W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > May 2001

Re: Action Item: proposal for posting resolved issues

From: Art Barstow <barstow@w3.org>
Date: Sat, 26 May 2001 11:17:29 -0400
To: Aaron Swartz <aswartz@upclink.com>
Cc: w3c-rdfcore-wg@w3.org
Message-ID: <20010526111729.A22925@w3.org>
On Fri, May 25, 2001 at 05:11:43PM -0500, Aaron Swartz wrote:
> Art Barstow <barstow@w3.org> wrote:
> 
> > o Link back to the original issue [the original issue could be
> > left intact]
> 
> Can the original issue link to the resolution also?

That would be fine with me.

> > o Link to discussion in the mail archive
> 
> This would be discussion in our mail archive, correct?

The w3c-rdf-core archive.

> > o Test case(s).  URI's would be nice but inline tests would
> > be better than nothing.  [Perhaps a separate agenda item
> > for the WG is test case creation and access.]
> 
> Yes, please put test cases at URIs! Perhaps in your rdf-tests directory
> where SiRPAC and other test currently live?

I'd take any URI. I would also be willing to create new
directory(s?) under rdf-tests and put tests there.  But,
for example, if DaveB wants to point to a directory at
an ILRT site, I think that would be OK.

> > At the top of the page in blinking lights and 50 point fonts
> > there must be a warning about clearly states that the information
> > in the document is not final and subject to change.
> 
> Heh. How would this page fit in with the errata document? And why would our
> resolutions be subject to change? If we resolve something, doesn't that mean
> we're done with it?

I don't think things like test cases and discussion links
belong in an errata.  But then again, until yesterday's phone
conference I thought errata content was fairly limited (e.g. 
fixing obvious typos) and would not contain things like
grammar changes.

However, the group decided the following change for rule [6.19] 
would be added to the errata:

 [6.19] Qname          ::= [ NSprefix ':' ] name

will be changed to:

 [6.19] Qname          ::= name

I think it would be helpful if the chairs would say something about 
what can and cannot be placed in the errata.

But even with this broad defintion of permissible errata changes, it seems
like some potential WG decision will be less clear - that is, not typo
fixes but more like interpretation changes.  As an example (and
please forgive me if I got this wrong), I believe as a resolution to:

 http://www.w3.org/2000/03/rdf-tracking/#rdfms-not-id-and-resource-attr

you essentially suggested in:

 http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0156.html

that the last clause of [6.12] be changed from:

 | '<' propName idRefAttr? bagIdAttr? propAttr* '/>'

to:

 | '<' propName idAttr? resourceAttr? bagIdAttr? propAttr* '/>'

This wouldn't be a typo fix, IMHO, as it would clearly break
all RDF parsers that conform to the 1.0 spec.  

I would not expect these type of changes to be placed in the
errata.  As a convenience to the WG (and the RDF community), I
think it would be useful to consolidate these type of changes
in a document as I have described.  All changes in that document
would be subject to some type of "it's not over till it's over"
provisio as I stated previously.

Art
---
Received on Saturday, 26 May 2001 11:17:30 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:35:52 EDT