RE: Re[2]: SUO: Language issues (formerly "Maintenance - related issues" )
Dear Leonid,
. First I must apologise for taking so long to reply.
I have been in the middle of some exchanges where my perspective was not
being well understood. Hopefully that situation has settled down a bit.
. Comments interspersed, prefaced "GH2> ".
Cheers Graham Horn
Australian Institute of Health and Welfare
===============================================
Phone: 02.6244.1094
Fax: 02.6244.1199
Email: Graham.Horn@aihw.gov.au <mailto:graham.horn@aihw.gov.au>
-----Original Message-----
From: Leonid Ototsky [mailto:leo@mmk.ru] <mailto:[mailto:leo@mmk.ru]>
Sent: Monday, 23 October 2000 11:07
To: Horn, Graham
Cc: 'Douglas McDavid/Boulder/IBM'; standard-upper-ontology@ieee.org
<mailto:standard-upper-ontology@ieee.org>
Subject: Re[2]: SUO: Language issues (formerly "Maintenance - related
issues" )
<< File: vcard.vcf >>
Dear Graham, Doug, John and other SUOs,
. Suppose your discussion around a "controlled English" has some
"terminological polemic" features. If a "natural language" and a "pure
logic" formalism to place on the opposite sides of a scale then of
course a "rational point" must be between them. But where ? Suppose you
are moving to it from the different sides. Although my views are "inverse"
from regarding " a man-on-the-street " as a client but suppose they are
not too differ from the Graham's ones. Such examples as "button" ,
"mouse" etc. are "class names" and of course such ones must be from a
"natural" language terminology (if it is an "external" representation
and may be a "conceptual" (one?) too ( but not an internal one !)) .
GH2> My perspective, and I think John Sowa has a similar view, is that a
properly structured controlled English is logically identical to any other
properly structured logic symbology. The difference is that while the latter
will tend to be easy to read by a student of logic, familiar with logical
symbologies, the former will be easier for ordinary people to read, since it
will comprise basically a subset of(the formal subset of) natural English.
GH2> This does not mean that they will necessarily even logically express
the same concepts in the same ways, or that what is succinct in one form
will necessarily be so in the other.
. But the expressions by my opinion must be much more close to the
"pure logic" end of the scale.
GH2> I would disagree with this statement. A properly structured
controlled English will be logically precise and unambiguous.
. Moreover I suppose that now "logic culture" and "logic scholarship"
must be learned from a childhood.
GH2> I believe people can learn concepts and principles at any age.
However, I agree that, if one has an understanding of principles of logic
(and not necessarily just or even the classically quoted ones), then one has
a better ability to comprehend the implications and/or potentials of a
situation or item. I also believe it helps to have a good grounding in the
logical structures underlying one's own communication language. This is why
I believe it is very valuable to have a sound grounding in formal
Graeco-Latin based grammar, or the relevant equivalent for one's own
language, preferably from one's school years.
. In our practice we used the formal class definitions in a form of
logic expressions (which we named as "class determinator" (following the
biology practice)) as a "middleware" for contacts with our clients from
Sales and Financial departments of our MMK and they liked such a rigour.
In our work we followed the "Duality Principle in a Classification Theory"
which was formulated in the middle 70th in Russia and which differed an
external "taxonomical" view founded on the set theory and "internal"
"meronomy" founded on the logic . (I?) Suppose the IIDEAS project community
(under the ISO TC184/SC4/WG10) in their EXIST means something the same
defining it as "EXpression of Information based on logic and Set Theory" .
GH2> I do not know the word "meronomy", and do not understand what
concept you are trying to communicate here.
. Here I actually interpreted one state of my paper "IT Strategy for
the 21st Century" published in Russian in #3,2000 of the "Open Systems
Journal" ( www.osp.ru/os/2000/03/076.htm
<http://www.osp.ru/os/2000/03/076.htm> ) (a source copy is on my website) .
Another one I gave in a previous letter when wrote about an "engineering
semiotics" .
. (I?) Suppose a further interpretation will be helpful for the SUO
too.
