Re: standard terms [was: Re: Trial A]

IJ:: "We've spent a lot of time discussing the term "content" in the

WL: And in the ATAG, ERT, GL, and a bit in the EO Working Groups. The
"conclusion" reached is essentially that even when we start talking
about pinning it down, there is no agreement that it can be pinned down.
The very title "Web Content Accessibility Guidelines" might be a case in
point. A major problem is that "content" is a word whose meaning is
without "ownership". A similar anarchy occured in ATAG with the word
"prompt" which has a fairly specific meaning in software engineering
circles and a rather different one in everyday conversation.

By putting one (actually two in the cited glossary reference!)
particular interpretation of how the term is used in one document just
won't have any effect on people reading this stuff who *know* what is
the "semantic content" of the word "content" - it means different things
at different times and in different contexts to different people under
various circumstances, if you take my meaning.

The little exercise attempted at: using
"content" was undertaken not in the sense of rigor demanded by a formal
document but as a means of explaining what to a great many people not
privy to our counsel what we mean when we (rather often and variously)
say "separate content from presentation" and all its cousins also
involving such things as "structure", "semantics", "styling",
"appearance", and probably other words that most people think they know
how to use (what they "mean") but are encountering in, to them, strange

To a great many people "presentation" *is* "content" as is "structure".
I don't expect to resolve the niceties of how to express all the
relationships in a "throw-away" note, just to begin making a few dark
things clear.

Ian says "However, we've spent a ton of time on what "content" means and
I would urge you not to assign meaning to the term other than the
document object is constituted of content." To which William replies
that the intended audience for "Trial A" comes from a population that
has spent statistically hugely more time deciding what "content" means
and they couldn't care less what we decide <g>. Further if one does a
search of all the WAI and various ML worryings about this matter it will
be seen to resist the sort of categorization we might seek. We might
call it the "content of content" problem. Or even the "'content' of
'content'" situation. 

I appreciate the advice but in the instant context I think the main
result of further messing with this particular word will be further
obfuscation in the very area I intended to clarify. Parenthetically (a
Table of Contents is often in fact a compendium of Structures with no
reference to "content"!) when I asked Prof. Goldfarb, whose urge to the
assembly at the WAI kickoff breakfast in Santa Clara was to take this
(what may be one of the last) opportunity to evangelize the separation
of content from presentation, what the SGML originators thought of this,
he made it clear that it wasn't all that clear. At least I think that's
what he said, or I'm pretty sure that's what he said, or maybe the
"Trial A" piece, Ian's email, Al's reply, and this somewhat convoluted
and self-reflexive response are reasonably probative examples of why
this has the same order of unresolvability as whether {ALT="" and ALT="
" discussions} have led to any conclusion? 

To the audience I thought I was trying to reach 


and I hoped to ease them into an awareness of what some of our jargon 
means and how it applies to enhancing a usable/accessible/semantic "Web
of Trust" and all that other high-sounding stuff.

Thank you all (mostly in advance) for your comments, suggestions,
encouragement and counsel. Keep them coming because one of the main
problems we encounter (particularly in GL and E & O) in getting visitors
to our documents to avoid CUS (cryptic usage syndrome) and ADE (arcane
definition effect).

I will bet that a landslide-sized majority of people who use <H1></H1>
not only think of it purely as a formatting shortcut, but aren't even
aware of what, in its context, "structure" means. It's just easier than
putting in all the font-size, alignment, etc. tags that come free with
that element. Loser buys the first round in Bristol.


Received on Saturday, 26 August 2000 20:57:05 UTC