Jeroen wrote:
The task of producing nice HTML / Printable versions of XML documents is further complicated by the highly verbose and somewhat unintuitive model of XSLT, which is presented as the most important tool for this task -- from the computer scientist purist point of view that might be true, but for many less gods, who think five lines of basic is already a lot, its functional programming model and verbosity is a real piss-off.
There is actually a fairly powerful "non-professional" alternative to the XSLT/XSL-FO approach to converting XML into PDF (or similar page-oriented layout): YesLogic's Prince product (soon to be at version 4.0 with optimized PDF output and embedded fonts -- wait until 4.0 is released in the next few days.) Prince uses the XML+CSS approach, and of course invokes the advanced CSS2 and some of the proposed CSS3 constructs. The founder of YesLogic, Michael Day, serves on the CSS Working Group of W3C, so he is quite aware of the power and limitations of CSS. Of course, there are a few knotty things that the current CSS2 cannot do, but YesLogic has added a few "custom" CSS constructs to fill in the voids, just as both Mozilla and Opera have (little known, btw). (I also want to add for those few here interested that the CSS parser in Prince is probably the best out there.) Now, I do agree that the absolute best outputs for print from XML sources via the XSLT/XSL-FO and Prince approaches require human intervention ("tweaking"), but the nice thing with a tool like Prince is that it gets one most of the way there, uses the slightly easier-to-use CSS, and allows for manual tweaking until the PDF is just right. Prince supports SVG and plans to add MathML support as well. They are a major supporter of the OpenReader System which I'm leading the development of: http://www.openreader.org . As an aside, for OpenReader I'm now building a supporter's/endorser's page, and any company, organization or individual willing to add their logo or name to the page, contact me in private email -- I'll send you the link to the current draft supporters page if you're interested in supporting/endorsing OpenReader. Maybe PG Foundation is interested? Greg? Michael? Btw, OpenReader plans to eventually natively support TEI-Lite (or maybe a well-defined subset of TEI or TEI-Lite) without need for conversion, including supporting constructs not supported in HTML web browsers such as inline notes and the like. Refer to the OpenReader web site for the details. Heck, we may even support ZML if it becomes popular as Bowerbird believes it will -- it'd be trivial to support ZML, actually (we'd internally convert it to XML and then present it using standardized CSS style sheets.) Jon Noring