On 10/20/2012 1:52 PM, James Adcock wrote:
All that has happened is that Lee has once-again confirmed that there are two camps of belief among readers of this forum.
Boy, do I wish that were true. In fact, there are almost as many "camps" on this list as there are participants. The largest group believes that presentational markup of PG texts is unnecessary or even wrong. This group apparently feels that if a human being can intuit the structure of a text nothing further is desirable--and consequently requiring that a human being always be "in the loop." These texts are not usable by humans or computers, they are /only/ usable by humans. There is an almost equally large group of users who feel that presentation is very important, but that a single presentation is adequate--indeed absolutely required--for each text. This group is highly fractured by the fact that each member seems to think that his/her own vision of what that presentation should be is perfect and inviolable. A third group believes that every PG text ought to be marked up in such a way that various formats (although not necessarily other presentations) can be derived from the master copy, but that the master copy cannot be usable by itself in any meaningful way. I believe that the argument of how best to present a text is simply irrelevant. No two people will agree on the best presentation, nor should they. The power of computers gives us the ability to easily and conveniently create as many "vanity publications" as we want, and to do it /algorithmically/. I believe that a text can be marked up using HTML in such a way that, 1. it can be viewed in any HTML user agent with an adequate, even if not perfect, presentation; 2. it can be combined with one of a number of external CSS files to create a presentation good enough to satisfy any individual reader; and 3. it can be programmatically transformed into any other non-HTML based format (e.g. PDF, RTF). Unfortunately, I don't believe that there is anyone on this list who agrees with me, so please don't accuse others of holding beliefs that are mine alone. But so far, no one has offered any evidence that I am wrong, so I guess I will go on being a "voice crying in the wilderness."
One group believes that they can simply do semantic markup "of everything" and then someone else can apply the formatting decisions later and everything will just "magically work."
One of the many things that Arthur C. Clarke is know for is the statement that "Any sufficiently advanced technology is indistinguishable from magic." Now I'm a programmer, I've read the specifications, I've participated in working groups, and I've written thousands of lines of code to manipulate and display HTML, CSS and ePub files. I can see how you might see browsers as magic, but to me it's just technology.
The other group doesn't believe this works, that on the contrary someone needs to make good "engineering judgment" decisions on formatting now rather than expecting someone else to do it later, and that in general there is not a limited set of semantic markups that can be uniformly applied such that the problem can be well-divided and the "engineering judgment" decisions about formatting can be made later.
I call this group the "ill-informed nay-sayers."
This second group says: "Hey, if the first group cannot even show good 'engineering judgment' about displaying their own semantic markup in an attractive and sensible way using ONE set of formatting decisions -- namely their own -- then how likely is it that their semantic markup is going to prove to be useful to someone else implementing a different set of formatting decisions at some latter date?"
Thankfully, this criticism does not apply to me, because the example I provided of McGuffy's reader is presented in an immaculate, pleasant, elegant and sensible way. It is inconceivable that anyone could not view that presentation as the epitome of fine markup.