W3C home > Mailing lists > Public > public-html@w3.org > September 2008

Re: ISSUE-54 (html5-doctype-vs-xslt): XSLT 1.0 can not generate HTML5 documents [HTML 5 spec]

From: Leif Halvard Silli <lhs@malform.no>
Date: Thu, 04 Sep 2008 20:53:45 +0200
Message-ID: <48C02EB9.6020400@malform.no>
To: "Michael(tm) Smith" <mike@w3.org>
CC: Jirka Kosek <jirka@kosek.cz>, Julian Reschke <julian.reschke@gmx.de>, public-html@w3.org, Henri Sivonen <hsivonen@iki.fi>, Ian Hickson <ian@hixie.ch>

Michael(tm) Smith 2008-09-04 15.40:

> [...] if we restrict the spec to only allowing <!doctype html>
> [...] then we have something very simple  for new authors [...]


Indeed.

 
> And the teacher -- if he or she wants to try to rationalize it 
> [...] might make a reasonable case that [...] it's just a way of 
> asserting that the document is meant to be conformant HTML" [...]

The only reason we do not drop DOCTYPE entirely, is to make it so 
that quirks-mode (always triggered by lack of DOCTYPE) formally 
gets linked to invalid/outdated code. Thus, it is a cornerstone in 
HTML 5's attemt to free us from the choice of rendering mode.

But I would argue that one could better make that rationalization 
if the docytpe was <!doctype html public ""> (public = official).

Do we, as you hint, need a rationalize-able spec? I think so. If 
both '<!DOCTYPE html>' and "<!DOCTYPE>" had triggered strict mode 
in all browsers, would we really have chosen the latter then?

> So now all those new authors would have to learn -- and their
> teachers would have to teach -- that, well, things are a bit more
> complicated than just <!DOCTYPE html> because, for certain cases
> that they really are not likely to have any good understanding of
> at the time they first learn it, they need to know that the
> doctype can optionally also be in the form
> <!DOCTYPE HTML PUBLIC "FUBAR"> (or whatever).


One could argue that the most pedagogical strategy would be to 
only nullify the FPI/URI option thing - and nothing more - as it 
is the choice of FPI/URI authors/teachers have struggled with.

Hence: '<!DOCTYPE html public "">', where "" was optional.

Based on history, such a solution would be simple to rationalize 
and explain. (A good teacher cannot forget about history.)

 From there, however, '<!DOCTYPE html>' could be permitted as 
well, rationalising it is as a synyonym to '<!DOCTYPE html public>'.

If there are plans, and good reasons, for making <!DOCTYPE html 
public ""> trigger quirks mode, then I can understand the 
reasoning behind not allowing it. (The HTML specs will for ever 
have to require UAs to respect the legacy mode triggerers.)

But if we do /not/ plan to define the 'public' tail as a quirks 
triggerer, then why dissallow it? To keep the option of using 
'public' as a symbol for HTML 6 or something?

Since all 3 doctypes [1] actually have the same effect, are easy 
to "rationalize" and helps XSLT, why only allow one of them? 
Comparably, the "XSLT-compat" idea only seems like a "forget your 
brain, just do as I say" proposal = appeal to authority.

[1] Summary of the doctypes I propose as synonyms:

<!doctype html public "">
<!doctype html public>
<!doctype html>
-- 
leif halvard silli
Received on Thursday, 4 September 2008 18:54:36 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:23 GMT