|Bert Bos writes:
|> I don't think it's a dream at all. I predict it will be reality sooner
|> than you think.
|
|The problem is that we've been saying this for over a year now, and
|I can't tell that it's any closer to reality now than it was then.
|
|Marc
If I'm permitted to take myself as an example:
I routinely test all my HTML pages with sgmls. In fact, many of them
are generated (i.e. translated) by sgmls, because I don't like HTML
and am writing in HTML+ or my own local adaptation of HTML most of the
time.
It's not only to deal with the variety of browsers out there, but also
because I like to make hardcopy of what I write. I do that by having
sgmls (& sgmlsasp) translate HTML to LaTeX. To make that work, the
document must be fully and correctly tagged.
I even maintain some Unix manual pages in HTML, using sgmls to
generate the nroff markup.
I agree that installing sgmls is not an option for most people, except
the few that know enough about SGML. But the suggestion by somebody
that browsers do not ignore markup errors, but report them (in a
friendly way, of course), could be an alternative. In my "vision" of
the future many, if not most documents will be multi-purpose, and that
means that people will simply have to write (correct) SGML.
Bert
--
_________________________________
/ _ Bert Bos <bert@let.rug.nl> |
() |/ \ Alfa-informatica, |
\ |\_/ Rijksuniversiteit Groningen |
\_____/| Postbus 716 |
| 9700 AS GRONINGEN |
| Nederland |
\_________________________________|