Paul,
The CCA (Component Collaboration Architecture) part of EDOC is, essentially,
an architectural meta model for SOA and is, I think, about 60 pages. What
you list below have various issues; (01)
Technology specific; WSDL, BPEL, ebXML (The ebXML Business Process Schema is
very similar but is "flat", no recursion). (02)
Not an SOA model; OWL, Topic Maps, BPMN (03)
So general that you need a profile (such as EDOC): UML, OWL, Topic Maps (04)
Not sufficiently specific to support an executable business models: BCN. (05)
The others I just don't know that well or are (IM(~)HO) just a redo of
existing work. EDOC is about 5 years old so it pre-dates the SOA buzz word,
but it is the same thing. What it allows is business centric
role/collaborations/interactions and recursive composition/decomposition of
both the roles (parties responsible for services) and interactions - the
interactions are what shows up on the surface as a "service" in the WSDL
sense. Since it is an architecture model you can map it to any number of
the technologies you mention and more. Frankly, none of the others do this
- it requires bits and pieces from multiple other specs and then you have to
figure out how to integrate them, which is why you want an architecture
model. (06)
What EDOC/CCA does NOT provide is semantic grounding - none of the
architecture oriented standards/tools do. This is an area we are very
interested in and trying to address and the reason for me causing trouble on
the onto-lists. The other area that needs improvement is the like to
business activity modeling, which is being done in another current effort. (07)
So, we use EDOC/CCA, I helped author parts of it - I am bias. I am not
saying it is the only way to do things, it works well for us, for some
others and for customers (like GSA). But, a lot of the angle-bracket stuff
is not nearly as complete or mature and is more complex - in general we just
think of that as stuff to generate with the MDA tools. We have but have not
yet published an OWL version of the same thing - the semantics matter, not
the representation. (08)
-Cory (09)
> -----Original Message-----
> From: soa-forum-bounces@xxxxxxxxxxxxxx [mailto:soa-forum-
> bounces@xxxxxxxxxxxxxx] On Behalf Of Paul S Prueitt
> Sent: Thursday, April 06, 2006 3:04 PM
> To: Service-Oriented Architecture CoP
> Subject: RE: [soa-forum] SOA and a peer review of mark - 3 technology
>
> One of the issues with these several hundred page "standards" is that by
> the
> time one gets to something of substance, the acronymns are so common that
> one can not fully understand without a major commitment of time. So I can
> not tell if these is a duplication of something (perhaps far simplier) in
> the BPEL (business process execution language) plus ebXMP plus etc...
>
> Why do you feel that this OMG effort covers something not covered in :
>
> > WSDL
> > ebXML
> > BCM
> > Topic Maps
> > OWL Full
> > BPEL
> > BPMN
> > UML
> > SOA-IM
> > SOA-CS
> > FERA
>
> or covers "things" in these 11 specification in a better way.
>
> Why do you make this recomendation? SOA-CS would seem to be sufficient
> for
> the refinement/enhancement of OASIS BCM into a community centric Service
> Methodology. This is what the list of 11 is suggesting to me.
>
>
>
> -----Original Message-----
> From: soa-forum-bounces@xxxxxxxxxxxxxx
> [mailto:soa-forum-bounces@xxxxxxxxxxxxxx]On Behalf Of Cory Casanave
> Sent: Thursday, April 06, 2006 11:15 AM
> To: 'Service-Oriented Architecture CoP'; 'Andrew S. Townley';
> rlballard@xxxxxxxxxxxxx
> Cc: 'Nathan Einwechter'; 'Dennis L. Thomas'; 'David RR Webber (XML)';
> 'John F. Sowa'
> Subject: RE: [soa-forum] SOA and a peer review of mark - 3 technology
>
>
> For your list of models of SOA, you should really include the
> OMG-EDOC-ECA-"Component Collaboration Architecture":
> http://www.omg.org/cgi-bin/doc?formal/2004-02-01
>
> > -----Original Message-----
> > From: soa-forum-bounces@xxxxxxxxxxxxxx [mailto:soa-forum-
> > bounces@xxxxxxxxxxxxxx] On Behalf Of Paul Prueitt (ontologystream)
> > Sent: Thursday, April 06, 2006 3:59 PM
> > To: 'Andrew S. Townley'; rlballard@xxxxxxxxxxxxx; Service-Oriented
> > Architecture CoP
> > Cc: David RR Webber (XML); Nathan Einwechter; 'Dennis L. Thomas'; John
F.
