Loading ...
Sorry, an error occurred while loading the content.

1466RE: [TaxoCoP] Re: Visualization technologies

Expand Messages
  • Roy Roebuck
    Jul 20 12:50 PM

      Hi Amanda:


      I do use Agilense WebModeler as the tool I was describing.  You made an astute observation on this, because I don’t think I mentioned Agilense in this forum.  Actually, Agilense has always communicated to me their willingness to let people evaluate their tool with a 45 day evaluation license, but typically under an NDA so that “reverse-engineering” is constrained.  Contact steve@... for further information.  Pass him this email/thread if you like.  Perhaps he has some online capability he can let you explore/evaluate.


      I identified in 2000 that it was the only tool I could find that was flexible enough to support the modeling I need.  I’ve found a few other tools that seem close, but I’ve not proven it yet.  Agilense is not perfect, requires a “techie” orientation to administer the repository and gain the full power of its feature set, and needs supplemental capabilities, built around what I call my “hub and spoke” integration model.  In this hub and spoke, my general/integrating ontology provides the “axle”, a repository like Agilense serves as the “hub”, various data/process integration technologies such as ESB serve as interfacing “hub-bands”, and then a variety of tools serve as “spokes” that support different functions of the enterprise or support different phases of the generalized spiral life cycle management technique we apply.


      Our approach enables rapid/adaptive changes in data, metadata, and meta-metadata, such as the composition of our seven reference catalogs (taxonomies), the relations between the catalog entries, and the attributes that describe these entries and relations.  The key to increasing the stability of this environment is to provide a mechanism for community/domain  consensus building, consensus decisions, and then subsequent variance identification and reporting to the variant author and broader community.


      As to the how, practices, etc., that’s the business I’m in, and I don’t normally disclose too many of its details outside of business talks and engagements.  


      You can review my EA site at http://www.one-world-is.com/beam, and if you’d like internal support with your EA, taxonomies, or ontologies using my approach, contract my services through my current employer, CommIT.







      CommIT Enterprises, Inc.

      Enterprise Architecture for Enterprise Management, Security, and Knowledge

      Roy Roebuck III
      Senior Enterprise Architect

      2231 Crystal Drive, Ste 501
      Arlingon, VA



      +1 (703)-598-2351
      +1 (703) 486-5540
      +1 (703) 486-5506


       Add me to your address book...


      From: TaxoCoP@yahoogroups.com [mailto: TaxoCoP@yahoogroups.com ] On Behalf Of Amanda Xu
      Sent: Thursday, July 20, 2006 2:08 PM
      To: TaxoCoP@yahoogroups.com
      Subject: RE: [TaxoCoP] Re: Visualization technologies


      Thanks, Roy , for showing me your way. What you did is
      amazing and seems very flexible in terms of
      anticipating changes and growth of your EA. What is
      the one tool that you are referring to for M3/M2/M1/M0
      layer modeling - not 'The Brain'? How do you like
      'Agile WebModeler'? I could not get a trial version
      from them.

      Replication and synchronization among M1/M0 is quite
      challenging. What tools do you use? Do you have
      samples or client websites that I can take a look?

      In addition to define object, class, instance, their
      associations and methods in open standards such as
      CWM, MOF, UML, LC classification scheme, etc., keep M2
      layer updated is also quite time consuming and labor
      intensive. I am wondering if you have any suggestion.
      What systems do you use to setup the repository - any
      DB, project server, CMS, version control, search
      services, etc. for the artifacts?

      Luckily, you have the freedom to define your own
      thesaurus with the consensus of end users. As long as
      it is working within your enterprise, it is fine. How
      often do you need to restructure everything due to the
      change of location, organization, resources, etc.? Do
      you have systems to decentralize the workflow for
      maintaining the thesaurus?

      As far I am concerned, we need top level ontology to
      be agreed upon by standard organization such as W2C,
      OMG, OASIS, LC, etc.

      For us, librarians, we have to reflect the truth of
      being or of the thing. Someone recommended UDC today.
      But most of our resources are already in LC
      classification and LC subject headings. Why do we
      need to reinvent the wheel?

      In short, I am still very confused. Thanks a million
      for any help!

      Amanda Xu

      --- Roy Roebuck <Roy.Roebuck@ commitent. com> wrote:

