You may be right, Peter, that there are no entirely pure abstractions
... .  Is this a good example for you?  I seem to recall that Einstein
put "tensors" to use in his theory of relativity, a mathematical
construct that had been developed much earlier, purely in the abstract,
without any apparent connection to physical reality at the time.  I'm
pretty sure there are other mathematical recreations that have been
worked out for which there is still no concrete application, other than
the entertainment of the inventor.

An interesting example, not entirely unrelated to the discussion, is
that the Dewey Decimal system of classification (DDC) attempted to
create a universal set of categories that would accommodate all
knowledge, past, present, and future.  One reason DDC has been largely
abandoned in US academic libraries is that, in reality, knowledge is not
isotropic across the categories.  The Library of Congress Classification
(LCC), on the other hand, was and is built in response to the collection
of books at hand.  There are no empty categories in LCC, while there are
many dis- or little-used categories in DDC.  When the USPTO was revising
its patent classification scheme (USPC) in the 1950's, they adopted the
LCC design philosophy, meaning that the USPC creates categories when and
only when there are inventions not accommodated by the current set.
This suggests that schemas that reflect the current reality (with just a
short view into the future?) are more likely to stay out of trouble than
schemas that attempt to predict the future, one of Roger's best
practices.

Bruce B Cox
Manager, Standards Development Division
USPTO/OCIO/SDMG
571-272-9004


-----Original Message-----
From: Peter Hunsberger [mailto:peter.hunsberger@g...] 
Sent: Monday, February 02, 2009 3:43 PM
To: Cox, Bruce
Cc: xml-dev@l...
Subject: Re: [xml-dev] RE: Keep business-process-specific data separate?

On Mon, Feb 2, 2009 at 2:07 PM, Cox, Bruce <Bruce.Cox@u...> wrote:
> Another try, after reading some entries in the OED
>
> Generic: General as opposed to specific; aspirin as opposed to Bayer;
>
> Abstract: Abstract as opposed to concrete; (a*a) + (b*b) = (c*c) "a
squared
> plus b squared equals c squared" as opposed to 3*3 + 4*4 = 5*5
>
> An abstraction might or might not be discovered by inspection of some
> instances, but an abstraction has an internal truth that is completely
> independent of whether it is ever instantiated.  However, aspirin is a
name
> for a collection of instances (with a common property) that has no
existence
> without those instances.

Sounds like a nice distinction on the surface, but just to continue to
play devils advocate for the moment; can you give me an example of an
abstraction that would make sense without some concrete instance
existing for reference purposes? I'm pretty skeptical that any such
pure, "unattached" abstraction -- which would be the ultimate
extension of your proposition -- is possible....

-- 
Peter Hunsberger



_______________________________________________________________________

XML-DEV is a publicly archived, unmoderated list hosted by OASIS
to support XML implementation and development. To minimize
spam in the archives, you must subscribe before posting.

[Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
Or unsubscribe: xml-dev-unsubscribe@l...
subscribe: xml-dev-subscribe@l...
List archive: http://lists.xml.org/archives/xml-dev/
List Guidelines: http://www.oasis-open.org/maillists/guidelines.php