[regrep] Re: [regrep-comresolve] Some other V2 Spec comments

[Farrukh Najmi <Farrukh.Najmi@Sun.COM>]

Joel,

Please recall that the decision of the team in the last meeting was to accept
any past comments from *you* that you re-submit
or send pointers to as formal comments. It was not a blanket YES to sifting
through all past email looking for
any question about the specs.

This process is getting out of hand and creates unnecessary confusion. I think
that we should only accept formal comments on a formal alias.
Otherwise we will be arguing over what is a formal comment or not.

Kathryn, is there already a formal email alias for submitting external comments?
If not then we can ignore my
suggestion above.

As to your comments that you feel are being ignored, could you please send a
link to the precise set of comments you would like to formally
submit as external comments. Kathryn, I would like to suggest that we dedicate
our next meeting to addressing Joel's comments.

As for yet another meeting prposal, I propose for conducting issue resolution by
email with each issue being assigned a number and initiated
in a separate thread by me (as the team's issue keeper). If anyone else would
like to be the issue keeper I will be glad to hand over that responsibility.

--
Regards,
Farrukh


"Munter, Joel D" wrote:

> To date, within his MS Access database approach, Farrukh has captured a
> single but vital comment from Duane Nickell about the V2 registry model
> being a potential bottleneck and that there is a potential for DoS attacks,
> and my two recent comments related to the normative V2 WSDL files affiliated
> with the RAWS approach.  Duane has submitted both issues within a single
> thread but for better visibility I believe that we should break them out
> into two.
>
> There are other comments which have come in during the formal review period
> that have not yet been captured.  Farrukh and/or others may have provided
> first level response to some of these, but I believe it is important that we
> discuss and agree with the disposition of these comments.
>
> Also, while the trivial ones have been addressed, I will again remind all
> that my own comments about the V2 spec continue to go ignored.
>
> Kit Ko, 7 March 2002,
> In "sec 6.6.2.2 Communication Bootstrapping for ebXML message
> Service", I think all this section is only applied to a "thin client" (as
> defined in sec 6.6.1).
> Am I right?!
>
> Kim Chaemee, 18 December 2001,
> (1) In Figure 6, there is no "Updated" status in lifecycle. Is it right or
> missed?
> (2) In 8.4.2. GetContentResponse Message Structure, Is it
> GetContentResponseMessage? In message fragment, there is <GetContentRequest>
> instead of Response.
> (3) In 9.7. Access Control, there is only 3 role as Content owner, registry
> administrator, registry guest.  However, there is no consistency between
> 5.3. Registry Users, Table1. Actors and Table11. Role. I think it's better
> to have some consistency to describe the role of actors.
> (4) Do you have a plan not to provide RIM DTD anymore? From Developer's
> perspective, sometimes we need DTD instead of Schema.  Some XML Binding SW
> doesn't provide Schema yet.
> (5) Content based query in SQL Query.  Is there specific request & response
> for content based query in SQL Query?
>
> Kyu-Chul Lee, 9 January 2002,
> I'd like to ask that the OASIS ebXML Registry V2.0 is backward compatible
> with V1.0 or not.
> There are already many implementations of ebXML Registry V1.0.
> I think it is required to guarantee the backward compatibility in order to
> save their investments.
>
> Nita Sharma, 11 February 2002,
> We(the ebXML BP Catalog team) had a long conversation last week with Kathryn
> about unique identification and what their scheme should be. We provided her
> with our requirements for unique identification that was not satisfied by
> the current UUID specification of regrep. The various things that we touched
> base upon were:
>
> 1. Meaningful verses meaningless identification scheme
> 2. multiple identification scheme for the same item based on various usages.
> 3. A standard organization (like UCC/EAN) to control the uniqueness and
> meaningfulness
> 4. Separate namespace for the various identification schemes
> 5. Analyze other schemes like IDEF, OID etc.
>
> And finally, when can we all meet to assign tasks and review the status of
> each of these issues/comments.  I propose either Friday 15 March 3-4pm MST
> (5-6pm EST) or Monday 18 March 9-10am MST (11-12pm EST).
>
> Thanks,
> Joel Munter
> Distributed Systems, Intel Labs
> joel.d.munter@intel.com
> (480) 552-3076
> (602) 790-0924 (cell)
>
> ----------------------------------------------------------------
> To subscribe or unsubscribe from this elist use the subscription
> manager: <http://lists.oasis-open.org/ob/adm.pl>