Re: IRIs [was: [Minutes] 17 Mar 2003 TAG teleconf...]

On Mon, 2003-03-17 at 17:34, Paul Grosso wrote:
[...]
> 
> The XML Core WG has at least three specs going to PR within the month:
> XML 1.1, Namespaces 1.1, and XInclude 1.0.  On what we thought was
> direction from the TAG, we put IRIs into all these specs.
> 
> Now that it's been said that the TAG didn't tell us to put IRIs into
> these specs, and given that having IRIs in these specs is causing
> trouble,

What sort of trouble?

>  what would be the effect of the XML Core WG's pulling IRIs
> out of these three specs and just going back to URIs?  Would that
> force us to go back to Last Call?  

The only way to get that question answered for sure is to ask
for PR status and get a Director's Decision that the spec
hasn't had sufficiently wide review since the change
to pull IRIs; i.e. that it needs another last call.
(or, conversely, that it's OK to go to PR).

I don't know, offhand, what the impact of taking IRIs
out of XML 1.1 etc. is. I suspect you know more than
I do. I'd expect to hear from the I18N on whether
it's acceptable. Hmm... that starts to sound like
another last call... hmm...


> We have specs that, presumably, the W3C membership would like to see 
> become Recs, but it is unclear to me how to make that happen. 

It seems there are 2 general options:
  * wait for the specs you depend on to reach the relevant
	level of endorsement
  * accept the risk that, even though the IRI spec isn't
	endorsed/baked, it won't change that much after
	your spec goes out. Convince The Director and The Membership
	to likewise accept the risk in your request for PR.

IRIEverywhere seems, to me, to be a complex issue that I have
spent a long time studying without figuring out exactly what
I think is the right answer. I don't believe there are any
easy answers. I suspect the best/right thing to do is to try
some things and see what hurts.


-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Monday, 17 March 2003 18:58:34 UTC