W3C home > Mailing lists > Public > www-html-editor@w3.org > July to September 2006

XHTML Basic 1.1: inconsistence on "target" and "value" attributes (+ other final remarks)

From: Alexandre Alapetite <alexandre@alapetite.net>
Date: Wed, 2 Aug 2006 21:32:51 +0200
To: <www-html-editor@w3.org>
Message-ID: <000001c6b66a$71b58d30$f9043f50@athlon1100>

Dear HTML editors,
I am surprised of the choice of XHTML Basic 1.1 to introduce 3 new attributes
[http://www.w3.org/TR/2006/WD-xhtml-basic-20060705/#abstract]. In any case, if they are to be used, the recommendation should
imho contain more explanations.

Among other things, those new attributes will make that XHTML Basic 1.1 is not purely based on XHTML Modularization 1.1, is no
longer a subset of XHTML 1.1 or even XHTML 1.0 Transitional. This will make the situation more complex imho, especially when
given the purpose of XHTML Basic, which should be simple and a kind of common denominator.


1) The "value" attribute for <li> tags was deprecated in HTML 4
[http://www.w3.org/TR/html401/struct/lists.html#adef-value-LI] and only allowed in the transitional version. It was forbidden in
XHML 1.0 Strict, XHTML 1.1, XHTML Basic 1.0 and is not in XHTML modularization 1.1. The current explanations about the use of
this "value" attribute are in the HTML 4 specification where it is deprecated.

If XHTML Basic intends to re-introduce <li value="123">, I think the specification should at least explain how to use it today
and why it has been un-deprecated.

Please note that the current specification and schemas allow a code with unordered lists such as:

 <ul>
 <li value="2">example</li>
 </ul>

I think this is wrong and if so, it should be stated in the specification.


2) XHTML Basic 1.1 introduces the "target" attribute
[http://www.w3.org/TR/1999/REC-html401-19991224/types.html#h-6.16]
as one of the default modules
[http://www.w3.org/TR/2006/WD-xhtml-basic-20060705/#s_xhtmlmodules].

Historically, the "target" attribute was neither in HTML 3.2
[http://www.w3.org/TR/REC-html32] nor in HTML 2.0
[http://www.w3.org/MarkUp/html-spec/]. It has been tolerated in HTML 4 Transitional
[http://www.w3.org/TR/1999/REC-html401-19991224/sgml/loosedtd.html] and XHTML 1.0 Transitional
[http://www.w3.org/TR/2002/REC-xhtml1-20020801/]. "Transitional versions" included the following recommendation: "Authors should
use the Strict DTD when possible".

And the "target" attribute has never been allowed in any strict HTML version, such as HTML 4 Strict
[http://www.w3.org/TR/1999/REC-html401-19991224/sgml/dtd.html] and XHTML 1.0 Strict
[http://www.w3.org/TR/2002/REC-xhtml1-20020801/]. Later XHTML schemas based on XHTML modularisation have not allowed it either:
XHTML Basic 1.0
[http://www.w3.org/TR/2000/REC-xhtml-basic-20001219/#s_xhtmlmodules] and XHTML 1.1
[http://www.w3.org/TR/2001/REC-xhtml11-20010531/doctype.html#s_doctype]. So the "target" attribute has never been strictly
allowed. See also
[http://www.w3.org/MarkUp/2004/xhtml-faq#target].

However, now XHTML Basic 1.1 introduces the "target" attribute as one of the default modules.

There have been long debates on the problems associated to the "target" attribute as well as frames. The target="_blank" is in
particular known to be problematic, as reported by the Web Content Accessibility Guidelines 1.0
[http://www.w3.org/TR/WCAG10-HTML-TECHS/#no-new-windows]. More generally, frames are not recommended
[http://www.w3.org/TR/2006/CR-mobile-bp-20060627/#iddiv351037376]
[http://www.w3.org/TR/xframes/#s_intro]. It is true that "target" can be used with e.g. <object> and not only frames, but the
specification does not provide with any example or recommendation. See also discussion on
[http://lists.w3.org/Archives/Public/public-bpwg/2006Jul/0012.html].

Here again, if XHTML Basic 1.1 intends to use the "target" attribute, the recommendation should imho be a bit more talkative in
this regard.


3) The "inputmode" is from XForms
[http://www.w3.org/TR/2006/REC-xforms-20060314/sliceE.html]. I think it is strange to pick up only one new attribute from XForms
(it is not in e.g. XHTML 1.1, XHTML 1.0, HTML 4) and bring it into XHTML Basic, especially since this attribute is unknown to
XHTML modularization 1.1. In my opinion, XHTML Basic should only contain elements and attributes defined in XHTML
Modularization.


4) A final minor remark: Current major W3C's recommendations follow the RFC 2119 "Key words for use in RFCs to Indicate
Requirement Levels" [http://www.ietf.org/rfc/rfc2119.txt]. This is in order to be clear about the signification of e.g. "must",
"should" etc. and those special words are in this case emphasized.

Examples:
- XHTML Role Attribute Module [http://www.w3.org/TR/2006/WD-xhtml-role-20060725/#s_conformance]
- XHTML Modularization 1.1 [http://www.w3.org/TR/2006/WD-xhtml-modularization-20060705/conformance.html#s_conform]
- XML 1.1 [http://www.w3.org/TR/2006/PER-xml11-20060614/#sec-terminology]
- XHTML 1.1 [http://www.w3.org/TR/2001/REC-xhtml11-20010531/conformance.html#s_conform]

But XHTML Basic 1.1 recommendation currently does not follow this model, as I think it "should".


Cordially,
Alexandre
http://alexandre.alapetite.net
Received on Wednesday, 2 August 2006 19:33:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:17:55 GMT