Re: stopping discussions on serializations and contributing constructively

On Jul 13, 2007, at 1:28 PM, Smylers wrote:

>
> Robert Burns writes:
>
>> On Jul 11, 2007, at 5:56 PM, Smylers wrote:
>>
>>> Dan Connolly writes:
>>>
>>>> I encouraged some discussion in this thread as input to the editor
>>>> of "HTML5 differences from HTML4", but I think it's time to bring
>>>> this thread to a close.  The editor has considered these
>>>> suggestions;  we're starting to repeat ourselves and get into
>>>> tangents.
>>>
>>> Obviously that ends the discussion on the HTML source of that
>>> particular document.  Do you also want us to stop the discussion on
>>> the wider issues that have come out of this thread ("tangents", if
>>> you like)?
>>
>> Smylers, rather than throwing out straw man arguments, it would be
>> better to find actual posts that reflect the sentiment you're arguing
>> against.
>
> In that mail I wasn't arguing against anything, but here are some  
> quotes
> from the mails I was alluding to:
>
> * Robert Burns said: "It might be worth considering making the non-XML
>   serialization of HTML5 a very XML-like serialization of HTML5.
>   Obviously, the UA conformance would include requirements for  
> inference
>   of opening and closing tags, but our recommendation for authoring
>   could require authors to always include closing tags"
>
> * Mynthon Gmail said: "My idea is to have compatible syntax, but xhtml
>   is xhtml with its own parse and html is html with its own parser.  
> Only
>   syntax is unified."
>
> * To which Robert Burns then agreed: "its hard for me to think of
>   downsides to just requiring of authors a very XML-like syntax for
>   HTML5's non-SGML / non-XML serialization."

Perhaps I was overstating the case there. I didn't mean required in  
the sense that its used in RFC 2119. I'll try to be more careful  
about using that word. However, there are again several issues here  
that relate to our toolbox of conformance options I've written about 
[1].  We can tell authors to: 1) always quote attributes, 2) to  
always include closing tags except for canonically empty elements;   
and 3) even include a solidus in canonically empty elements (e.g.,  
<img />). Or we could do none of that. However, even if our author  
conformance recommends or requires one or all of those, there's  
nothing to say the conformance checker needs to throw up errors for  
those things. Perhaps the conformance checker should parallel our  
author conformance criteria  but it could do so by distinguishing  
between errors, warnings and comments.

>>> For example, there've been proposals that the non-XHTML should
>>> require closing tags, or quotes round attributes,which would be
>>> significant changes from the current spec.  Is it still OK to
>>> discuss these proposals?
>>
>> I don't think anyone expressed that.
>
> The above quotes appear to do that.

I hope I've clarified that now.

>>> And there have been claims that XHTML is preferred over HTML, which
>>> is not reflected by the current spec, which says:
>>>
>>>  Generally speaking, authors are discouraged from trying to use XML
>>>  on the Web, because XML has much stricter syntax rules than the
>>>  "HTML5" variant described above, and is relatively newer and
>>>  therefore less mature.
>>
>> This is not from the current "spec". This is from the current
>> "draft": which is what we're all here to work on.
>
> Well yes, obviously.  I presume that everybody on this mail list knows
> that the spec is currently in draft state.  That's what I meant by
> "current spec": the spec as per its current draft.
>
>> You can't cite the current "draft" as evidence that proposals would
>> conflict with current standards.
>
> Actually, quite the opposite: I mentioned it to show that should the
> group decide to embrace the opinions I referred to then the spec would
> need changing to reflect them.  Hence why I asked whether it's  
> still OK
> to discuss such changes.

I don't know how to reconcile your previous two statements. The only  
thing we might discuss on this list that wouldn't involve changing  
the draft would be statements like: "Great draft". Or "I whole- 
heartedly endorse section 2". While I don't want to discourage anyone  
from making such statements here, the meat of our work has got to  
involve critical reviews of the draft and recommendations for better  
wording or new wording or added wording or deleted wording. Otherwise  
why else are we spending our time here.

>> That's entirely counter-productive.
>>
>> Again, however, no one has expressed anything about guiding authors
>> to to prefer XML over text/html.
>
> This thread started with people suggesting that the document  
> produced by
> this working group should be have the closing tags and quote marks
> required by XML, for example Sander Tekelenburg:
>
>   The current mix of sometimes using optional closing tags and  
> sometimes
>   not is probably a bad signal to authors -- at the very least it  
> should
>   be consistent, but preferably it would use explicit closing tags  
> even
>   when they're not required. ... W3C setting the example by always
>   explicitly closing everything is probably a good idea.
>
> If this working group agrees on that "good idea" then we will need to
> update the spec accordingly.

On that issue that would mean update the spec only in terms of its  
source code. But that's an entirely separate issue from whether our  
recommendations guide authors that they must use xml-like syntax or  
(and this is what I was speaking to here) require authors to only use  
xml serialization and delivery. In other words I was saying: "no one  
has expressed anything about guiding authors  to to prefer [xml  
serialization and delivery, i.e.,]  XML over text/html."

>> One way to reduce the traffic on the list would be to stop these  
>> straw
>> man and knee-jerk arguments entirely. When posting a reply consider
>> whether you're post will shape the conversation in a positive way. If
>> you're posting to just register your opposition to statements without
>> constructively shaping the discussion, just choose not to comment.
>
> Thanks for the advice.  In this case I wasn't opposing anything,  
> merely
> seeking advice from the chairman on whether the issues I mention are
> on-topic.

I"m sorry I misunderstood. I thought your post was trying to guide  
the chair to rule these discussions out of order, rather than respond  
to the substance of the discussion in a way that contributes to our  
understanding one another and improving the draft. If that was not  
your intention, than my advice does not apply to you.

Take care,
Rob

[1]: <http://lists.w3.org/Archives/Public/public-html/2007Jul/0697.html>

Received on Saturday, 14 July 2007 20:58:45 UTC