W3C home > Mailing lists > Public > public-evangelist@w3.org > November 2004

XHTML Best Practices - a project

From: Karl Dubost <karl@w3.org>
Date: Tue, 2 Nov 2004 14:46:12 +0100
Message-Id: <8F2BF636-2CD5-11D9-8D7E-000A95718F82@w3.org>
To: 'public-evangelist@w3.org' <public-evangelist@w3.org>
Hi,

The HTML/XHTML specification defines semantics for the elements but the 
given definitions or the companion examples are sometimes not very 
explicit or might lead to confusion. So the idea of trying to set a 
kind of Best Practices has sprung on public-evangelist mailing-list 
recently.

HTML 4.01 [1] defines the semantics of HTML 4.01, XHTML 1.0 [2] and 
XHTML 1.1 [3] elements (if not redefined locally)
XHTML 2.0 [4] WD defines its own semantics. XHTML 2.0 has not yet 
reached last call, so it has to be considered as an unstable 
technology, understand not yet finished and might change.

[1] http://www.w3.org/TR/1999/REC-html401-19991224
[2] http://www.w3.org/TR/2002/REC-xhtml1-20020801
[3] http://www.w3.org/TR/2001/REC-xhtml11-20010531
[4] http://www.w3.org/TR/2004/WD-xhtml2-20040722

* STRATEGY *

1) What the members of the list would prefer develop scenarios for HTML 
4.01 semantics family or for XHTML 2.0?
	1.a) HTML 4.01: It means we can start now and with a good 
organization, develop scenarios.
	1.b) XHTML 2.0: It means that we have to wait at least for Last Call, 
and then develop scenarios.

I see advantages and drawbacks for both cases. Maybe, we could use the 
HTML 4.01/XHTML 1.0 cases to refine what could be proposed for XHTML 
2.0 and indeed ask for clarification to the HTML WG when a definition 
or an example seems obscure.

* POSSIBLE WORK ORGANIZATION *

This is just a proposal. The first one is based on a vocabulary basis.

* Work 1
Element name: [name of an element]
Source: [URI]
Definition: [extracted from the specification]
Possible Meaning: [to define what is understood from the spec, 
including the false assumptions.]
Possible Usage Scenarios: [Real use case with examples; possible ideas 
for implementations in different contexts.]

The second one is slightly different not based on a vocabulary basis, 
but more on a scenario constraint.
	How do I make a quotation?
	How do I create a table of content?
	How do I insert code in a page?

* Work 2
Possible Usage Scenarios: [Real use case with examples]
	-> What does that imply for the elements and attributes?
	-> What are the possible implementations?

I think a kind of combination of both might be useful. The first one is 
systematic and easier to follow, the second one brings richer 
propositions BUT also more arguments, let's say debate.

Things you have to remember in this kind of exercise:
	- we are here to share ideas
	- we are working in an international context with multi-cultural 
needs. Don't forget to listen other cultures (basically don't do a 
north-american only best practices guide).
	- No arguments, no nicknames, no trolls but proposals, explanations, 
etc. Do not assume the others know, just explain.
	- be patient.

* HOW TO MOVE ON.

One person take one element (on the wiki? on the list?) fill the 
template, one week later proposes to the list the possibilities 
associated with the template with this subject if you deal with the 
code element.
	 [XHTML Best Pratices] element code
It's debated here on the list for two weeks. The subject is closed and 
filled in a HTML page. The wiki can be a support for it too.

* BACKGROUND MATERIALS *

SimpleQuiz: A journey through HTML semantics
	http://www.simplebits.com/notebook/simplequiz/index.html
	Which contains mistakes like
	http://www.simplebits.com/notebook/2004/08/04/sq.html
	“address” element is not made for postal address in HTML.

A touch of Class, Tantek Çelick
	http://tantek.com/log/2002/12.html#L20021216t2238






-- 
Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager
*** Be Strict To Be Cool ***

Received on Tuesday, 2 November 2004 13:46:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 15 July 2011 00:13:22 GMT