W3C home > Mailing lists > Public > www-html@w3.org > February 2005

Re: Nesting <HTML>...</HTML> tag pairs

From: Asbjørn Ulsberg <asbjorn@tigerstaden.no>
Date: Fri, 25 Feb 2005 19:11:53 +0100
To: neal.p.murphy@alum.wpi.edu
Cc: HTML List <www-html@w3.org>
Message-id: <opsmrl53fhlo81gp@quark>

On Mon, 21 Feb 2005 21:22:41 -0500, Neal Murphy  
<neal.p.murphy@alum.wpi.edu> wrote:

> That's fair enough, and is what I expected. My question remains, though.  
> Would it be worth ... enhancing the spec to allow HTML documents to be
> nested?

Since HTML documents should always start with a <!DOCTYPE>, wouldn't it be  
natural to extend it to allow nesting of <!DOCTYPE>s as well? If so, one  
afaik has to extend SGML as well to allow for this. And XML, to allow for  
it in XHTML. And either way, I'd say -- no, screa -- «not worth it»! I'm  
sorry, but this is a really, really bad idea.

The HTML specification should not take into consideration how people may  
abuse the language, even less so the language HTML is built on; SGML.

> Second, converting a random MS Word (or Word Perfect, OpenOffice or  
> other) document, a text document or an HTML document requires non-trivial
> effort; file formats are rarely openly defined.

OpenOffice's XML documents are openly defined and are being standardized  
as we speak. It's also rather trivial converting a heading of level 1 in a  
Word document to a heading of level 1 in an HTML document when saved as  
one. Making good HTML is really, really simple. And if Word was able to  
export to sane, valid XHTML, it would be simple to extract everything  
beneath <body> and plug it into the final document.

> To do so (converting documents) on a recurring basis is even less  
> trivial,
> because there are no guarantees that the source document will be  
> consistent
> over time.

Then this is another problem you have to tackle instead of trying to patch  
up all the symptoms it causes.

> Granted, in an ideal world, every web page should be written perfectly  
> from scratch.

No, not really. In an ideal world, all HTML producing software would  
generate sane, semantic and valid markup. And users not interested in  
learning HTML shouldn't have to. They should be able to use Word,  
Frontpage or whatever and be certain that the tool they are using isn't  
abusing the HTML specification in every possible way, but adhering  
strictly to it.

-- 
Asbjørn Ulsberg     -=|=-    http://virtuelvis.com/quark/
«He's a loathsome offensive brute, yet I can't look away»
Received on Friday, 25 February 2005 18:11:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:02 GMT