W3C home > Mailing lists > Public > public-evangelist@w3.org > May 2005

Re: Valid XML

From: Karl Dubost <karl@w3.org>
Date: Fri, 20 May 2005 16:24:54 -0400
Message-Id: <3D0EB0FD-C878-49C7-A27A-B19EEA3C8EAF@w3.org>
To: "'public-evangelist@w3.org' w3. org" <public-evangelist@w3.org>



Small story:
Jim, you are waking up in the morning, you are maybe taking a coffee.  
Your cup is falling over you and you have coffee all over you. Should  
I kill Jim Ley, Should I suppress the cup? Should I not try to drink  
coffee in the morning because I'm clumsy. :)

There are certainly plenty of solutions, there is no perfect world.

The small story is just to express. There will always be problems and  
I will not argue against that. Take it granted that. There will  
always be

     - incompatibilities
     - conflict between specifications
     - conflict of interests
     - bad implementations
     - good and bad professionals

The work which has been done for QA Specification Guidelines [1] is  
one part of this  attempt to improve things. Björn has, for example  
by his participation, showed very interesting problems (normative  
references). He has sometimes maybe too strict opinions, but I really  
appreciate his abilities for identifying nasty issues.

[1] http://www.w3.org/TR/2005/WD-qaframe-spec-20050428/

So Jim and others, if your point is to say: W3C is not perfect and  
there are issues in specifications. I will reply yes to both and I  
will add it will always be the case.

This understood let's move forward.

Le 05-05-20 ā 14:16, Jim Ley a écrit :
> Bjoern was just asking for similar help with XHTML guidelines.   
> Arguments seem rather thin on the ground other than "well it works  
> in both browsers so all is okay" which is something I can't agree  
> with.

:) It's where we have a strong disagreement. I'm not living in a  
mathematical universe with everything predictable. So yes I wish for  
more quality, for better implementations, but if you think, we will  
achieve perfection, I can answer you right now, that it will never  

> For example I'm embarrassed that I can't show the W3's homepage on  
> my pocket device, and I'm not embarrassed for it, it doesn't claim  
> to render XHTML, the W3's own XHTML Appendix C guidelines state it  
> won't be shown correctly.

I'm not embarrassed at all if it's not displayed correctly on some  
devices. I could be embarrassed if it was a software I was writing  
myself and was claiming it to be perfect.
It would be like Greenpeace being embarrassed because someone is  
doing the wrong thing. We can try to explain, do do outreach, we can  
pursue the desires of helping people, but we will not save the world  
or stop the war :)

>> You have the right to send "XHTML 1.0" as text/html.
> You only have the right if you follow the Appendix C.  Guidelines,  
> they are _not_ being followed by the W3 themselves,  so are you  
> saying the W3 homepage is wrong here?  Or are you saying something  
> else?

:) Hehe. What you are looking for is a statement to pinpoint and say  
"Look! look! be ashamed".  I can answer if you really want:
     - W3C Home Page is wrong
     - Appendix C. is wrong
     - I am wrong
     - etc :)))

*grin* I'm not sure it will achieve anything in the end or we will  
not make any progress. It's not a question of black and white, or 0  
and 1. At least, I do cherish when we are able to validate a  
document, to check that it has been used with conformance, etc. I  
cherish even more when we had successfully convinced someone to  
improve, fix an implementation, when we have found the good strategy  
for more interoperability.
But again a perfect world will never exist. If it's what you are  
looking for, I will not be able to answer.

> I'm not stressing on any problem, I was asking a simple question  
> about WCAG 1.0, and why it's appropriate to ignore the  
> specificaition?  I don't think it's appropriate to.

*I* (Karl Dubost) would say that it's a dependency bug with normative  
references. An issue that Bjoern has raised for many specifications  
and I'm very grateful, he pushed the QA WG to include that in the QA  
Specification Guidelines.

Good Practice 08: When imposing requirements by normative references,  
address conformance dependencies.

I don't know if an issue has been sent to the WCAG WG, and if they  
think that it should be included in the erratas.

Have you asked them? What did they answer?

>> http://www.w3.org/TR/2004/WD-WCAG20-20041119/
>>     I don't see such requirements in WCAG 2.0 WD.
> That is a working draft, and has not been reviewed by a great many  
> people, I think due to the poor responsiveness of W3 working groups  
> whereby issues are only ever addressed at last call, many people,  
> including myself, don't waste our time reviewing working drafts, I  
> realise this isn't a good thing, but many year old issues against  
> specifications just don't encourage it.

Hehe :) Imagine what I do every days ;)
The reason why I was pointing the new document is because it seems  
they fixed that, so they might be aware of such issue.

About Issues not fixed:

     This is trickier, because it involves people, social contracts,  
process, etc. All these small pieces of dust which stick everywhere  
along the life.
     - Karl Dubost might have an opinion different from Jim Ley.  
Björn might have a third opinion incompatible with the two first,  
etc. At a point we will need to find a solution, and there might be  
still disagreement. It's the result of collective discussion as we  
exactly do now.
     - There are also questions of time, resources, economical  
reasons, etc. Many things which are making our life difficult every  
day at work or in our personal life. W3C is not an absolute entity,  
it's an organization with people and resources constraints.

     For example, now, I have many things to do, and you certainly  
have, but we took the time to answer more than doing other things  
like reviewing another specification for me to try to identify  
errors, but I'm just human and I might miss some ;)

>> What is an "XHTML User Agent"?
> A user agent which claims to render XHTML - most XHTML user agents  
> will also render a great many other content-types.

:)  More precisely.
An "XHTML 1.0 User Agent" is

And an "XHTML 1.1 User Agent" is

I'm not sure someone can claim to have a conformant XHTML 1.0 or  
XHTML 1.1 user agent. But that would be interesting to show if time.

>> a) XHTML 1.1 application/xhtml+xml
>> b) XHTML 1.0 application/xhtml+xml
>> c) XHTML 1.0 text/html
>> Do you mean you don't know any user agents which does a)  
>> correctly  and not b) correctly?
>> So you are saying some user agents are unable to do a) and b)
> No, I'm saying that all that do b), also do a) - so XHTML 1.1 is as  
> supported as XHTML 1.0

for application/xhtml+xml, yes :)
Well I have plenty of other issues on HTML 4.01 and then XHTML 1.0 ;)  
but on this particular mime type aspect yes.

>> Do you know a user agent which is unable to do c) ?
> Yes, Internet Explorer 6.0, I gave the file (with corrections) in  
> the previous example.

So you are saying that IE 6.0 doesn't display (render) XHTML 1.0  
served as text/html ?

Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager
*** Be Strict To Be Cool ***
Received on Friday, 20 May 2005 20:24:59 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:16:19 UTC