RE: [xml-dev] Defining an XML vocabulary: specify syntax, semantics, and

["Costello, Roger L." <costello@mitre.org>]

Bruce Cox wrote:

> It seems to me that whether or not to specify behavior 
> with vocabulary would depend on the context.  

> Roger's book markup is a good example, because it appears 
> to me essentially devoid of behavior.  That can work well 
> for books, because the instances would be used in an industry 
> where the behavior is very well established and documented, 
> even standardized, in other places.  The markup does not need 
> to encode the behavior, because almost anyone using it would 
> already know what to do.  

> In other cases, if the behavior is not encoded with the data, 
> it could happen that no one would ever understand what it was 
> supposed to represent or what to do with it (for example, the 
> radio messages in Carl Sagan's "Contact").

Excellent Bruce!  I particularly appreciate your example of Carl
Sagan's "Contact" movie.  That hit home for me. 

For those who have not seen the movie: Earth received radio signals
from a highly advanced civilization from a distant galaxy.  Embedded
within the signals are both data and instructions for using the data
(it described how to build a device to transport humans at
faster-than-light, as I recall).

Without a specification of the behavior, the "Contact" data would have
been unfathomable to the Earthlings.  I think this is an excellent
example of the need to specify:

   - syntax
   - semantics
   - behavior (instructions)


Thanks Bruce!

/Roger

-----Original Message-----
From: Cox, Bruce [mailto:Bruce.Cox@USPTO.GOV] 
Sent: Tuesday, April 15, 2008 7:11 PM
To: rob@koberg.com; Len
Cc: bryan rasmussen; noah_mendelsohn@us.ibm.com; Costello, Roger L.;
Fraser Goffin; xml-dev@lists.xml.org
Subject: RE: [xml-dev] Defining an XML vocabulary: specify syntax,
semanti cs, and BEHAVIOR?

Down to lower levels, and out to higher levels as well.  Every XML
vocabulary exists in a social and technological context that provides
the motivation and justification for its existence.  When we were
devising the DTD's for patent documents, we soon realized that, while
we
could reach consensus on the elements, there were long-standing
cultural
and legal differences in how the content would be processed in
different
patent offices.  When needed to correctly structure the DTD, behavior
was taken from the rules of the Patent Cooperation Treaty, to which all
the offices involved were party.  In one sense, the PCT "standardized"
the larger, abstract "machinery" that would ultimately process the
content of the DTD instances.  Still, we hoped for interoperability,
that is, we hoped that the same patent application could be processed
successfully by any of the offices, as well as the PCT, without
modification.  We soon learned that a common vocabulary, while
necessary, is not sufficient for interoperability.  To achieve
interoperability, there are many other layers of the patent business
that require harmonization, none of them in the domain of IT.  That
part
is in progress, but vastly more slowly than what has been accomplished
on the IT side.  The Patent Cooperation Treaty will go a long way in
that direction, but it might be a decade before it enters into force.

It seems to me that whether or not to specify behavior with vocabulary
would depend on the context.  Roger's book markup is a good example,
because it appears to me essentially devoid of behavior.  That can work
well for books, because the instances would be used in an industry
where
the behavior is very well established and documented, even
standardized,
in other places.  The markup does not need to encode the behavior,
because almost anyone using it would already know what to do.  In other
cases, if the behavior is not encoded with the data, it could happen
that no one would ever understand what it was supposed to represent or
what to do with it (for example, the radio messages in Carl Sagan's
"Contact").  Is it correct that the extemporaneous aggregation of
services over the internet face this problem?  WSDL's and other pieces
standardize representation of the context required for machine
processing, I believe, but still, it's the larger social envelope that
motivates the investment and drives the desired behavior.  

The degree to which the various layers of the overall business context
have been standardized (as well as the representation thereof in XML)
seems likely to be proportional to the potential benefit to the
investors.  Google is a fine example of a business growing from a
common
starting point into new territories that others have neglected, at
least
in part by extending structured behavior and vocabularies
(http://www.news.com/8301-10784_3-9917421-7.html?tag=nl.e433) .

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

The opinions expressed in this message are solely those of the author
and must not be construed as the official views of the U.S. Patent &
Trademark Office.

-----Original Message-----
From: Robert Koberg [mailto:rob@koberg.com] 
Sent: Friday, April 11, 2008 10:17 PM
To: Len
Cc: 'bryan rasmussen'; noah_mendelsohn@us.ibm.com; 'Costello, Roger
L.';
'Fraser Goffin'; xml-dev@lists.xml.org
Subject: RE: [xml-dev] Defining an XML vocabulary: specify syntax,
semanti cs, and BEHAVIOR?

Surely you can take this past the processor down to lower lever code,
then down to the machine, then to the organic on down to the sub atomic
and beyond, right?

You are drawing Zeno's arrow, but only letting it fly half way.

-Rob


On Fri, 2008-04-11 at 20:26 -0500, Len wrote:
> The rules of the XML processor are parsing rules.
> 
> The rules of the language encoded in XML aren't.
> 
> It's only oxymoronic if the language only conveys the rules of XML.
> 
> That isn't all that useful. ;-)
> 
> len
> 
> 
> From: bryan rasmussen [mailto:rasmussen.bryan@gmail.com] 
> 
> If I have an application that takes all XML in, I look for first a
> specific format handler or then fall back to the default handler for
> unrecognized XML in my application. The default handler is the XML
> format handler and the format is XML.
> 
> If I have an application that handles two formats and each of them
has
> a handler then if I get a format X that I don't recognize and I dump
> it, then the format isn't handled and the format isn't from the
> viewpoint of my application a format.
> 
> If I think of format handler in these ways then obviously there can't
> be a format without a format handler. I'm not sure if that was what
> Len meant though because it renders the statement somewhat
oxymoronic.
> 
> 
> 
>
_______________________________________________________________________
> 
> 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@lists.xml.org
> subscribe: xml-dev-subscribe@lists.xml.org
> List archive: http://lists.xml.org/archives/xml-dev/
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
>