia@cs.manchester.ac.uk>
- Date: Mon, 16 Feb 2009 00:35:56 +0000
- To: www-tag@w3.org
Replying to: <http://lists.w3.org/Archives/Public/www-tag/2009Feb/0120.html > Jeff, my email address is <bparsia@cs.man.ac.uk>. I understand you were "me to-ing" some of Thompson's statements, but you did claim a certain epistemic status in virtue of your particular experience. I dispute that status. I also dispute both Thompson's sentiments(?) and the evidence he's thus presented. And yours as well. :) For example, I don't think there is a "notable" lack of people advocating XML5. Indeed, it's hard to draw much from the advocacy of a technology, but it's especially hard to draw conclusions about the usability of a technology by the lack of advocacy of an easier one. The questions are largely orthogonal. For example, from Thompson's email: ""But I think the world has already voted with its feet on the XML5 question, in that there is a notable _lack_ of folk advocating it. And there's good reason for that: XML actually _is_ usable by authors and authoring well-formed XML is _not_ hard.""" Huh, what? There are many many explanations that are easily at least as good as this, if not far more likely, than it is usable. It's, in fact, weird to appeal to such an abduction. You claim, based on your teaching experience, that it is not hard to learn to create valid and wellformed XML consistency. There's a lot of evidence against it as a general claim and without specifics of your sample, its hard to draw sensible conclusions. Indeed, there should be, contra what seems to be the presumption here, a pretty strong presumption *against* XML's ease. (Just on general grounds of how people fare with formal languages; how we handle error in natural language, etc.) In other words, methodologically, it seems safer to presume that people will do poorly at authoring well formed XML, and then be surprised when they don't. Of course, it's a good idea in general to play devil's advocate to your own position. By the by, I'm not convinced that XML5 per se is a win. I've not worked it through at all. But shared, defined error handling below the well-formed layer seems worth pursuing. Cheers, Bijan.
Received on Monday, 16 February 2009 00:36:31 UTC