ontolog-forum
[Top] [All Lists]

RE: [ontolog-forum] Where does a taxonomy fit into a conceptual datamod

To: "[ontolog-forum] " <ontolog-forum@xxxxxxxxxxxxxxxx>
From: Rex Brooks <rexb@xxxxxxxxxxxxxx>
Date: Sat, 4 Mar 2006 07:58:18 -0800
Message-id: <a06230919c02f5e5fa03d@[67.103.37.134]>
Thanks Matthew,    (01)

for the excellent technical discussion.    (02)

I wanted to add some standards and application discussions to this topic.    (03)

In the OASIS Emergency Management Technical Committee (EM TC) we have 
developed the Emergency Data Exchange Language Distribution Element 
(EDXL_DE) Specification . It is currently in the final days of a 
second 15-day Public Comment (PC) Period, subsequent to the initial 
required 60-day  PC. The DE part of EDXL_DE means that it is a 
subspecification of the EDXL family of specifications. This is a 
taxonomical relationship.    (04)

Within the DE we have specified certain "Distribution Types" of EDXL 
Messages, such as Report, Update, Request, Response, etc. These are a 
subset that we determined it was necessary for us to specify.    (05)

However, for other elements and subelements we determined that it was 
not within our scope to specify, especially where groups exists that 
either have already  published their own terminologies with 
taxonomies within them, or else there is significant reason for us to 
expect that there were or could be conflicts in how different groups 
might need to define similar or identifical terms. So, for those 
elements, such as Event Types, we specified that messages must cite a 
keyword which is associated with a published and maintained list from 
which an Event Type for that message is used.    (06)

We specified that these lists would each be accessed by through a 
value list urn, which is available through the usual means.    (07)

This kind of structure implicitly calls for a taxonomy but, because 
we must also make our specifications usable by different kinds of 
devices, including handhelds, we did not restrict this further.    (08)

We work with practioner-developer groups as we do this work, to 
develop these taxonomical relationships in such a way that building 
applications that use our specifications will be facilitated. At 
least that is our aim at the same time that providing a standard aims 
to make such applications interoperable. However, that is another 
topic. The point is that, as with other fields where much standards 
activity is going on, it is likely that the use of taxonomies and 
ontologies will increase as experience develops and the practical 
benefits are documented.    (09)

That last statement is another key to building the momentum of 
adoption of standards generally, and taxonomies/ontologies in 
particular.    (010)

Regards,
Rex    (011)


