[Typo3-dev] XHTML compliancy, old TypoScript and beyond..

[Christopher]

Hi Michael,

--- Michael Scharkow <mscharkow at gmx.net> wrote:
> [this is basically a reply to an old post by Masi but I wanted it to be 
> on top in your newsreaders ;)]
> 
> Martin Kutschker wrote:
> 
>  > I'm a bit disappointed that besides a few use "xhtml compliancy" quite
>  > often nobody bothers to point out any specific bugs. Especially as it 
>  > is said that some users have even patches.
> 
> Hi Masi,
> 
> we've just discussed the compliancy issue in the bugfixers list, but I 
> have not found any definitive decision on the issues of what should be 
> supported. Maybe we can re-animate this thread in order to get stuff 
> sorted out for 3.8
> 
> 1. What is the current policy for TYPO3 3.8+?
> According to the T3 CGL, we require XHTML (transitional) / CSS 
> compliance. Practically, at the moment, we semi-support HTML4 and XHTML 
> 1.x in TYPO3, resulting in things like xhtml_cleaning and various 
> conditional stuff which is either implemented already, or requested for 
> cases in which valid xhtml is invalid html4.
> Various people, including Wolfgang Klinger and me, have proposed to 
> concentrate on XHTML only, dropping HTML4 support. I find no sensible 
> objections to that, other than the typical legacy issues ("We started 
> with HTML4 in 1999 and want to continue that...")
> Supporting both valid HTML4 and XHTML still requires a lot of work I'd 
> rather spend somewhere else.


I don't buy this objection. HTML 4.01 is a current standard, and IMO the
onus is on whoever would like to drop support for HTML 4.01 to come up with
compelling reasons for not supporting all current versions of (X)HTML
instead of asking for others to come up with reasons for continuing to
support non-deprecated versions of HTML.

One of the very nicest things about T3 from a development point of view
right now is that it interferes relatively little with whatever coding
methods are chosen. Furthermore, when it comes to the differences between
the non-strict versions of xhtml and html, the differences are actually
fairly minor (e.g. both allow most presentational attributes). If what
we're talking about here is moving to xhtml 1.0 strict then, even though I
would love the web to suddenly be an xhtml strict place, I still have to
object.



> 2. Once we do settle on a policy, can we provide the users with the 
> appropriate, consistent, valid, bug-free content rendering? CSC has come 
> a long way, but there still seem to be quite a lot of issues with 
> "buggy" TS.
> As a result, I see the following todos:
> 
> 1. Making -core and sysexts completely XHTML 1.x compliant real soon now.
> 2. Removing the obsolete, non-XHTML TS from default installations (like 
> content(default), cSet, basically most of the static templates, as 
> listed in http://typo3.org/documentation/document-library/doc_statictmpl)



It's my opinion that this problem could be approached another way; instead
of supporting one changeable standard or another in the core, we should
probably abstract all Typo3's html generation to a separate class that can
be more easily upgraded/extended (though perhaps this might be an
unreasonably large task at this point?) Of course backwards compatiblity
would need to be maintained, but abstracting the markup generation would
allow different doctypes to be more easily supported.


> 3. Creating new example templates that are XHTML, use CSC for rendering 
> and show more of the new features, thus replacing the old glück, green, 
> business templates. Can we distribute those with the new import 
> extension, or should we create an example templates extension, like the 
> accessibility template?


Agreed - these are good ideas.


> I know this has been discussed a few times, but solving this issue would 
> spare us a lot of user requests ("I want to XHTMLize my 
> content(default)") and bug reports. Plus, a more rigid approach (like 
> the plone people have) would make it newbie-friendlier with valid and 
> modern out-of-the-box-templates.
> 


May be a good idea. As long as the rigid approach does not interfere with
flexibility :)

-Christopher



		
__________________________________ 
Do you Yahoo!? 
Make Yahoo! your home page 
http://www.yahoo.com/r/hs