> > Sowa
> > Subject: [soa-forum] SOA and a peer review of mark - 3 technology
> >
> > Andrew,
> >
> > Your effort to understand what Ballard has been doing is paying off, it
> > is
> > easy to see. You are covering ground that takes some time for those who
> > are
> > prepared, and can not be traversed by those who are not prepared.
> >
> > The dependency of the full reality of a "concept" on an experiential
> (see
> > work by Peirce and others in semiotics) or perceptual act assigns to
> this
> > full reality a "pragmatic" axis where category formation is "in process"
> > and
> > under re-creation. Thus "knowledge representation" of a concept is both
> > relative to the experience and is not (fully) stationary.
> >
> > But to the extend that the categories forming are strongly similar to
> > "invariance" in other concepts (also being experienced), then we can see
> a
> > shared scope to "something". We also see the bases for QSAR.... as
> > discussed in Chapter 4 and 6 of
> >
> > http://www.bcngroup.org/area3/pprueitt/book.htm
> >
> > The second school calls this perceptual invariance, and "semantic
> covers"
> > are related to (in my mind) Sowa, Ballard, Adi, and other theories of
> > semantic primitive.
> >
> > At this point we face some technical difficulties. Should the real time
> > measurement of social discourse use the Sowa, the Ballard, or the Adi
> set
> > of
> > primitives - (or are each of these sufficient in specific ways and not
> > sufficient in other ways)?
> >
> > Over the next 30 -45 days I am focusing my effort on the knowledge
> > management community and practices that have been developed and used for
> > the
> > purpose of knowledge elicitation.
> >
> > The knowledge elicitation might be fully specified as a OASIS standard,
> > and
> > we are looking at this carefully as a means to support community centric
> > service methodology (CCSM). We might also see a supporting
> specification
> > on
> > SOA with Topic Maps.
> >
> > A successful OASIS CCSM specification will need to show relationship and
> > use
> > of the following
> >
> > WSDL
> > ebXML
> > BCM
> > Topic Maps
> > OWL Full
> > BPEL
> > BPMN
> > UML
> > SOA=IM
> > SOA-CS
> > FERA
> >
> > I do not see ebSOA distinctly and have left this out of my list for now,
> > since I feel that ebXML and FERA and BPEL cover the space.
> >
> > Stratification of the individual conceptualization (as in the BCM lower
> > layer of the four BCM layers) and automated aggregation of a community
> > layer
> > (BCM's "business layer") is the key to the technological innovation
> > suggested in the BCNGroup RoadMap.
> >
> >
> >
> >
> >
> >
> >
> > -----Original Message-----
> > From: Andrew S. Townley [mailto:andrew.townley@xxxxxxxxxxxxxxxx]
> > Sent: Thursday, April 06, 2006 1:26 AM
> > To: rlballard@xxxxxxxxxxxxx
> > Cc: Dennis L. Thomas; Paul Prueitt (ontologystream)
> > Subject: RE: a peer review of mark - 3 technology
> >
> >
> > Hi Dick,
> >
> > Thanks for answering. No problems on the delay.
> >
> > Please forgive the naive questions, I'm still on a bit of a vertical
> > learning curve with this stuff. However, with the help of Google and
> > archived comments from yourself, I think I answered my fundamental
> > question: how does your work deal with the idea that the understanding
> > of a concept is dependent on the person understanding/perceiving that
> > concept within their current context (as stated by David Bohm in
> > "Wholeness and the Implicate Order")?
> >
> > Please correct me if I'm wrong, but I think the answer is: it doesn't.
