Hi Brad --
You make some excellent points about human motivation making things work.
A quibble, though. You wrote...
note that "English" is NOT a solution if coalition
interoperability is of any interest at all (as french speakers will no doubt
attest). And if english might do, wouldn't <Java, or Haskel, or ...> work
better, since at least those languages are executable?
....
The knowledge to make systems interoperate resides in the heads
of the domain experts who built them.
Yes, a mixture of natural languages in a project is a problem. I
believe international air traffic control solves this by standardizing
on English.
Once you do this, it's technically feasible to make open vocabulary
English executable, without painstaking dictionary and grammar
construction, and with strict semantics [1].
(There's also an underlying formal semantics [3].)
Thus, the knowledge needed to make systems interoperate can now reside in the heads of domain experts and in executable English.
So, English concepts can be typed into a browser, and immediately "run" to see what they do.
This may sound too good to be true, and there is a trade off. I
hope you may have time to view, edit and run the online examples [2] to
see for yourself. You are cordially invited to write and run your
own examples too.
As befits any kind of Wiki, shared use is free.
-- Adrian
PS This kind of lightweight executable English technology also
works in French, German and so on. But that does not solve the
natural language interop problem that you point out.
[1] ww.reengineeringllc.com/A_Wiki_for_Business_Rules_in_Open_Vocabulary_Executable_English.pdf
[2] Internet Business Logic Executable open vocabulary English
Online at
www.reengineeringllc.com
Shared use is free
[3] Backchain Iteration: Towards a Practical Inference Method that is Simple
Enough to be Proved Terminating, Sound and Complete. Journal of Automated Reasoning, 11:1-22
Adrian Walker
Reengineering
Phone: USA 860 830 2085
On 11/13/06, Brad Cox, Ph.D. <bcox@xxxxxxxxxxxxxxx> wrote:
As promised, these are a few reflections on several subjects I briefly
mentioned in the teleconference.
The first is the implication, intended or otherwise, that we're expecting
technologies to solve a problem that human beings have struggled with for ages
and never mastered, translation of meaning across domains (be they network
domains or language/cultural boundaries).
The biggest problem with this view is a truism: tools (technologies) never
accomplish anything without human effort to wield them. More about the human
side of this later. But note that "English" is NOT a solution if coalition
interoperability is of any interest at all (as french speakers will no doubt
attest). And if english might do, wouldn't <Java, or Haskel, or ...> work
better, since at least those languages are executable? And even with english
(or esperanto), wouldn't an upper ontology wind up as incomprehensible
legalese that only ontology lawyers could produce or understand, thus locking
out domain experts who generally speak the specialized language of their domain?
Another is that cross-domain inteoperability is obviously impossible if that
means some push-button widget ("technology") that automagically translates
anything to everything, and better yet if it costs nothing to acquire.
To shift from semantic quibbling to solutions, it seems to me the core of this
problem is human, and that the search for technical solutions is evading the
core problem. The knowledge to make systems interoperate resides in the heads
of the domain experts who built them. This implies less expectations on
technology as the solution and much more on incentivizing them to make their
knowledge available in forms that others can use. Whether they do so in
english, or french, or java, or haskel should be a pair-wise decision between
provider and consumer, not dictated by central planning.
As for what such an incentive mechanism might look like, that's a matter of
collaborative design by groups such as this. I do have definite ideas on
approaches that might work and a prototype solution at
http://virtualschool.edu/mybank. That system, however, measures incentives in
dollars, which clearly won't do for government employees (contractors are
another matter).
But consider this as a plausible alternative for government workers... OMB's
PRM (performance reference model) tracks performance down to the department,
agency, project, and potentially even the individual worker level. A SOA
infrastructure should be able to collect response time measurments which could
be used as the basis for a fine grained metric of individual performance. This
could be rolled that up into coarse-grained group performance, potentially
even to the PRM level. Might that suffice to incentivize domain specialists to
put their knowledge into forms that others could use? I don't know. I only
know that its never been tried.
