Re: standardization of semantics
rmb@santafe.edu
Date: Mon, 15 Aug 94 19:36:36 MDT
From: rmb@santafe.edu
Message-id: <9408160136.AA21903@moqui.santafe.edu>
To: srkb@cs.umbc.edu, ginsberg@t.uoregon.edu
Subject: Re: standardization of semantics
Cc: jksharp@sandia.gov
Sender: owner-srkb@cs.umbc.edu
Precedence: bulk
In reply to Matt Ginsberg's question:
Why is there this rush to standardize on things that aren't even in
actual use?
Many standards (information processing or otherwise) get started by
ratifying an existing practice, but others are established when people
know they have to reach agreement to make things work together (some
kinds of physical connectors, for example).
All kinds of standards are being developed to exchange data (product
data, business data, geographic data, you name it...). Few of these
standards have an existing codification that everybody would agree to
use as is. Actual use is not the standard that carries the greatest
weight when people decide to share information.
The ISO and ANSI X3 conceptual schema committees are attempting to help
people share information using semantic agreements similar to the
linguistic conventions they already follow when they communicate without
computers. First-order logic is a subset of the standards we follow
when we attempt to reach an unambiguous understanding. I don't see any
harm in standardizing a few initial symbolic forms so that a wider
community can begin making more deliberate use of its own semantics.
Much of the earlier ANSI X3 work was very insistent that its main goal
was not to standardize any particular notation for encoding of abstract
semantics, but only to establish a starter set of one or more notations
expressive enough to capture semantics of other commonly used languages
and exchange formats. The standards process is open to any proposals
for languages that have this power, which they can demonstrate by
showing the ability to define the semantics (or lack thereof) of any of
their rivals.
All standards committees in the United States are legally constrained
to be remain open to anyone who wishes to participate. If you'd like
to attend the Seattle meeting, I'd encourage you to do so.
Roger Burkhart