> >
> > I found some information from a discussion between Paul, yourself and
> > John Sowa from back in 2001 along with the PDF version of the FAQ and
> > "First Commercial Knowledge Asset Production Process" from your
> > website. In digesting this, I think I understand the following:
> >
> > * a knowledge base is initially created to address a specific need,
> > therefore it is only representing the theories which directly relate to
> > the business needs of the stakeholders.
> >
> > In relation to my question, this means that you have created/defined a
> > shared understanding of reality within a given domain.
> >
> > * Within this established knowledge base and set of "compatibile"
> > theories, it is clear that there can be only one unique definition for
> > each concept, otherwise the fundamental theories would not be
> > compatible.
> >
> > * Paul's "choice points" that he's been talking to me about are really
> > each n-ary concept node with connecting nodes in the overall concept
> > graph.
> >
> > If I understand what I read, these connections are actually based on
> > assumptions or constraints which have been identified as related during
> > the creation of the model and populated during the ingest of information
> > into the system.
> >
> > * From the above, the ontology represented by the knowledge asset is
> > domain-specific.
> >
> > * The knowledge base is not intended to explore new themes outside the
> > original domain or provide considerations of alternate views of
> > "reality", because they would change the domain scope of the ontology.
> >
> > I'm lead to this conclusion from the phrase in the FAQ:
> >
> > "until every question the knowledgebase was intended to answer." (bullet
> > #1, page 15)
> >
> > >From this, I am assuming that adding additional information into the
> > knowledge base will create new connections and relationships within the
> > concept graph, but these will be based on the theories codified into the
> > creation of the "knowledge operating system" when the asset is
> > designed. From this, I am curious how easy it is to expand the theories
> > within a knowledge asset to new domains. At what point does the
> > cohesiveness and uniqueness of a concept break down? Is this possible,
> > or is it explicitly or implicitly prevented within the design of the
> > system?
> >
> > In the interests of full disclosure, I am only beginning to be exposed
> > to KM, KR, semantics and ontology, so maybe these are questions that are
> > obvious once I have the background. Also, feel free to ignore or
> > answer/address any of the above as you see fit. I'm sure you're very
> > busy.
> >
> > >From what I've read over the last couple of days, I really do think
> your
> > work is very interesting. As a human, some of the future implications
> > of it if it becomes as successful as you want are a bit scary, but I can
> > certainly relate to the drivers you mention originating from
> > conventional software system design.
> >
> > At any rate, all this has gotten me interested enough in aspects of this
> > field to try and learn more about it. Having read the syllabus from
> > your Winter Quarter 2005 course, I hope you reach the audience you
> > intend. It is clear that even through exposure to the material in the
> > briefest way, it expands the way one thinks.
> >
> > Thanks very much for your time,
> >
> > ast
> >
> > On Wed, 2006-04-05 at 13:07, rlballard@xxxxxxxxxxxxx wrote:
> > > Paul and Andrew
> > >
> > > Sorry to be so slow in responding.
> > >
> > > We are getting the new course organized today with some uncertainty as
> > to
> > > UCI Extension's ability to reach the audience we would like. In past
> > > iterations Dennis put the class together out of senior project
> > management
> > > specialists and business owners. This time we are trying more
> diligently
> > to
> > > reach into the younger labor pool of want-to-be knowledge engineers.
> > >
> > > From my perspective as a physical scientist, the significant
> > contributors
> > to
> > > any n-ary are "degrees of freedom" -- decisive choices seemingly
> > available
> > > to any decision maker. As with any use of theory, the ultimate
> > correctness
> > > of the list they compile may be guilty of mixing apples with oranges.
> > >
> > > Still we can expect that techniques like factor analysis, sensitivity,
> > etc.
> > > will progressively speak of first order, second order, etc.
> > approximations
> > > to one or more operant theories. Engineers typically focus on
> "decision
> > > drivers" -- the presence of compelling theories ("conops", concepts of
> > > operation) with the power to force a decision into nearest first
> > alignment
> > > with the goal sought. Once in the ballpark, then all other issues may
> be
> > > examined for their particular beneficial or adverse consequences in
> > > assessing their ultimate decision impact.
