Subject: Re: [xsl] metrics for evaluating xsl-t?|
From: Tony Graham <tkg@xxxxxxxxxxxx>
Date: Wed, 23 Aug 2006 11:08:05 -0400
Wendell Piez <wapiez@xxxxxxxxxxxxxxxx> writes: > At 09:03 AM 8/23/2006, you wrote: >>The future conference paper that I'd like to see (and present) is the >>one with lots of manager-friendly graphs relating schema complexity >>and XSLT complexity from a survey of real projects. If that existed, >>it would be easier to explain to your manager why a new stylesheet for >>a new, complex schema could take longer than a week to write. > > Maybe. Or maybe just convince him he understood, or give him a face-saving > reason to take your word for it. In my experience, it just has to be colourful. > But the graphs could be generated out of XSLT. It sounds like a fun talk. So I can ask you for XML-to-SVG help when the time comes? > Badgering you from the floor, the wrinkle I'd complain about is that the > complexity of the XSLT can't really be correlated directly to the complexity > of the schema. A simple schema with complex and variable usage could require How do you know until you measure both? > complex XSLT. A fairly complex schema whose usage is well-enforced and > consistent might allow certain useful tasks to be done with simple XSLT. So > the real-world relation of the schema to the dataset, as well as the nature of > the task to be done, are both critical variables. Or? My educated guess is that there would be clusters of similar results, with the occasional unusual result. Inasmuch as many shops would have multiple distinct stylesheets for data marked up with the same schema, of course there would be variations. And, as I said in a recent post, I would expect different clusters in the measured complexity of XML-to-XML, XML-to-(X)HTML, and XML-to-FO stylesheets. There might even be trends in the complexity of XML-to-SVG stylesheets. Regards, Tony.