At 10:34 AM +0000 3/4/06, West, Matthew R SIPC-DFD/321 wrote:
>Dear Kathleen,
>
>Well as the person who is responsible for developing Shell's
>conceptual data model fot its downstream business I will try
>to answer this question.
>
>See below
>
>
>Regards
>
>Matthew West
>Reference Data Architecture and Standards Manager
>Shell International Petroleum Company Limited
>Shell Centre, London SE1 7NA, United Kingdom
>
>Tel: +44 20 7934 4490 Mobile: +44 7796 336538
>Email: matthew.west@xxxxxxxxx
>http://www.shell.com
>http://www.matthew-west.org.uk/
>
>
>>  -----Original Message-----
>>  From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
>>  [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx]On Behalf Of Kathleen A
>>  Ellis
>>  Sent: 03 March 2006 19:48
>>  To: [ontolog-forum]
>>  Subject: [ontolog-forum] Where does a taxonomy fit into a conceptual
>>  datamodel?
>>
>>
>>
>>
>>
>>
>>  This week, a member of our TaxCop (Taxonomy Community of
>>  Practice) group
>>  posted this question.
>>
>>      "I am working on developing an enterprise wide conceptual
>>  data model. I
>>     wanted to know how developing a taxonomy is useful for the
>>  data model.?
>>
>>  My background is microbiology and library science. I have
>>  only a vague idea
>>  about what an "enterprise wide conceptional model" is.
>
>MW: A conceptual data model will be an entity-relationship
>model consisting of entity types (classes) and relationships
>between the entity types. The relationships may have cardinality
>constraints. You also find subtype/supertype relationships
>between entity types. They are traditionally presented in
>diagramatic form, but can also be in a lexical form.
>
>Data models are usually developed as the basis for a database
>design, where entity types potentially become database tables.
>
>The link below should give you more information about data
>models and how to develop them.
>
>http://www.matthew-west.org.uk/Documents/princ03.pdf
>
>>  Since
>>  the membership
>>  of Ontolog forum seems to be very technical, I would like to
>>  know how you
>>  would answer this question? Here is some more background for
>>  the question:
>>
>>     Here is what I have to do. I have several budgets such as
>>  Finance, Human
>>     Asset etc. Now I am categorizing my Finance subject area
>>  into following
>>     :
>>
>>     Fund Management
>>     Payment Management
>>     Cost Accounting
>>     Receipt Management
>>     Asset Management
>>     Now I can sub categorize any of these category , i.e. Fund
>>  Management is
>>     categorized into following:
>>     Account Structure
>>     Budget
>>     GL Account
>>     Audit Finding
>>     Available Funding    etc
>>     Now you can see, I am developing this category, so where
>>  is Taxonomy
>>     here?  Here is description of my Fund Management :
>
>MW: Well many of these things are activities. In principle, they
>could become entity types, but looking at these, they are
>rather general groupings of things. The critical question would
>be "Does the organization want to hold information about an
>instance of a (say) fund management activity?" If the answer
>is "yes" then you make it an entity type. If the answer is
>"no" then you need to decompose fund management into its component
>activities and ask what you want to hold about that.
>
>MW: GL Account is a good example of something where you will
>want to hold the instances. You will want to hold the names
>of the GL accounts and know which transactions have been posted
>to them. So GL account would be an entity type.
>>
>>     "Funds Management activities capture the highest level of
>>  financial data
>>     and summarizes and maintains account balances by fund
>>  structure. Funds
>>     Management is also the primary tool for carrying out the Agency
>>     responsibility of establishing a system and maintaining
>>  official records
>>     for ensuring that it does not obligate or disburse funds
>>  in excess of
>>     those appropriated and/or authorized. Information captures
>>  processes
>>     involved with Funds Management include budget preparation, budget
>>     formulation, funds allocation, budget execution and funds control."
>>
>>
>>     Does this description mean to Taxonomy?
>
>MW: Not much. This is an activity description. You need to pick out
>the "things of interest to the business" from it (and more probably
>from the definitions of its component activities). They are your
>candidate entity types.
>
>MW: Taxonomy is split into two parts when you are dealing with
>data models. The first part is in the subtype/supertype relationships
>between entity types. So you might have in a data model entity types
>
>Organization
>   Company
>   Department
>   Public Administration
>   Charity
>
>Where the latter would be subtypes of the former shown in the data
>model.
>
>You might also have an entity type called Product Type. Instances
>of this entity type would themselves be classes, and will probably
>have a subtype/supertype hierarchy. These instances will be held
>as data records in a data base, and as such are usually referred
>to as Reference Data (or sometimes Master Data). Reference data
>is stuff that changes slowly in the business and is usually
>contrasted with Transaction Data which is about things like sales.
>Most reference data are classes, and so reasonably the subject of
>taxonomy, but some are individuals, like organizations and countries.
>
>So after all that explanation, the answer to give to your data
>modelling colleagues is that taxonomy is found in the subtype/
>supertype structure of the conceptual data model and in the
>hierarchies of the Reference Data held in the databases it
>defines.
>
>>  __________________________________________
>>  Kathleen Ellis - Assistant Senior Information Specialist
>>  LINK -- Lilly Information and Knowledge
>>  Eli Lilly and Company, Indianapolis, IN 46285
>  > Email: kaellis@xxxxxxxxx Voice: 317-277-4071 Fax:317-276-4418
>>
>>  _________________________________________________________________
>>  Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
>>  Subscribe/Unsubscribe/Config:
>>  http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
>>  Shared Files: http://ontolog.cim3.net/file/
>>  Community Wiki: http://ontolog.cim3.net/wiki/
>>  To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>> 
>>
>>
>
>_________________________________________________________________
>Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
>Subscribe/Unsubscribe/Config: 
>http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
>Shared Files: http://ontolog.cim3.net/file/
>Community Wiki: http://ontolog.cim3.net/wiki/
>To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (012)


-- 
Rex Brooks
President, CEO
Starbourne Communications Design
GeoAddress: 1361-A Addison
Berkeley, CA 94702
Tel: 510-849-2309
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Unsubscribe/Config: 
http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (013)

<Prev in Thread] Current Thread [Next in Thread>