> > >
> > > Clearly the situational degrees of freedom are easiest to argue as
> > > potentially most relevant, unless the decision path studied explicitly
> > voids
> > > their consideration or need.
> > >
> > > The necessary presence of human decision makers with varying degrees
> of
> > > experience and authority is an early requirement and necessity in
> > avoiding
> > > asset liabilities beyond those acceptable as simple helpful advice.
> > > Obviously the early race to reference dominance will begin as soon as
> > Mark
> > 3
> > > hits the market and demonstrates its competitive virtues. Thereafter
> the
> > > first stylistic race is on -- in defining the unique style and format
> to
> > be
> > > associated with "patterns of thought." Decisions to quibble on that
> > > characterization and its unique, short-term memory requirement are
> > likely
> > to
> > > be increasingly fruitless -- here forward.
> > >
> > > Once the earliest styles and virtue are accepted, then the reference
> > > dominance race will favor those pushing market closing initiatives
> based
> > > upon existing, non-electronic knowledge assets and previously
> marketable
> > > reputations for authority. This race will be hard and costly because
> the
> > > scope of resources needed will jump quickly to whole libraries.
> > Publishers
> > > are typically most conservative and they compete constantly to realign
> > and
> > > re-factor their asset lists competitively -- so publishing mergers and
> > > acquisitions should become endemic.
> > >
> > > Many vendors will step forward and compete heavily on cost and massive
> > > strategic acquisitions from available public sources. Of necessity
> these
> > > must concentrate first in specific job related categories. The
> earliest
> > > offerings will offer easy pickings for everyone claiming scholarly
> > > authority, but that kind of nitpicking can be overwhelmed by the
> massive
> > > scale of non-electronic holdings available. The more constructive
> > approaches
> > > are those that favor professional detailing of theory-acceptance and
> > > outcomes base evidences of successful practice.
> > >
> > > The most important early work fill be aimed at modeling
> standardizations
> > > based upon conventional 2-dimensional representations (mediating
> > structures)
> > > re-expressed in n-dimensional forms. That activity forces us to
> > radically
> > > rethink our linguistic bias toward subsumption hierarchies -- the
> > subject
> > of
> > > my "first lesson's learned letter."
> > >
> > > These are extraordinary times for new leadership to step forward --
> jump
> > in,
> > > the field is wide open.
> > >
> > > Dick
> > >
> > >
> > > -----Original Message-----
> > > From: Paul Prueitt (ontologystream) [mailto:psp@xxxxxxxxxxxxxxxxxx]
> > > Sent: Wednesday, March 29, 2006 2:41 PM
> > > To: 'Andrew S. Townley'; rlballard@xxxxxxxxxxxxx; Dennis L. Thomas
> > > Subject: RE: a peer review of mark - 3 technology
> > >
> > >
> > > Dick and Dennis
> > >
> > > Andrew is asking a question about the uniqueness of the semantic n-ary
> > >
> > > I am finding that Andrew asks many of the same type of questions as I
> > > do/did.
> > >
> > > Perhaps, Dick; if you have time, you could reflect on why this
> question
> > came
> > > up, and how you might answer.
> > >
> > >
> > >
> > > -----Original Message-----
> > > From: Andrew S. Townley [mailto:andrew.townley@xxxxxxxxxxxxxxxx]
> > > Sent: Tuesday, March 28, 2006 5:36 PM
> > > To: Paul Prueitt (ontologystream)
> > > Subject: Re: a peer review of mark - 3 technology
> > >
> > >
> > > Hi Paul,
> > >
> > > Thanks for the slides. They were very interesting. Wow. If that's
> > > what you folks have been up to, it's very, very cool. I didn't think
> it
> > > made sense to respond to everyone, because I just had a question maybe
> > > you could clear up a bit.
