XML 1.0 5th Ed and related PEs [was: Minutes for XML Core WG telcon of 2007 December 5]

 

> -----Original Message-----
> From: public-xml-core-wg-request@w3.org On Behalf Of François Yergeau
> Sent: Wednesday, 2007 December 12 20:11

> 
> > ACTION to Francois:  Generate XML 1.1 3rd Edition with approved 
> > errata (and warning about upcoming PE about LEIRIs).
> 
> Done, see 
> http://www.w3.org/XML/Group/2007/12/PER-xml11-20071212-review.html. 
> Please check the new note in 4.2.2.
> 
> 
> > ACTION to Francois:  Generate a PE against 1.1 about allowable
> > values of the XML decl's version attribute.
> 
> Done, this is PE162.  Please check if I did the right thing, I'm not 
> sure I remembered correctly.

I think it's right, but I suspect we may want to say something
more in prose, but whatever it is, it would match what we decide
to say in XML 1.0 5th Ed (more below).

> >> 8. XML 1.0 5th Edition
> >>
> >> ...
> > 
> > ACTION to Francois:  Generate XML 1.0 5th Edition with namechar
> > and XML version changes and approved errata (and warning about
> > upcoming PE about LEIRIs).
> 
> I created PE160 detailing the major changes we want: naming rules 
> and version number. 

I hate ask you to do more work, but I really think we should
separate the name char PE from the version number PE, so could
I ask you to break out the "Section 2.8 Prolog and Document Type 
Declaration" bit from PE160 and make that a separate PE please.

---

As far as the version number change, it seems to me that we need
more prose here about what 1.0 processors should do with, say,
version="1.1", but I'm not sure what it should be.

Richard, John, others, comments on this please.


> Note that I ended up deleting the whole 
> Appendix B and adding a new Appendix J "Suggestions for XML Names", 
> following the lead 
> of XML 1.1.  This Appendix is in fact exactly the same as XML 1.1 
> Appendix I, except that
> 
>      "...derived from [XML-1.0] Appendix B."
> becomes
>      "...derived from Appendix B in previous editions of this 
> specification."

I think this is fine--does anyone else have a problem with this?

> 
> Draft 5th Edition is at 
> http://www.w3.org/XML/Group/2007/12/PER-xml-20071212-review.html
> 
> 
> The status of this draft and that for 1.1 3rd Ed. are as follows:
> 
> - All published errata integrated.
> 
> - Outstanding PEs also integrated.  I will back out/modify as 
> required if these PEs are finally not resolved as currently
> proposed.
> 
> - I still need to validate everything and check pubrules.
> 
> - I took a first stab at the Status sections, taking inspiration
> from the previous PERs and RECs. 

When mentioning the xml-editor archives, add the word "public"
before the linked word "archives" (so that people realize any
comment they make will be public).

Change "An implementation report" to "A preliminary implementation 
report", since when we publish the PER, we may not have any
implementations (so the prelim IR will be just a stub), and we
almost certainly won't have all that we will have when we go
to Rec.

We will need to augment the test suite to test all the errata,
especially the name char and version ones, before we can go
to PER.  I'm hoping Richard, perhaps with help from Daniel,
will be able to supply the needed test cases.

> Noteworthy:
> 
>     - Last time around we offered implementation reports.  
> Perhaps our plans will be different this time, at least
> for 1.0 where we want to have CR-like exit criteria? 

If we have any implementations before we actually go to PER,
we can list them, otherwise, the preliminary IR will just be
a stub saying there are none yet.

Just before the IR paragraph, add (my suggestion):

 The XML Core WG wishes to ensure continued universal
 interoperability for XML 1.0.  To this end, the WG will 
 not request that this XML 1.0 Fifth Edition become a
 Recommendation until the following criteria are satisfied:

 1.  At least three months have passed since the publication
     of this PER.

 2.  There are at least three implementations that pass the
     test suite for each of the errata that have been newly
     applied to the 5th Edition.

Does anyone have comments on my suggestion above?

> What about 1.1?
> 
>     - I took the Patent Policy language from the 2006 Recs.  
> There is a difference between 1.0 and 1.1: the latter has
> this sentence "This document is governed by the 24 January 2002
> CPP as amended by the W3C Patent Policy Transition Procedure.",
> which 1.0 doesn't have. 
> Is that kosher?  PP experts, please!

Whatever is in the 2006 ones should be correct and should
be carried on into the latest versions.  One does not change
PP for a new edition of the same (version of the same) spec 
(I remember Ian telling us that before).  

XML 1.0 1st Ed came out before there was any official PP, 
so that's why it has (or doesn't have) what it does.

paul

Received on Thursday, 13 December 2007 15:31:24 UTC