xml-dev - RE: Realistic proposals to the W3C?

[Didier PH Martin <martind@netfolder.com>]

Hi Cavre,

Thread 1:
(3)
Didier said:
You probably mean here conformance test suites. To obtain the stamp of
approval the W3C members would have to run the test suite in order to check
their implementation conformance to the recommendation. If successful, the
vendor may keep its W3C membership and put the W3C logo on their product.
Did I got you right? If yes, this is an other mechanism that could help
re-enforce conformance and provide a help protect the users.

Cavre replied:
Yes indeed but why limit it to just W3C Members.  In truth any developer
ought
to have the right to submit any application for review and maybe earn the
W3C
logo.  Not just members alone.  However maybe members could get a free
review,
where as non-members would have to pay a evaluation fee of some sort.

Didier replies:
So, in that case, W3C has to provide a low fee access to the consortium to
open the doors. I personally think that the probability that this will
happen is very low since the requirement to reduce the fees to open up the
consortium occurred several times in the past and from this community...
nothing changed. Maybe we should conclude that it is not in the W3C intents
to democratize further the membership by reducing the fees (like OASIS did).

Thread 2:
Didier said:
I guess this is already the case but it does not re-enforce any good
behavior form the W3C members. (2) and (3) does.

Cavre replied:
I agree but if you read through number 4 of my suggestions I think you
will find I am only refering to "REC"'s.  Developers are humans and as
such each developer will have his's or her's own ideas and views about
any particular "REC" or "standard".  Their application may not need all
the bulk a standard may require to be included just to meet "standards"

Didier replies:
But this is precisely what some vendors are doing and then, by doing so,
jeopardize inter-operability by supporting a subset. Just remember the
browser's case with <div> and <layer> both bowsers where (and are)
supporting subset of the recommendation but a document may work on one but
not on the other. So, my take is that we should ask for full support of a
recommendation (or modules as it is the case for XHTML). This way, we have
some assurance that a document will be properly interpreted on the
"compliant browsers". However, the vendor may add some extras, but the users
is fully aware of that the document will be dependent on this vendor. This
may be useful, the importnat point to underline here is that at least the
user has the choice. Just consider the <div> and <layer> case, to keep some
inter-operability the Microsoft's browser had to include the <layer>
element. Thus, the only inter-operable element is the proprietary one :-)).
Conclusion: we have to be careful no to be contented only by "partial
conformance". So I fully agree with your last statement:

    "I still believe that if your going to support a "standard" you support
this
"standard" in full.  No exceptions allowed.  Any developer has freedom
of expression by supporting the "REC".  It's that freedom of expression
that I wish to keep as open as possible as much as possible. But a
"standard"
is a "standard" and we all need to follow it.  Please consider HTML Strict
for a good example."

Cheers
Didier PH Martin
----------------------------------------------
Email: martind@netfolder.com
Conferences: xml devcon 2000 (http://www.xmldevcon2000.com)
		 Wireless Summit NY (http:www.pulver.com)
	       XML 2000 DC (http://www.gca.org)
	       xml devcon 2001 London (http://www.xmldevcon2000.com)
Book: XML Professional (http://www.wrox.com)
column: Style Matters (http://www.xml.com)
Products: http://www.netfolder.comProducts: http://www.netfolder.com