> > >
> > > In the slides, the implication is that there is always one single
> > > semantic definition for a concept or thing. How does this relate to
> > > context? On page 16, it seems to be involved as assumptions, but what
> > > about perception? How does this fit in with the likes of Bohm and
> > > Feynman (and even Heraclitus for that matter) on reality and truth not
> > > being fixed notions?
> > >
> > >
> > > The KM/OASIS reading is on the agenda for tomorrow. Should be able to
> > > get through a lot of what I have.
> > >
> > > Thanks again for the slides--fascinating...
> > >
> > > ast
> > >
> > > On Tue, 2006-03-28 at 17:35, Paul Prueitt (ontologystream) wrote:
> > > > Azamat,
> > > >
> > > >
> > > >
> > > > You may wish to consider this presentation and to make a principled
> > > > discussion about the way in which Dick Ballard is using his terms..
> > > >
> > > >
> > > >
> > > > To assert specific implications.
> > > >
> > > >
> > > >
> > > > I agree with John Sowa on many elements of analysis, particularly
> > > > about the need to pull back into a conservative position, with
> respect
> > > > to use of language to assert things implicitly.
> > > >
> > > >
> > > >
> > > > For example "machines that know each other" or "machines that know".
> > > > can both be ok in context to what are the implicit assertions IF a
> > > > second school position is assumed. So we (the second school) mean
> > > > that through human use the n-ary information structure in the Mark 3
> > > > will use structural information to bring finite state machines to a
> > > > specific state. These state transitions and the representation of
> > > > structure within information (represented in the n-ary form) is
> > > > clearly a simplification of the current (XML registry/repository)
> > > > standards.
> > > >
> > > >
> > > >
> > > > That the nary transforms is conducting a type of "knowledge
> > > > processing", I would concur. But I am close to the boundary - and
> do
> > > > not wish to step into first school language use.
> > > >
> > > >
> > > >
> > > > A "knowledge operating system" is possible (as Dick, Don Mitchell
> and
> > > > I have been discussing since 2000).
> > > >
> > > >
> > > >
> > > > No buzz here just a proper and clear description of the magic that
> > > > could be available if n-ary ontology is used properly ( as I feel
> the
> > > > Mark 3 will allow). But there is no need to give the assertion that
> > > > the machine becomes endowed with a soul .. (An interesting
> discussion
> > > > could be engaged here regarding quantum computing and the emergence
> of
> > > > a machine spirit due to the non-locality effects).
> > > >
> > > >
> > > >
> > > > So the trick is to stay away from inferred assertions that end up
> > > > bothering some people, even business people.
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > You know my position well, so perhaps you might orient your thinking
> > > > to second school viewpoint (as much as you feel is comfortable). and
> > > > see if there is some minor correct to terminological use that would
> > > > help both Ballard and the knowledge science revolution.
> > > >
> > > >
> > > >
> > > > Your work could quite easily be converted into the n-ary form and
> > > > reside as a Mark 3 resource.. As a matter of fact.
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > I have included in the bcc some who may wish to join a discussion
> > > > about this. but which I place in bcc so that they do not feel a need
> > > > to respond unless they have time to do so.
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> ______________________________________________________________________
> > > >
> > > > From: Dennis L. Thomas [mailto:dlthomas@xxxxxxxxxxxxxxxxxxxxxxxx]
> > > > Sent: Monday, March 27, 2006 8:06 PM
> > > > To: Linda M Wynott
> > > > Cc: Richard L. Ballard
> > > > Subject: Re: Presentation PDF - Delphi II event, Phoenix
> > > >
> > > >
> > > >
> > > >
> > > > Linda,
> > > >
> > > >
> > > >
> > > >
> > > > Please find attached Dr. Ballard's Creating Systems That Know
> > > > presentation in PDF form. This is a 3.1MB file.
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > Please know that as a result of the high caliber crowd that will be
> > > > attending the Phoenix event, Dr. Ballard has decided to unveil his
> > > > entire Knowledge Science as a lead into the F/A-18
> database/simulation
> > > > integration project.
