David Starner wrote:
On 9/27/05, Lee Passey <lee@novomail.net> wrote:
To be honest, I haven't looked at any of these competing PDF expressions of Alice, because I can't see how PDF has any value at all to an end user.
I spend enough time sitting in front of a computer; I'd love a nice looking printout that I can take and read where ever I want. What end users want varies quite a bit.
I spend virtually no time reading while sitting in front of a computer (if you don't count figuring out someone else's source code), except when I am forced by circumstances beyond my control to refer to a PDF file. For _all_ of my recreational reading I download the file to my hand-held device. For many people (and you appear to be one of them) the hand-held device consists of an ordered collection of sheets of paper. Your comments are so obvious, and so correct, that you shouldn't even need to make them, yet it is surprising how often they are forgotten: WHAT END USERS WANT VARIES QUITE A BIT! And this is the problem with PDF. Having a PDF file does not necessarily mean you will get a nice looking printout, it only means you will get a printout that looks like what the document author wanted you to have. If you find widows and orphans disconcerting (and by this I mean typographical widows and orphans, not those caused by mis-guided foreign policy) you will probably not be happy with the Perathoner XSLT to PDF version of Alice in Wonderland, and if you don't like monospaced, serifed fonts you will probably not be happy with the BowerBird ZML to PDF version. Despite Mr. Perathoner 's assertions that any styles which don't match his criteria can be dismissed as bad taste, it remains virtually axiomatic that _de gustibus non disputatum est_. The problem of finding the styling that satisfies the plurality of people ("the greatest good for the greatest number") only exists when, as with PDF, the ultimate rendition is in a fixed, immutable form. And the best way to avoid the problem is to postpone the rendering to the last possible moment, preferably when the "greatest number" has been reduced to one. XSLT is a good method of postponing rendering. Indeed, if everyone had access to a tool whereby you could easily mix one part document (in a master file format) with one part XSL script (of the user's choice), shake well, and end up with a result reflecting the end user's preferences and in the format best suited to the end user's tool set (including an ordered collection of sheets of paper) XSLT may even be the best method. Tools to perform XSL transformations, however, are still far from widespread, and the last time I used them (about 2 years ago) I couldn't even find one that was a complete enough implementation of the spec to do some of the things I wanted to do. XML+CSS is also a good method of postponing rendering. CSS is certainly not as powerful as XSL (being, as it is, merely a style sheet as opposed to a scripting language), but support is much more wide-spread than support for XSLT, permitting end users to do the very kind of mix- and match-ing that I envisioned. Using tools like YesLogic's Prince you can even go from XML+CSS directly to high-quality print or PDF. When it comes to a master file format, what should be selected is the one which (1) permits end-users to postpone rendering decisions (what many people refer to as 'tyopgraphy') to the last possible moment, which (2) allows the end user to have the maximum amount of input as to the which rendering decisions are made, and which (3) permits the end user to use the widest range of tools possible. ZML certainly does not satisfy these criteria, and PDF is even worse. At the moment I believe that TEI+CSS best satisfies these criteria, but this could change as new technologies come on line. But arguing about whether line lengths should be 66 characters, or 30 em, or 11 words or 10 cm (as suggested in "Huey, E. B. (1968). _The psychology and pedagogy of reading_. Cambridge, MA: MIT Press.") is completely irrelvant. Line lengths should be whatever the reader wants them to be, and we should try our best to give the reader that choice.