See also: IRC log
<pratul> agenda at http://lists.w3.org/Archives/Public/public-sml/2008Apr/0142.html
Pratul: any objections to
approving the minutes?
... no objections heard. Minutes approved.
Pratul: In their 4/21 call, the XML CG approved our F2F meeting in Redmond from 10/28 to 10/30.
msm: The section title 'mapping from schema' is actually correct because 'schema' does not have to mean only schema document.
<MSM> I propose that I add, as an individual, the following comment to the bug, and then that the WG 'endorse' it (if we choose to).
<MSM> Strictly speaking, I think the premise that elements and attributes
<MSM> always originate in schema documents is at fault here. The section on
<MSM> mapping from schema is talking about the mapping from an annotation
<MSM> component (NOT an annotation source declaration), the contents of
<MSM> which are element and attribute information items.
<MSM> In the usual case, of course, the elements and attributes will have
<MSM> originated in a schema document, but we wish also to cover the case
<MSM> that the annotation component was born binary. Even in the latter
<MSM> case, however, it will contain element and attribute information
<MSM> items.
Pratul: does anyone object to
endorsing this comment?
... no objections heard. I will go ahead and endorse this
comment on behalf of the WG.
msm: explains the procedures
involving addressing reviewer's comments.
... I believe we should add bare-name support to SML URI
scheme.
Kumar: Henry's comment does not explain why he thinks our response is FUD. Our response does talk about bare-names which his example shows.
Pratul: explains the key aspects of the WG resolution during f2f.
<scribe> ACTION: Kumar to look at minutes of f2f and summarize the discussion on this issue. [recorded in http://www.w3.org/2008/04/24-sml-minutes.html#action01]
<trackbot-ng> Created ACTION-181 - Look at minutes of f2f and summarize the discussion on this issue. [on Kumar Pandit - due 2008-05-01].
Kumar: Henry's use of the word
'target' is different from what the SML spec defines. The spec
defines target to be an element. In order to find if the target
element is present, one must descend in to a document.
... Msm's response to 5562 addresses this. One could define a
ref scheme that acts as 'XML lens' (converts non-XML target
into some XML representation). Such a scheme would allow
targetRequired to apply to non-xml documents.
msm: We shouldn't use 'XML lens' idea as an argument here. The XML lens requires a lot of work (both implementation wise and conceptual) therefore may not be of interest to a lot of people. We should instead state that this is a trade-off between complexity and functionality.
Pratul: The WG discussed this and decided to stick to the original resolution. We should add a comment to the bug with a summary of our discussion.
<MSM> [I wonder if HT has confused himself a bit -- he says "you don't allow fragids (so you don't support 3986)", but what he means, I think is probably something more like "you don't allow all the fragids defined by the XML RFC, because it says barenames are OK and you don't allow them"]
Kumar: Henry's comment (bullet 1, comment# 3) is incorrect. He says that "You don't allow fragids". This is not correct because we do allow frag ids encoded as smpxpath1() scheme.
<pratul> Here's what I plan to enter in the bug
<pratul> The WG reviewed Comments 3 and 4 and decided that no change is needed since SML does support fragids using smlxpath1() scheme
<MSM> [I finally located the draft replacement for RFC 3023, but it's expired and the IETF site won't show the text. https://datatracker.ietf.org/drafts/draft-murata-kohn-lilley-xml/]
<pratul> The WG reviewed Comments 3 and 4 and believes that no change is needed since SML does support fragids using smlxpath1() scheme
resolution: The WG decided to add the comment typed by Pratul as a response to this bug.
Kumar: I had noted down some comments on RFC 2557. I will send a summary.
<scribe> ACTION: Kumar to send comments on RFC2557/SML-IF comparison and start an email discussion thread. [recorded in http://www.w3.org/2008/04/24-sml-minutes.html#action02]
<trackbot-ng> Created ACTION-182 - Send comments on RFC2557/SML-IF comparison and start an email discussion thread. [on Kumar Pandit - due 2008-05-01].
<pratul> The WG appreciates your flexibility on this issue and expects that the proposed note will facilitate interoperability
<MSM> perhaps 'interoperability of implementations of the XLink reference scheme'
Pratul: any objections to adding
this comment to the bug as the WG's response?
... no objection heard. I will update the bug with this
comment.
<MSM> On 5562, I had an action to draft a response. Draft at http://lists.w3.org/Archives/Public/public-sml/2008Apr/0126.html
<MSM> Comments from Kumar and Kirk later in the thread.
Kumar: Msm has written an excellent response to this bug.
msm: Kirk has made useful suggestions regarding the response text.
Pratul: can you add the response to the bug so that Henry gets to read the response soon?
resolution: Msm to add the response to the bug. The WG will decide during the next call whether to endorse it as-is or with some updates.
<MSM> [For the record, I should say I'll revise the response in light of KDW's comments before posting to Bugzilla. Thanks, Kirk!]
<MSM> I think DE updated the wrong bug
<MSM> I think DE intended to endorse 5541
Kumar: I do not understand David's response in comment# 10. It appears to be aimed at some other bug.
msm explains the context behind David's comment.
msm to summarize his understanding of David's comment and paste into IRC.
<MSM> There may be some confusion here.
<MSM> Henry's original comment raised, in a single paragraph, two distinct
<MSM> technical issues. The issue of XLink support is now being tracked as
<MSM> bug 5561; the issue of schemaless identification of references is now
<MSM> bug 5541.
<MSM> We agree with the XML Schema WG's view that it should be possible to
<MSM> specify sml:ref attributes by adding them, with a default value to the
<MSM> schema. Further details may be found in bug 5541.
<MSM> The issue of base URIs is tracked in bug 5542.
resolution: the comment pasted by msm above is the official response of the WG. Pratul will paste that into the bug.
<pratul> Sorry, Comment 1 had some incorrect boilerplate text. The non-normative note is being drafted and we'll update this bug with a link to the note when it is ready
resolution: The WG decided to add the response typed by Pratul as an interim response to this bug.