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > This is significant because for the first time Ballard presents his
> > > > entire KNOWLEDGE SCIENCE, defines what KNOWLEDGE is based on this
> > > > science, what KNOWLEDGE MANAGEMENT means from the perspective of
> > > > knowledge science, how KNOWLEDGE can be measured, and how machines
> can
> > > > capture knowledge at absolute bit limits - and reason with that
> > > > knowledge RATIONALLY like humans do. This presentation also explains
> > > > how Knowledge Foundations' N-Dimensional technology integrates
> > > > unlimited concepts, ideas, thought patterns and the theory that
> gives
> > > > them meaning into PREDICTIVE SEMANTIC WEBS.
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > Your interest in this valuable work, and your invitation to present
> at
> > > > Delphi Phoenix, is very much appreciated.
> > > >
> > > >
> > > >
> > > --
> > > Join me in Dubrovnik, Croatia on May 8-10th when I will be speaking at
> > > InfoSeCon 2006. For more information, see www.infosecon.org.
> > >
> > >
> >
> **************************************************************************
> > **
> > > ***********************
> > > The information in this email is confidential and may be legally
> > privileged.
> > > Access to this email by anyone other than the intended addressee is
> > > unauthorized. If you are not the intended recipient of this message,
> > any
> > > review, disclosure, copying, distribution, retention, or any action
> > taken
> > or
> > > omitted to be taken in reliance on it is prohibited and may be
> unlawful.
> > If
> > > you are not the intended recipient, please reply to or forward a copy
> of
> > > this message to the sender and delete the message, any attachments,
> and
> > any
> > > copies thereof from your system.
> > >
> >
> **************************************************************************
> > **
> > > ***********************
> > >
> > >
> > >
> > --
> > Join me in Dubrovnik, Croatia on May 8-10th when I will be speaking at
> > InfoSeCon 2006. For more information, see www.infosecon.org.
> >
> >
> **************************************************************************
> > **
> > ***********************
> > The information in this email is confidential and may be legally
> > privileged.
> > Access to this email by anyone other than the intended addressee is
> > unauthorized. If you are not the intended recipient of this message,
> any
> > review, disclosure, copying, distribution, retention, or any action
> taken
> > or
> > omitted to be taken in reliance on it is prohibited and may be unlawful.
> > If
> > you are not the intended recipient, please reply to or forward a copy of
> > this message to the sender and delete the message, any attachments, and
> > any
> > copies thereof from your system.
> >
> **************************************************************************
> > **
> > ***********************
> >
> >
> > _________________________________________________________________
> > Subscribe/Unsubscribe/Config:
> http://colab.cim3.net/mailman/listinfo/soa-
> > forum/
> > Shared Files: http://colab.cim3.net/file/work/soa/
> > Community Portal: http://colab.cim3.net/
> > Community Wiki: http://colab.cim3.net/cgi-
> bin/wiki.pl?AnnouncementofSOACoP
>
> _________________________________________________________________
> Subscribe/Unsubscribe/Config:
> http://colab.cim3.net/mailman/listinfo/soa-forum/
> Shared Files: http://colab.cim3.net/file/work/soa/
> Community Portal: http://colab.cim3.net/
> Community Wiki: http://colab.cim3.net/cgi-bin/wiki.pl?AnnouncementofSOACoP
>
>
> _________________________________________________________________
> Subscribe/Unsubscribe/Config: http://colab.cim3.net/mailman/listinfo/soa-
> forum/
> Shared Files: http://colab.cim3.net/file/work/soa/
> Community Portal: http://colab.cim3.net/
> Community Wiki: http://colab.cim3.net/cgi-bin/wiki.pl?AnnouncementofSOACoP (010)
_________________________________________________________________
Subscribe/Unsubscribe/Config: http://colab.cim3.net/mailman/listinfo/soa-forum/
Shared Files: http://colab.cim3.net/file/work/soa/
Community Portal: http://colab.cim3.net/
Community Wiki: http://colab.cim3.net/cgi-bin/wiki.pl?AnnouncementofSOACoP (011)
|