xhtml vs html 4 strict

[Jukka K. Korpela]

Re: xhtml vs html 4 strict

Jan Roland Eriksson <jrexon@newsguy .com> wrote:
[color=blue][color=green]
>>There is no SGML declaration for XML,[/color]
>
> I beg to differ.
>
> http://www.y12.doe.gov/sgml/wg8/document/1955.htm[/color]

It describes XML as if it were SGML, but in fact XML has been defined
independently of SGML - for rather obvious reasons.

It says: "XML documents implicitly contain the following SGML
declaration." That's hypothetical language, describing how XML _could
have been defined_ on SGML basis. (And this has some practical value
of course in software design.)
[color=blue]
> Without the 'Web SGML TC', and the work behind it, there would be
> no XML in the first place.[/color]

I don't think so. The marked demand for a trivialization of SGML was
too big.

--
Yucca, http://www.cs.tut.fi/~jkorpela/
Pages about Web authoring: http://www.cs.tut.fi/~jkorpela/www.html

[David Håsäther]

Re: xhtml vs html 4 strict

Jukka K. Korpela <jkorpela@cs.tu t.fi> wrote:
[color=blue][color=green]
>> http://www.y12.doe.gov/sgml/wg8/document/1955.htm[/color]
>
> It describes XML as if it were SGML, but in fact XML has been
> defined independently of SGML[/color]

I don't think so. XML is just a restricted form (a subset or profile)
of SGML. The SGML declaration is fixed, and many features that are
seldom used in SGML can't be used at all in XML
(http://www.w3.org/TR/NOTE-sgml-xml.html lists all differences AFAIK).
I don't think this is anything "SGML people" disagrees on, as with the
case of "HTML as an SGML application".
This is my understanding at least.
[color=blue]
> for rather obvious reasons.[/color]

What would those obvius reasons be?

--
David Håsäther

[Andy Dingley]

Re: xhtml vs html 4 strict

On Thu, 02 Jun 2005 03:18:27 +0200, Jan Roland Eriksson
<jrexon@newsguy .com> wrote:
[color=blue]
>We are going downhill from there...[/color]

But XML (by volume, out in the real world) isn't replacing elegant SGML,
it's replacing nasty unreliable CSV.

I've never used SGML. I once had an obvious "SGML project" (1997) and
went shopping for SGML vendors to sell me tools to make it all work.
Even with a blue chip budget behind me, I couldn't find a viable SGML
solution that was affordable, completable in a reasonable time, and
comprehensible. . Did it with PDFs and string in the end. These days I'd
throw XML at the problem and have it done in a week.

There never _was_ an SGML "golden age". Maybe SGML always was "right",
but it wasn't useful (outside Boeing).