Hi Brad --
Thanks for your comments.
As I understand it, you raise two main issues.
1. Is it really English?
2. Propriatryness
I'll try to address these in order.
1. Is it really English?
You say that Executable English is "based on english only in the sense that Java, Objective-C and Cobol are" and "this is as much english as cobol was, albeit with less procedural, more rule-oriented semantics
"
Well, I guess a key question is whether it's useful for current purposes, rather than whether or not it's like Cobol.
(By the way, changing the order of the rules does not change what is
derivable -- a rule just means what it says. This is a different
rule semantics from any other I know about, and arguably better for
current purposes.)
So, getting back to "like Cobol"..... I don't see anyone
proposing to do ontology-based interoperability in Cobol. The
language of choice so far seems to be some flavor of FOL.
As mentioned, the Internet Business Logic system maps EE automatically
to and from an internal FOL-like notation. The hoped for
advantage of writing in EE rather than in FOL is that person B will
have a better chance of actually understanding what person A wrote.
(It's famously difficult to stare at a complex FOL _expression_ and make
an accurate English gloss of what it's supposed to mean). One
could think of EE as just a good way documenting FOL, and making sure
that the documentation and the FOL never get out of step. It also
provides support for explanations for non-technical users.
A curious property of the approach is that EE is somewhat agnostic
about what you write. So, if you actually need to write more or
less directly in executable FOL, you can do so. For example, in
www.reengineeringllc.com/demo_agents/RelBioOntDefn3.agent
there are rules such as
some-C and some-C1 are two different Non-process classes with instances
not : (E c,t) [ that-C c t and not (E c1) [ that-C1 c1 t and c part_of c1 at t ] ]
------------------------------------------------------------------------------------------------------
(A c,t) [ that-C c t => (E c1) [ that-C1 c1 t and c part_of c1 at t ] ]
You are welcome to run the agent. Of course, the resulting
explanations will be mainly of interest to technical folks who know how
to read this sort of stuff.
2. Propriatryness
You wrote "People are sensitive to
proprietary languages/compilers because they're locked to that language
and that vendor by the programs they write."
Agreed, propriatryness is a problem for many folks. It's
hopefully mitigated by having the Internet Business Logic system
available free for shared use on the web, as opposed to shrink wrap or
fee-for-download. (As mentioned, I have suggested that
the language could be of interest for a W3C standard
**. If that happened, there would then be different
implementations of the standard, and presumably some of them would be
proprietary, with competition on inference speed and other features.)
"wiki pages are just text that can be exported if need be and used anywhere"
An EE file can be copy-pasted and used anywhere -- used in the sense of
documentation that has actually been tested. It's a fairly easy
task to translate the file manually into FOL. (The reverse
translation can be quite difficult.)
I hope the above helps with the issues you rightly raise. To what
extent do you think that open sourcing the Internet Business Logic
system would solve the proprietaryness problem?
Cheers, -- Adrian
** www.w3.org/2004/12/rules-ws/paper/19
Adrian Walker
Reengineering
Phone: USA 860 830 2085
On 11/17/06, Brad Cox, Ph.D. <bcox@xxxxxxxxxxxxxxx> wrote:
Adrian, please help me understand. I don't see why this isn't a proprietary
language with web-accessible demo compiler that is based on english only in
the sense that Java, Objective-C and Cobol are.
People are sensitive to proprietary languages/compilers because they're locked
to that language and that vendor by the programs they write. The wiki issue
isn't germane because wiki pages are just text that can be exported if need be
and used anywhere, which isn't the case with source code. There might be an
issue if we were using its semantic capabilities to build significant
ontologies, but we're not using it that way and I know of no plans to.
I mean by all that that this is as much english as cobol was, albeit with less
procedural, more rule-oriented semantics:
person some-CID some-name is classified in driver eligibility as Young Driver
person that-CID that-name has Marital Status Single in his or her description
not : person that-CID that-name resides in a state that is exceptional for
driver premium purposes
--------------------------------------------------------------------------------------------------------------------------------
person that-CID that-name is subject to a $ 300 premium increase
This isn't to be confrontation; just trying to understand.
--
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: "common upper ontology working group" <cuo-wg@xxxxxxxxxxxxxx>, "Cory
Casanave" <cbc@xxxxxxxxxxxxxxxxxxxxxxx
>, "Schoening, James R C-E LCMC CIO/G6"
<James.Schoening@xxxxxxxxxxx>
Sent: Fri, 17 Nov 2006 14:44:15 -0500
Subject: Re: [cuo-wg] Executable English vs FOL for all domains run time
interoperability
> Hi Cory --
>
> Wow, some good questions within ten minutes of posting! (:-)
>
> I'll try to answer them one by one.
>
> >
> > Is EE...
> >
>
> Well, EE is the language, and Internet Business Logic is the thing that
> computes with it.
