Re: supporting both formats html5 & xhtml5 re: http://www.w3.org/html/wg/html5/#xhtml5

Shawn Medero wrote:
> On 12/19/07, Dean Edridge <dean@55.co.nz> wrote:
>
>   
>> If there are any genuine reasons why both serialisations can't be
>> supported by all user-agents I'd like to hear them.
>>     
>
> This is all I am asking... when putting together an issue in the HTML
> WG Tracker, I need to collect data that presents as much info about
> the pros & cons.
>   

Perhaps be more specific next time then. You never said that you were 
the "WG tracker guy". I'm so used to my comments being knocked down by 
straw-man arguments, didn't realise you were trying to mediate sorry.

> I'm not in a position to make proposals, my experience with HTML &
> XHTML involves authoring, parsing, and info retrieval against large
> data sets... I've not written any aspect of a general user-agent
> implementation to date and have no plans to. On the other hand, I'm
> familiar enough with the concepts that I've volunteered to collect,
> refine, and track issues for the working group.
>
> With that in mind, I can think of two use cases that are sorta linked:
>   

> 1. Mac OS X Dashboard Widgets
>   

Well surely this should be discussed as a group instead of just one 
person (Ian) deciding that XHTML support be optional and writing that 
into the spec. I would have thought that by default all features in the 
spec should be supported by all user-agents. I don't think I'm being 
unreasonable here; XHTML has been a W3C recommendation since 1999, I 
don't think it's too much to ask that user-agents support it. Three out 
of the "big four" UAs do so already (Firefox, Opera, Safari/Webkit).

> 2. HTML editing software
>   

"HTML editing software" is not a user-agent. And anyway, if HTML editing 
software did not support XHTML; XHTML could still be used on the web so 
this is not a major problem.

> Apple wanted HTML desktop widgets but didn't want to get into of the
> business of XML. If I had to guess, they did this because:
>
> 1. Authors generally know how to write HTML but not XML
> 2. HTML editing software that has to understand but HTML and XML is
> more complex to write, test, and support. To suggest that support both
> HTML and XHTML is no big deal underplays the problems with things like
> serialization and namespaces.
>
> These seem like valid concerns with the benefit that Dashboard is more
> stable for OS X users and Dashboard widgets are easier to write for
> Apple's many 3rd party developers. I'll leave the exercise of vetting
> these details to people much smarter than myself.
>
> <rant>
>
> A general meta-reflection about raising issues on the mailing list:
>
> I really wish people would somewhat extract themselves from the issues
> (It is impossible to be entirely emotionally detached from them, we
> are human after all.)... 

Actually Shawn, you've got a nerve saying that. I'm insulted by your 
suggestion that my comments are influenced by my emotions/issues. Why 
don't you go to the source of the problem first Shawn? Can I suggest the 
next time you want to be the "list mom" (your words not mine) that you 
take a look into the source of the problems with this group. Not the 
person (me) that has the unfortunate job of bringing the groups 
attention to the problems. If the spec was written in a democratic and 
open way I wouldn't need to complain about it's content and if *other* 
people would "extract themselves from the issues" we wouldn't have an 
unbalanced and "sold out" spec. The reason that I need to post emails to 
this list is because the spec is unbalanced and heavily influenced by 
people such as Ian Hickson, and his group. These people are too 
"emotionally attached" to classic HTML and the way the web *was* 
yesterday to have the vision to see what the web *should be* tomorrow. 
These are the people that can't "extract themselves from the issues", 
not me. These are the people that have the "It's our spec, what do you 
mean you don't like it." attitude. It's not my "emotions" or "issues" 
that are stopping the web from "reaching it's full potential" or the 
HTML WG from being productive, it's theirs.

If Ian Hickson acted in a more professional, independent and objective 
manner without his "It's my way or the highway" attitude and stopped 
putting his own personal wants/feelings in front of what's best for the 
web; I wouldn't need to be here spending my time taking out all the "I 
hate XHTML" and other errors from the spec.

> when someone asks for more info ... it is not
> a personal attack against you or your beliefs. The majority of
> specification change request come with too little information to be
> actionable.

That's where you are missing the point Shawn. It shouldn't be a 
"specification change". By default; both serialisations should be 
supported by all (X)HTML5 user-agents. Does the CSS2.1 spec say that 
user-agents only need to support half of the CSS specification because 
it's inconvenient for Apple inc, to support all of it? I doubt it.
If it is inconvenient or impossible for a particular UA or 
Implementation to support all of the spec; then it should be discussed 
openly within the group, not just automatically left out of the spec.

>  (I don't expect this to ... nor is it clear that it
> should... change.) So the discussion has to start on the mailing list,
> irc, etc to fill in all the gaps.
>
> I think it is great that you are passionate about your areas of
> interest Dean, without that spirit this (and other) specification
> would go nowhere. 


> I'd just like to see cooler heads prevail in most
> cases.

You mean you want people to overlook the hijacking of the web by Google, 
Apple and Opera so we can all have a more peaceful mailing list? Have 
everyone just go along with the dictatorship led by Ian Hickson and his 
sponsors? Because this is really what it's all about isn't it? If 
everyone just goes along with Google's, Opera's and Apple's spec and 
don't ask for any changes everything will be just fine.
Well, I don't think so. I'm not selling out.

>  (In fact, I may be overreacting as response to the unnecessary
> blow up over the <video> container & codec issues...)
>   

Yes you are in fact overreacting Shawn. If you don't want me to email in 
every week attempting to correct the errors in the spec I suggest that 
you talk to Ian and ask him to do his job and correct the errors in the 
spec himself. Otherwise we might as well do everything Ian's way; remove 
other peoples ideas, and call the spec HML5 (Hixie's Markup Language 
version 5).


[1] http://lists.w3.org/Archives/Public/www-style/2007Dec/0094.html

Dean Edridge

Received on Monday, 24 December 2007 16:16:06 UTC