[
Lists Home |
Date Index |
Thread Index
]
- To: "Bjoern Hoehrmann" <derhoermi@gmx.net>
- Subject: RE: [xml-dev] What is the rule for parsing XML in a namespace inside HTML?
- From: "Joshua Allen" <joshuaa@microsoft.com>
- Date: Tue, 13 Jul 2004 19:29:55 -0700
- Cc: <xml-dev@lists.xml.org>
- Thread-index: AcRpRaqmOfPdKxc7RA60iQyPDQIdUwAA/AVQ
- Thread-topic: [xml-dev] What is the rule for parsing XML in a namespace inside HTML?
> >First, the poor support for XHTML extends far beyond Microsoft. The
> >market for tools which consume and produce HTML is vastly more
mature.
> emit things that clearly ain't HTML. So, is this just your impression
> or did you do some representative research on this matter? I did not,
There is no point even having a conversation with someone who tries to
argue that the market for XHTML tools is even close to as mature as for
HTML.
> web that do not pass the W3C MarkUp Validator seems to tell a
different
Staggering lack of user-empathy. 99% of people writing web pages do not
care whether it validates in some markup validator. This is not what
people care about.
> How is XSLT exactly relevant here? As far as I understand, XSLT would
> serve to transform some XML to some other XML or other format, what
> format do you have in mind here? Surely it would be some kind of
> recognized markup language that provides means to link other documents
> or to offer the user means to communicate with the site in some sens,
> e.g. through forms, so it seems that you would send some arbitrary
> markup to the browser only to have it transform it to HTML or XHTML,
> so, why not just deliver that HTML/XHTML to the browser?
Ugh; I'll repeat myself verbatim: "Users should either stick with HTML;
or if they want to have a pure machine-processable architecture, move to
XML+XSLT+CSS."
How is that confusing? If you want to render it in a UI, use HTML. If
you want to parse it using some machine process (like a news aggregator)
use XML. XHTML is just some crazy in-between Frankenstein. It is
neither good at machine processing (try to build an ecosystem of news
aggregators on the brilliant idea of screen-scraping RSS from XHTML),
nor is it good at rendering in user agents. Choose the right tool for
the task -- in this case, the right tool will be either HTML or XML; not
some abominable hybrid.
Next we will have the nabobs crying that PNG is inferior because it is
not well-formed XML...
|