      > Amanda: Thanks for the feedback.
      > The following text is probably a bit overwhelming
      > for most, and even for me sometimes, so I also have
      > diagrams (which I’m also told are often
      > overwhelming) to aid the discussion if needed.
      > Please note that I typically use one tool to do the
      > M3/M2/M1/M0 modeling and model
      > aggregation/ integration/ unification across those
      > layers.
      > Unfortunately, I still have to procedurally use
      > multiple tools (e.g., intelligence mining) and
      > techniques (e.g., user surveys) to efficiently and
      > effectively populate/aggregate the initial M1 layer
      > with the client/user world view (i.e., their
      > ontology) as an independent M1 domain (e.g.,
      > representing a software application, a business
      > process, a form, a database, a concept map), to
      > populate the resultant metadata-modeled M1 domain
      > with relevant M0 client/user data, and to keep the
      > M1 and M0 synchronized with the client’s
      > ever-flowing operational and analytical data and
      > metadata.
      > I then take the resultant independent domain (but
      > now aggregated in my single repository and
      > namespace) M1 layer world views/ontologies (e.g.,
      > process models, data models, UML sequence and class
      > diagrams) and corresponding M0 virtualized or
      > migrated class repositories and/or databases, and
      > relate their class and subclass structures to one or
      > more of my seven generalized reference classes
      > (i.e., my generalized “physics-based”
      > taxonomies) using one or more of my seven relation
      > types with my M2 Layer “Reference Catalog”.
      > From the generalized and integrated ontologies, I
      > then build up my specialized type of
      > “thesaurus”, to unify the diverse and
      > ever-changing vocabulary of the diverse and
      > ever-changing independent world views/ontologies
      > into a unified vocabulary with preferred terms
      > (built largely by consensus of clients/users) and
      > alternate terms (specific to each independent, but
      > now federated, domain).
      > Physical aspects: energy, time, space, matter,
      > intelligence (or dynamic-self- referencing pattern or
      > concept, if you prefer).
      > General Reference Catalogs: Location, Organization,
      > Organization Unit, Function, Process, Resource,
      > Mission Requirement
      > General Relation Types: Categorization (inheritance
      > hierarchy of “like things”), containment (tree
      > or structural “hierarchy” of “unlike
      > things”), sequence (flow), change (flow), variance
      > (flow), equivalence (categorization) , descriptive
      > (has a, about a, attribute, property)
      > Generalized Role Types (for Sequence Relation
      > Types): Customer, Supplier, Authority, Performer,
      > Outsource, Partner, Public)
      > These are sufficient to my “enterprise management
      > intention, but obviously open to infinite
      > variability and discussion. I choose to just be
      > practical, not linguistic or philosophical.
      > CommIT Enterprises, Inc.
      > Enterprise Architecture for
      w:st="on"> Enterprise Management,
      > Security, and Knowledge
      > <blocked::http://www.commiten t.com>
      > Roy Roebuck III
      > Senior Enterprise
      > 2231 Crystal Drive, Ste 501
      > Arlingon ,
      <blocked::http://maps. yahoo.com/ py/maps.py? Pyt=Tmap& addr=2231+ Crystal+Drive, +Ste+501& csz=Arlingon, +VA&country= us>
      > 22202
      > roy.roebuck@ commitent. com
      > <BLOCKED::mailto:roy.roebuck@ commitent. com>
      > mobile:
      > fax:
      > direct:
      > +1 (703)-598-2351
      > +1 (703) 486-5540
      > +1 (703) 486-5506
      > Add me to your address book...
      <blocked::https://www. plaxo.com/ add_me?u= 34360117809& v0=756286& k0=2046596413>
      > ____________ _________ _________ __
      > From: TaxoCoP@yahoogroups .com
      > [mailto:TaxoCoP@yahoogroups .com]
      On Behalf Of axu789
      > Sent: Wednesday, July 19, 2006 11:32 AM
      > To: TaxoCoP@yahoogroups .com
      > Subject: [TaxoCoP] Re: Visualization technologies
      > Thanks Bob for the URL of ‘The Brain’.
      > I agree with Roy
      that it
      > would be ideal to have modeling tools that are
      > compatible with each
      > other for all layers of MDA, at least, from M4, M3,
      > M2, M1 to M0,
      > etc.
      > I would hesitate to use a tool that requires me to
      > piece together
      > different modeling tools for each layer of MDA, in
      > addition to keep
      > track of the upgrades of each version of the
      > individual tools, and
      > compatibility of each manifestation generated by the
      > tools.
      > In addition, I view this infrastructure level data
      > modeling tool to
      > be open in terms of Web enabling standards and
      > technologies, to be
      > independent in terms of the separation of process
      > and application
      > from build to runtime enviornment, to be comfortable
      > in terms of my
      > physical and virtual working environment, where
      > creativity,
      > collaboration, and productivity are enhanced, etc.
      > How many times do I find myself who would rather
      > take the road that
      > I knew than the potential new one which seems to be
      > faster but has
      > too many unknown variables? As far as I am
      > concerned, selecting and
      > buying tools are not simply for the tools, but also
      > for the building
      > of long term partnership with your tool providers.
      > Cheers,
      > Amanda Xu
      > --- In TaxoCoP@yahoogroups .com
      > <mailto:TaxoCoP% 40yahoogroups. com> , "
      w:st="on">Roy Roebuck "
      > <Roy.Roebuck@ ...>
      > wrote:
      > >
      > > I've used TheBrain for several projects since its
      > first beta
      > release,
      > > and still use it. Be aware that while TheBrain is
      > an excellent
      > tool for
      > > modeling and tracking "networks" of ideas/content,
      > it is less
      > capable at
      > > creating "hierarchies" (i.e., classification
      > structure with
      > inheritable
      > > attributes) of content, which are fundamental to
      > taxonomies and
      > > ontologies.
      > >
      > >
      > >
      > > I typically categorize relationships (i.e., verb
      > phrases) into one
      > of
      > > seven "Relation Types": categorization,
      > containment, sequence,
      === message truncated ===

      Amanda Xu
      58-11 197th St .
      Fresh Meadows, NY 11365
      718-990-6716 (voice)
      axu789@yahoo. com (email)

      ____________ _________ _________ _________ _________ __
      Do You Yahoo!?
      Tired of spam? Yahoo! Mail has the best spam protection around
      http://mail. yahoo.com

    • Show all 27 messages in this topic