Appendix A of C14N 1.1 [was: Minutes for XML Core WG telcon of 2007 October 10]

 

> -----Original Message-----
> From: public-xml-core-wg-request@w3.org 
> [mailto:public-xml-core-wg-request@w3.org] On Behalf Of Grosso, Paul
> Sent: Wednesday, 2007 October 10 11:16
> To: public-xml-core-wg@w3.org

> > 3.  C14N 
> > 
> > The C14N 1.1 Candidate Recommendation is published at
> > http://www.w3.org/TR/2007/CR-xml-c14n11-20070621

> Interoperability testing was performed on 27 September.  
> A report of the outcome is at:
>
http://lists.w3.org/Archives/Public/www-xml-canonicalization-comments/20
07Oct/0000

> Point 3 complains about appendix A being too difficult to 
> understand, partly because it uses 3986 language which is
> also hard to understand.
> 
> We have three choices:
> 
> 1.  say hard to understand isn't wrong, so leave it as is.
> 2.  delete appendix A altogether, possibly adding to the
>     main text if there is anything "normative" only mentioned
>     in the existing appendix.
> 3.  rewrite appendix A.
> 
> ACTION to Konrad:  Send us pointers to suggestions for rewriting
> appendix A.
> 
> One such pointer is 
> http://lists.w3.org/Archives/Member/member-xmlsec-maintwg/2007Sep/0017

I am really not excited about working further on Appendix A.
This business of trying to describe xml:base fixup has already
been a big time sink, and every time we make some changes, more
need to be made later.  I worry we may never come to closure.

I am not at all happy about trying to use the wording in
http://lists.w3.org/Archives/Member/member-xmlsec-maintwg/2007Sep/0017
I have no reason to believe that wording is accurate or that it 
will be easier to come to agreement with that as the starting point.

Starting with 3986 seems the safest thing to do, and appendix A
represents our best attempt at that so far.

My preferred solution is to say that, if there are no errors pointed
out in Appendix A, then we leave it as is.

My next preference is to attempt to describe the changes from the
3986 algorithm in prose in section 2.4--because currently it seems
clear that we have not described xml:base fixup fully in section 2.4,
so we cannot just delete appendix A--and then delete appendix A.

I'm happy to entertain other ideas if they are presented in email
before our next telcon AND they make forward progress in closing
this issue. 

paul

Received on Monday, 15 October 2007 21:03:41 UTC