OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

regrep message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: [regrep] [Fwd: [regrep-semantic] IBM BI-ICS]


Joe,

OK - despite Scott saying that BC-ICS is not IBMs answer to CAM,
the fact remains that if you have CAM, which we do, CAM is available
in open source - and an OASIS spec' - then there is absolutely no
earthly reason why you would want to do ICS instead.

It's like saying well - I know you have a car - but I'm still going to
ride this bicycle.

Also notice that CAM is highly extensible - fits into
all kinds of other longterm upgrades - like agent directed mapping,
OWL directed discovery of available CAM templates, industry
catalogues of business processes, building domain dictionaries
and vocabularies, then payload validation from ebMS or BPEL,
and on and on.  It's the old adage - once you have this in XML syntax
it can be hugely leveraged.  ICS is frankly a deadend alley - its does
what it does, and that's it; its a one trick pony.  Worse - whatever
you are doing in ICS - you will eventually going to have to re-do
into CAM - because you are going to need what CAM offers
beyond what ICS can offer in an ebSOA deployment.

The place Scott started from with ICS was back last September
at XML2003 in Philly when he saw how much progress we'd
made with CAM - and in his value judgement - he thought he'd
make something simpler and more limited that can be put
together really quickly and simply - works with XSD, and
does not use an ebXML Registry...

So - OK he did that - and he has it.  And he's not having
to use CAM - which was his objective achieved.  That's
fine for Scott - and the quick-fix-it uses he doing with it.
Long term also - he does not want any potential avenue
that leads to any ebXML Registry applications any
time whatever.

However - when you look at the full-up needs you realize
that CAM just does everything better and more
comprehensively than ICS.  BTW - you do NOT have to
use an ebXML Registry with CAM - it functions in
"local definitions" mode just like ICS - if that is all
you want to do with it.

However - people pretty soon realize that
CAM ContentReference section is the key to
building industry dictionaries (like the UDEF folks
are seeing already) and ideally equipped to allow
fullup regisrty directed semantic noun definitions -
and moving them to next-generation
information system integration.  And - yes - the
UBL folks should see that CAM is also CCTS
aware - as we've built all those 'hooks' into CAM
too.

That is what we are working on enabling with
the Registry SCM team today - nouns and
business catalogues that are CAM enabled
and deliver on the vision of dictionary driven
design and information integration.

Last but very much not least is CONTEXT.  Support
for dynamic content assembly and validation is quite
simply where CAM has absolutely no equals today.

Context driven interactions is absolutely the corner
stone of ebSOA - next generation IMHO - that
completes the ability to have agile information
interchanges - that we did not do in ebXML V1.0
because of time-to-market constraints.

So people can follow Scott and his ICS - just
so long as they realize the background and the
focus it has.

Those people interested in fully context driven
information systems and agile SOA implementations
should look at slide #26 in my presentation here,

 http://drrw.net/presentations/ebXML%20Today%20-%20March%2004.zip

and look at the two BPEL use cases for CAM here:

 http://www.oasis-open.org/committees/documents.php?wg_abbrev=wsbpel

and then you realize that ICS cannot touch any of this
to the same level and sophistication that is really
needed and that CAM provides.

That's why I put this presentation together - because
I saw this all coming down the pike...

I like Scott - he's a smart fellow - and he's also a
very good IBM employee.

OK - I'll get off my soap-box now!

Thanks, DW.

----- Original Message ----- 
From: <david.burdett@commerceone.com>
To: <chiusano_joseph@bah.com>; <ebsoa@lists.oasis-open.org>;
<regrep-cc-review@lists.oasis-open.org>; <cam@lists.oasis-open.org>;
<regrep-semantic@lists.oasis-open.org>
Sent: Wednesday, March 17, 2004 5:39 PM
Subject: RE: [ebsoa] IBM BI-ICS


There's more information on this initiative available at [1] and [2] and a
specification at [3]. However Business Payload Composition bit sounds to me
awfully similar in scope to work being done by UBL see [4] which says "The
TC will then design a mechanism for the generation of context-specific
business schemas through the application of transformation rules to the
common UBL source library".

... or am I missing something.

David

----- Original Message ----- 
From: "Chiusano Joseph" <chiusano_joseph@bah.com>
To: <regrep@lists.oasis-open.org>
Sent: Wednesday, March 17, 2004 5:27 PM
Subject: [regrep] [Fwd: [regrep-semantic] IBM BI-ICS]


> [I meant to send this to the main Registry TC]
>
> -------- Original Message --------
> Subject: [regrep-semantic] IBM BI-ICS
> Date: Wed, 17 Mar 2004 17:26:17 -0500
> From: "Chiusano Joseph" <chiusano_joseph@bah.com>
> Organization: Booz Allen Hamilton
> To: ebsoa@lists.oasis-open.org, CCRev
> <regrep-cc-review@lists.oasis-open.org>,CAM
> <cam@lists.oasis-open.org>,Registry TC - SCM SC
> <regrep-semantic@lists.oasis-open.org>
>
> This morning at the XML.gov monthly meeting[1] I saw a presentation
> given by Scott Hinkleman of IBM on IBM's new "Business Integration -
> Information Conformance Standards (BI-ICS)" specification[2]. I found it
> quite interesting, and thought I'd provide a few comments here. Although
> not available as of this e-mail, the presentation should be available at
> the XML.gov site[3] within the next few days, if IBM does make it
> available.
>
> BI-ICS essentially lives at what IBM calls the "Business Level" of the
> Web Services stack, above "Service Composition" (where WS BPEL,
> WS-Transaction, etc. live). I know - it sounds like the "ebXML area" of
> the stack, so we might very well expect some overlap in this and future
> specs - as BI-ICS is apparently one of 4 planned "specification areas"
> for this "framework" (my own use of this term) which is as of yet
> unnamed.
>
> Another area will be "Business Payload Composition", which - Scott
> Hinkleman stated - "has its roots in ebXML Core Components", and
> specifies a "context-driven approach" to payload composition (overlaps,
> anyone?). Scott did, however, state that BI-ICS is "not IBM's answer to
> OASIS CAM".
>
> IBM is soliciting interest in advancing BI-ICS to an open standards
> consortium.
>
> Let the discussions begin...
>
> Kind Regards,
> Joe
>
> [1] http://xml.gov/agenda/20040317.htm
> [2] http://xml.coverpages.org/ni2004-02-02-a.html
> [3] http://xml.gov/presentations.asp
>
> To unsubscribe from this mailing list (and be removed from the roster of
the OASIS TC), go to
http://www.oasis-open.org/apps/org/workgroup/regrep/members/leave_workgroup.php.
>
>



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]