Subject: Re: Special note for CPP members
There is consensus in the TP team not to define a negotiation protocol. [As
a participant
in both,] I am not in favor of the proposed TA wording due to this fact .
It does certainly effect the *level* of interoperability with specific
ebXML parts, but I see
no "make or break" issues in ebXML --parts will be able to be used to some
levels of interoperability,
with or without other parts, achieving the goal of loosely-coupled modules.
This is much more analog than binary.
If such protocol would eventually be defined it is far from designing an
implementation, so I would
think it to be a candidate in some future life.
Thanks,
Scott Hinkelman, Senior Software Engineer
XML Industry Enablement
IBM e-business Standards Strategy
512-823-8097 (TL 793-8097) (Cell: 512-940-0519)
srh@us.ibm.com, Fax: 512-838-1074
Martin W Sachs/Watson/IBM@IBMUS on 01/30/2001 07:51:40 AM
To: Duane Nickull <duane@xmlglobal.com>
cc: "Welsh, David" <David.Welsh@nordstrom.com>, "Bob Haugen (E-mail)"
<linkage@interaccess.com>, "Brian Hayes (E-mail)"
<Brian.Hayes@Commerceone.com>, ebXML-StC <ebxml-stc@lists.ebxml.org>,
matt@xmlglobal.com, ebxml-tp@lists.ebxml.org
Subject: Re: Special note for CPP members
Duane,
There was a better way. You could have phoned me to discuss before hitting
me with proposed TA text that is really in the province of the Requirements
team to begin with (for example: "A CPA negotiation protocol SHALL be
defined...").
Had you phoned me, I would have said that:
The TP team is planning a non-normative appendix describing what
knowledge we are developing about CPA composition and negotiation
techniques. It is likely that this will be sufficient to guide the
implementers.
The TP team is still arguing about some fundamental structural issues,
so that even the non-normative appendix is in the future (though most
likely, it will be in V1.0).
While I am extremely pleased that companies are already working on
reference implementations of composition/negotiation, they should take
heed of the current version number (0.29) since it is a heads-up that a
lot could change before V 1.0.
...and, as I did say, the TP team is very sensitive to the need for a
CPP structure that is composable into the CPA. The best way to handle
the questions that have been raised is for those people who are working
on reference implementations to feed their questions and concerns to the
TP team via its listserver since many (if not all) of those issues
should be addressed in the design of the CPP itself rather than by
putting together a normative specification that is really software
design and does not involve interoperability as we usually understand
interoperability.
Regards,
Marty
Regards,
Marty
*************************************************************************************
Martin W. Sachs
IBM T. J. Watson Research Center
P. O. B. 704
Yorktown Hts, NY 10598
914-784-7287; IBM tie line 863-7287
Notes address: Martin W Sachs/Watson/IBM
Internet address: mwsachs @ us.ibm.com
*************************************************************************************
Duane Nickull <duane@xmlglobal.com> on 01/29/2001 07:28:01 PM
To: Martin W Sachs/Watson/IBM@IBMUS
cc: "Welsh, David" <David.Welsh@nordstrom.com>, "Bob Haugen (E-mail)"
<linkage@interaccess.com>, "Brian Hayes (E-mail)"
<Brian.Hayes@Commerceone.com>, ebXML-StC <ebxml-stc@lists.ebxml.org>,
matt@xmlglobal.com
Subject: Re: Special note for CPP members
Marty:
I agree. I also acknowledge that you have stated all along that the CPA
negotiation is outside of the scope of your group.
But...
as people are starting to actually implement this type of software,
they are seeing that this is potentially a make or break issue.
Problem:
You are right that it is very unlikely such an effort could be
effectively started (nevermind completed) before May. It will likely be
totally dependant on the work your group is doing and probably cannot be
effectively scoped beforehand.
I think this issue is rather large therefore the Steering committee was
notified.
I don't have a solution either so let's hope someone out there is
listening and has some spare time to ponder...
Duane
Martin W Sachs wrote:
>
> Duane,
>
> Since your last email was copied to the Steering Committee, I believe
that
> it is appropriate for the Steering committee to see my response in full
> along with your original email (below).
