Re: Versioning re-visited (was : mixed signals on "Writing HTML documents", tutorial, etc.)

Stephen Stewart wrote:

 > Maybe you're making a wider point which is lost to me (I use boilerplate
 > and I'm too lazy to think about that) but how does shortening the
 > doctype & making it simpler, or even getting rid of it and the <html>
 > tags for that matter, make any difference to the quality of what we do
 > here or what authors do with the final spec?

Nothing we do can significantly affect what authors do with the final
spec (unless we make it so incomprehensible that none are able to
understand it) but the "quality" of what we do here is very much
affected by versioning or the lack thereof.

If we have the humility to accept that, no matter what we agree,
hindsight will allow those who come after us to make improvements,
then it is vital that documents created based on /our/ work identify
themselves as such.  If they do not, and simply claim to be "HTML",
then at some point in the future a document that is today "Valid HTML"
will cease to be valid and/or its semantics will change because the
specification has changed (something added, something removed,
something altered).  What we are therefore doing, if we insist that
every instance of an HTML 5 document start with (at the very least)

	<!DOCTYPE HTML 5> [1]

is ensuring that, when at some point in the future that document
is rendered, there is a formal specification with which it can
be compared, and from which both the syntax and the semantics
can be unambiguously determined.  Without versioning information,
documents that we write today based on HTML 5 will, at some point
in the future, become either invalid or meaningless or both, because
such documents will contain no unique reference to their syntactic and
semantic specification, and the specification will by then have mutated
so far from that on which the document was based that it will no longer
be possible to deduce the document's intended meaning from it.

Of course, those who come after us may well once again realise
that versioning is crucial to ascertaining a document's
validity and meaning, and may well re-introduce it even if
we drop it, in which case documents starting

	<!DOCTYPE HTML>

will be easily identifiable as HTML 5 and treated as such.
But if those who came before us believed in the importance
of versioning, and if those who come after do the same, then
perhaps we should just accept that they are probably right
and retain versioning without further debate :-)

Philip Taylor
--------
[1] I would personally far prefer to continue with the more formal
DOCTYPE/DTD specifier, as in

	<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 5.0 Strict//EN">

assuming that we do indeed ultimately produced a DTD (yes, I know
this is far from certain, but I for one would be very happy if
HTML 5 continued to be based on SGML rather than apparently being
designed /ex nihilo/).

Received on Thursday, 21 June 2007 22:01:32 UTC