Bottom line, technology alone will not solve this problem. I'm more confident
that human beings could, if we figure out how to incentivize them to
cooperate. How to do that is an open question that I believe deserves more
consideration that it generally receives in technical circles.
--
Work: Brad Cox, Ph.D; Binary Group; Mail
bcox@xxxxxxxxxxxxxxx
Home: 703 361 4751; Chat brdjcx@aim; Web http://virtualschool.edu
---------- Original Message -----------
From: "Adrian Walker" <
adriandwalker@xxxxxxxxx>
To: James.Schoening@xxxxxxxxxxx, cuo-wg@xxxxxxxxxxxxxx
Sent: Mon, 13 Nov 2006 11:26:25 -0500
Subject: Re: [cuo-wg] Draft Agenda for 13 Nov CDSI Conference Call
> Jim --
>
> Thanks for the pointer to the Markoff article.
>
> This is a request please to add to the list of candidate technologies:
>
> "Internet Business Logic -- A kind of Wiki for executable English
> ontologies --
> www.reengineeringllc.com -- shared use is free"
>
> (There seem to be problems with directly editing the
> www.visualknowledge.com/vkgw/vk page at the moment -- Application Error)
>
> I look forward to the conference call this morning.
>
>
Best, -- Adrian
>
> Adrian Walker
> Reengineering
> Phone: USA 860 830 2085
>
> On 11/7/06, Schoening, James R C-E LCMC CIO/G6 <James.Schoening@xxxxxxxxxxx
>
> wrote:
> >
> > CDSI Subscribers,
> >
> > A draft agenda for the 13 Nov (1130-1230) kick-off conference call
> > for the Cross Domain Semantic Interoperability WG is posted at:
> >
> > *
> >
http://www.visualknowledge.com/vkgw/vk?ObjectKey=A23781S1028&ObjectMessage=renderForm:&formContext=A72456S3496911
> >
*<http://www.visualknowledge.com/vkgw/vk?ObjectKey=A23781S1028&ObjectMessage=renderForm:&formContext=A72456S3496911
>
> >
>
> Please send any
suggested changes, or new topics if you'd like to
> > present anything.
> >
> > James R. Schoening
> > U.S. C-E LCMC CIO/G6 Office
> > Voice: DSN 992-5812 or (732) 532-5812
> > Fax: DSN 992-7551 or (732) 532-7551
> > Email: James.Schoening@xxxxxxxxxxx
> >
> > _________________________________________________________________
> > Message Archives: http://colab.cim3.net/forum/cuo-wg/
> > Subscribe/Unsubscribe/Config:
> >
http://colab.cim3.net/mailman/listinfo/cuo-wg/
> > To Post: mailto:cuo-wg@xxxxxxxxxxxxxx
> > Community Portal: http://colab.cim3.net/
> > Shared Files: http://colab.cim3.net/file/work/SICoP/cuo-wg/
> > Community Wiki:
> >
http://colab.cim3.net/cgi-bin/wiki.pl?SICoP/CommonUpperOntologyWG
> >
> >
> >
> >
------- End of Original Message -------
_________________________________________________________________
Message Archives: http://colab.cim3.net/forum/cuo-wg/
Subscribe/Unsubscribe/Config: http://colab.cim3.net/mailman/listinfo/cuo-wg/
To Post: mailto:cuo-wg@xxxxxxxxxxxxxx
Community Portal: http://colab.cim3.net/
Shared Files:
http://colab.cim3.net/file/work/SICoP/cuo-wg/
Community Wiki: http://colab.cim3.net/cgi-bin/wiki.pl?SICoP/CommonUpperOntologyWG
_________________________________________________________________
Message Archives: http://colab.cim3.net/forum/cuo-wg/
Subscribe/Unsubscribe/Config: http://colab.cim3.net/mailman/listinfo/cuo-wg/
To Post: mailto:cuo-wg@xxxxxxxxxxxxxx
Community Portal: http://colab.cim3.net/
Shared Files: http://colab.cim3.net/file/work/SICoP/cuo-wg/
Community Wiki:
http://colab.cim3.net/cgi-bin/wiki.pl?SICoP/CommonUpperOntologyWG (01)
|