>
> So, is Internet Business Logic a
>
> > - Proprietary Product?
> >
>
> only in the sense that the Wiki we are using for CDSI is also a pp. That
> is, anyone on the web can write (and run) their own material by pointing a
> browser to the shared area, and the material is then open to anyone on the
> Web. As befits a Wiki, shared use is free.
>
> - Standard?
> >
>
> No, but I've suggested to W3C that it would speed up adoption of the
> Semantic Web. Please see
>
www.w3.org/2004/12/rules-ws/paper/19
>
> - Established open approach in some community?
> >
> - Emerging open approach in some community?
> >
>
> Early days yet. Maybe it will in time become an established, semi-open
> approach in this community.
>
> - Prototype?
> >
>
> It's at Beta.
>
> - Idea?
> >
>
> Implemented system, live online. The first publication of the ideas that
> led to the system was back in 1981 (!). There are over 20 papers since
> then, some of them in refereed publications, such as
> "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
>
> I suggest we be clear about the above as we submit these options and that
> > only non-proprietary approaches are eligible for consideration.
> >
>
> Hmm. Would that, in your view, also rule out the use of the current CDSI
> Wiki -- as Peter Yim has suggested?
>
> > Also, it sounds a lot like the "business semantics of business rules"
> > standard of OMG (http://www.omg.org/cgi-bin/doc?dtc/2006-08-05)-
which
> > takes a very structured English approach. Do you know what the relationship
> > is?
> >
>
> Glad you asked. There is a Business Rules and OMG SBVR Presentation --
> www.reengineeringllc.com/Business_Rules_and_OMG_SBVR_Presentation.pdf .
> (Jim -- would that presentation be of interest for a conference call?)
>
> In particular, EE differs from other approaches in that the vocabulary is
> open, and there is no external dictionary construction required, yet the
> English semantics are strict. The English syntax is also mostly open. This
> means that one can freely use things like government acronyms, jargon
> syntax, and so on. There's a trade off, of course, that one can evaluate by
> using the system.
>
> (As you may know, brittleness, and related requirements for dictionary and
> grammar maintenance appear to have kept most natural language query systems
> away from major commercial and government use.)
>
> Thanks!
> >
>
> Thanks for the good questions!
>
> Adrian Walker
> Reengineering
> Phone: USA 860 830 2085
>
> Cory Casanave
> >
> > ------------------------------
> > *From:*
cuo-wg-bounces@xxxxxxxxxxxxxx [mailto:
> > cuo-wg-bounces@xxxxxxxxxxxxxx] *On Behalf Of *Adrian Walker
> > *Sent:* Friday, November 17, 2006 1:57 PM
> > *To:* Schoening, James R C-E LCMC CIO/G6; common upper ontology working
> > group
> > *Subject:* [cuo-wg] Executable English vs FOL for all domains run
> > timeinteroperability
> >
> > Hi Jim --
> >
> > This is to try to address a little better how Executable English (EE) may
> > help to answer your requirement for 2N interoperability over all compliant
> > domains.
> >
> > The EE approach for all compliant domains would be to find a canonical,
> > pivot body of knowledge so as to get to 2N, and to represent the knowledge
> > in EE rather than FOL. (But see below for the automatic, bi-directional
> > mapping between EE and FOL).
> >
> > For example, in
> >
> > www.reengineeringllc.com/demo_agents/OntologyInterop2.agent
> >
> > a canonical set of units is chosen. Then, anyone using other units must
> > use the suggested adapters to map to the canonical units. This is 2N. The
> > agent computes the adapters that are needed.
> >
> > As another example, in
> >
> > www.reengineeringllc.com/demo_agents/SemanticResolution1.agent
> >
> > anyone wishing to do business must have an adapter (2N again) that maps
> > his or her internal terminology to a more general, canonical "upper" set of
> > terms. So, in this example, the pivot is a taxonomic hierarchy.
> >
> > EE is automatically mapped into and out of an FOL-like notation for
> > inference. But that's done inside a black box called Internet Business
> > Logic, which also does the inference, and provides the English explanations
> > of what's going on. It gets scalability by automatically generating and
> > running networked SQL "under the covers".
> >
> > So, as an analogy, EE is to FOL as Java is to Assembly Language. In this
> > analogy, the Internet Business Logic system corresponds to a Java compiler.
> >
> > Hope this makes sense. If folks have time to run some examples**,
> > comments would be much appreciated. You can write and run your own examples
> > too.
> >
>
>
Thanks, -- Adrian
> >
> > ** Just point a browser to www.reengineeringllc.com and click on Internet
> > Business Logic. The system works better with Mozilla or Firefox than with
> > IE. For IE, the "browsers" page has some suggested settings.
> >
> > Adrian Walker
> > Reengineering
> > Phone: USA 860 830 2085
> >
> >
> > _________________________________________________________________
> > 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)
|