Re: Agreement on IRI "processing spec" moving to W3C

Hello Addison, others,

On 2011/11/21 4:04, Phillips, Addison wrote:
>>
>> On 2011/11/20 15:11, Chris Weber wrote:
>>> During IETF 82 an announcement was made that the IRI "processing spec"
>>> would move to the W3C for creation as a self-contained document. See
>>> <http://trac.tools.ietf.org/wg/iri/>  for the minutes.
>>>
>>> Are IRI WG members in agreement on this decision?
>>
>> Some procedural questions:
>>
>> What WG will handle this? Will there be a mailing list with reasonably limited
>> scope? (rather than a mailing list where a vast array of unrelated issues is
>> discussed on a daily basis?) Could we maybe even use this mailing list for this
>> work, because it's hosted by W3C and the relevant people should already be
>> here? Or would that be too complicated, because it would have to run under
>> two sets of rules (IETF and W3C)?
>
> The I18N-WG is chartered to produce Working Group Note documents and is willing to host the document, as long as "Working Group Note" status is an acceptable document status. It may be that the HTML-WG would prefer to host it.

In my opinion, if possible, a more normative status (i.e. W3C 
Recommendation) would be desirable.

What I didn't ask yet, but also should be considered:
In the IETF IRI WG, we didn't get much work done on processing because of:
- Lack of critical mass
- People with time to devote
- Disagreements on the approach to take

What is the expectation about these issues when moving to the W3C? 
(Maybe the I18N WG has lots of people with free time, or if the HTML WG 
is in charge, there won't be any disagreements on approach, or whatever.)

Somewhat related, I'm not sure why this got announced as "the IRI 
"processing spec" will move to the W3C" (like a fait accompli), but then 
it's not clear what exactly that means, and we are asked to say whether 
we agree or not. As an example, what will happen if there is strong 
disagreement here?

I personally don't care too much where this is done, or whether it takes 
the W3C a week or two to figure out how they would do it. I also think 
that the IETF IRI WG doesn't have a very successful track record for the 
IRI processing spec, so it may not be a bad idea to try and do it 
somewhere else, even if just to give it a try. But ideally, it would be 
something more than "just give it a try".

>> Also, how will we make sure that the work on the main spec (3987bis) and
>> this processing spec work is going to be coordinated? There are some points
>> where such coordination should happen, e.g. around encoding issues in
>> query parts.
>>
>
> (individual comments) For definition of "what is an IRI", the processing spec should obviously be subordinate to the main IRI spec.

Fully agreed.

>  The processing spec, of course, has to deal with error conditions and 
out-of-bounds items.

Yes. What I was hinting at with query parts is that between "what is an 
IRI" and "error conditions and out-of-bounds", there are issues that 
need closer coordination.


Regards,    Martin.

Received on Monday, 21 November 2011 03:54:38 UTC