Grr, therein lies the problem entirely with HTML:
How do I know if that H1 is the title of the page, or just something a person
wanted really big?
How does one know if that "<P>" tag indicates a block of addresses, a break
between paragraphs, or what?
How do I know if the "<pre>" tag indicates a block of commands to be executed
by the user, or just a chunk of a note someone threw into the page and didn't
have time to mark up?
The key thing is: what audience are we targetting with our documentation? If
the audience is *solely* web-based, yeah, standardizing on HTML is better
than nothing. But what if someone needs to print them out? HTML looks
butt-nasty when printed, I'm sorry. What if a publisher wishes to reprint
the documentation in book form? He will have an enormous time investment by
someone to convert it to a format the publisher can use. What if you want to
implement a search engine? The engine will pick up titles, meta tags,
comments, etc. and have no clue what type of data it is looking at, thus
continuing to make the Web more of a morass than it already is.
Semantic markup is vital to progress -- although I understand the concerns
that people should not be required to write in DocBook (and I quite agree),
it should be allowed and supported because those DocBook XML authors have
invested the time to make the world a better place : )
Mozilla supports XML, too, if I recall correctly -- DocBook is also an XML
DTD and, if the DTD is correctly installed on the users computer, should be
readable...
--
Matthew P. Barnson Manager, Systems Administration
Excite@Home [EMAIL PROTECTED]
"There is no spoon" -- Neo