>
> Regards,
> Marty
>
> Martin W Sachs
> 01/29/2001 06:48 PM
>
> To: Duane Nickull <duane@xmlglobal.com>
> cc: "ebxml-tp@lists.ebxml.org" <ebxml-tp@lists.ebxml.org>
> From: Martin W Sachs/Watson/IBM@IBMUS
> Subject: Re: Special note for CPP members (Document link: Martin W.
> Sachs)
>
> Duane,
>
> The TP team discussed CPA generation from CPPs and concluded that the
> generation process is outside the team's scope as initially constituted.
> The team did set itself a requirement of defining the CPP and CPA such
that
> composition and negotiation are possible.
>
> There is a high level issue with your proposal. I believe that a lot of
> people will argue that defining the CPA composition process at that level
> of detail is designing the implementation. Since there is no
> interoperability issue in the CPA composition software, it is not at all
> clear that ebXML should define a standard that is for all practical
> purposes software design in an area where there is no harm in two
different
> CPA composers doing it differently. I believe that a decision to define
a
> composition standard should be reviewed at the highest levels of ebXML.
>
> If there are specific concerns that people designing CPA composers have,
> they can bring these to the attention of the TP team. It may be that
> specific concerns can be addressed with specific changes to details of
the
> CPP/CPA specification without having to design a composer.
>
> In any case, this team only began its work in August, giving it just 5
> months to settle on requirements and then complete a version 1.0 spec
> (measured from August to the start of the QR cycles for April, a deadline
> that we did not meet). So even if ebXML approves such a requirement on
the
> TP team, the team could not possibly start work on it until after version
> 1.0, which means after May, 2001 (assuming that ebXML continues to exist
> beyond May).
>
> I urge you not to add this requirement at this time since it can't
possibly
> be fulfilled. If ebXML continues after May, a CPA composition standard
can
> be discussed at the May or July 2001 meeting.
>
> Regards,
> Marty
>
> P.S., the terms are "Collaboration Protocol Agreement" and "Collaboration
> Protocol Profile". I believe that I pointed this out in my comments to
the
> TA spec.
>
*************************************************************************************
>
> Martin W. Sachs
> IBM T. J. Watson Research Center
> P. O. B. 704
> Yorktown Hts, NY 10598
> 914-784-7287; IBM tie line 863-7287
> Notes address: Martin W Sachs/Watson/IBM
> Internet address: mwsachs @ us.ibm.com
>
*************************************************************************************
>
> Duane Nickull <duane@xmlglobal.com> on 01/29/2001 05:40:18 PM
>
> To: "ebxml-tp@lists.ebxml.org" <ebxml-tp@lists.ebxml.org>
> cc:
> Subject: Special note for CPP members
>
> Hello all:
>
> As we conclude the TA Specification and the disposition of comments, it
> has become apparent that there is a potential shortcoming on
> specifications regarding the Trading Partner issues, specifically
> concerning CPA generation from CPP's and business processes. In order
> to facilitate CPA negotiation, people who are building reference
> implementations have informed us that they believe it is necessary to
> observe a standard protocol for deriving a CPA from CPP's.
>
> Therefore, we have added two small sections to the technical
> architecture specification (NOTE: not officially approved by the TA team
> yet) which read as follows:
>
> "CPA negotiation SHALL be strictly defined. Issues such a precedence,
> prioritization and the mechanics of the negotiation process SHALL be
> addressed in the ebXML Specifications governing Collaborative Protocol
> Agreements."
>
> "A CPA negotiation protocol SHALL be defined by the ebXML TP Project
> Team."
>
> Comments please? (Today if possible)
>
> Duane Nickull
> TA Team
>
>
*************************************************************************************
>
> Martin W. Sachs
> IBM T. J. Watson Research Center
> P. O. B. 704
> Yorktown Hts, NY 10598
> 914-784-7287; IBM tie line 863-7287
> Notes address: Martin W Sachs/Watson/IBM
> Internet address: mwsachs @ us.ibm.com
>
*************************************************************************************
Powered by
eList eXpress LLC