GH2> English would be the best language for me. I'd need to refer to
dictionaries, to go back to French or German.
Best regards,
Leonid
GH2> PS: the "GH" in "GH> " are my initials. The normal
approach is to tag lines with the initials of the author, so that the text
below would normally be tagged "DM>", "DMD>", or something similar.
Friday, October 20, 2000, 1:06:18 PM, you wrote:
HG> Douglas,
HG> . Comments interspersed, prefaced "GH> ".
HG> Cheers Graham Horn
HG> Australian Institute of Health and Welfare
HG> ===============================================
HG> Phone: 02.6244.1094
HG> Fax: 02.6244.1199
HG> Email: Graham.Horn@aihw.gov.au <mailto:Graham.Horn@aihw.gov.au>
<mailto:graham.horn@aihw.gov.au <mailto:graham.horn@aihw.gov.au> >
HG> -----Original Message-----
HG> From: Douglas McDavid/Boulder/IBM [mailto:mcdavid@us.ibm.com]
<mailto:[mailto:mcdavid@us.ibm.com]>
HG> Sent: Friday, October 20, 2000 2:14 AM
HG> To: Horn, Graham
HG> Cc: standard-upper-ontology@ieee.org
<mailto:standard-upper-ontology@ieee.org>
HG> Subject: RE: SUO: Maintenance - related issues
HG> Graham-
HG> Your explanation is helpful, and I appreciate that you took the trouble
to HG> provide it. I'm in a mode of thinking about your proposition, rather
than a HG> mode of having taken a position on it, one way or another. I
guess I should HG> state, however, that it never occurred to me that the
work of ontology HG> building would ever have a man-on-the-street clientele.
GH>> Yes, It was the change of perspective to our client focus I HG>
mentioned previously, with our "Knowledgebase", that prompted me most to HG>
pick up this perspective. Our modern society means far more education is HG>
around, and there is far more opportunity for the bulk of society to HG>
participate in information areas. I feel placing things in readable English
HG> facilitates that enormously.
HG> The community I have had in my mind has primarily been domain analysts
and HG> software developers. Having said that, I think back to my roots,
having HG> spent the first part of my career as a librarian. In my mind the
librarians HG> in the libraries of record (like the Library of Congress)
have the role of HG> ontologists. They create and extend ontological works
such as the Dewey HG> Decimal System, the Library of Congress Classification
scheme, the Library HG> of Congress Subject Headings, and the Sears Subject
Headings. The equivalent HG> of my analysts and developers would be the
librarians in public, corporate HG> and academic institutions who make use
of those ontological products to HG> classify actual works, and use the
classifications in information retrieval. HG> Of course a goal of all of
this work is to make the classifications and HG> their application
semantically accessible to non-librarians, so that as much HG> information
retrieval as possible can be accomplished on a self-service HG> basis. So,
clearly, your concern for the man-on-the-street user has HG> precedence in
my previous life's experience.
GH>> Yes, you are right on the wavelength that inspires me. As an HG>
engineer by training, I see your background as being from one of the great
HG> traditional information management areas, and the Dewey Decimal System
as HG> one of the important steps in addressing issues of general purpose
HG> information classification.
GH>> Furthermore, I see the leverage for direct "human machine HG>
interaction" as providing a whole further dimension of potential. The HG>
following is just a quick example off the top of my head, but I think it HG>
gets across the sort of potentials I see. Consider how much easier it would
HG> be to develop intelligent interactive software able to respond with:
HG> "I didn't understand what you meant by 'I need a licence for driving my
HG> unregistered car to the service station for repairs'. Do you need:
HG> 1. a drivers licence,
HG> 2. a temporary permit for your unregistered car,
HG> 3. both, or
HG> 4. something else (if so please restate your question differently
in
HG> the space below)?"
HG> Still thinking, and thanks for the explanatory work on my behalf.
HG> Doug McDavid
HG> Certified Executive Consultant
HG> Business Innovation Services - IBM, US
HG> Member of IBM Academy of Technology
HG> mcdavid@us.ibm.com <mailto:mcdavid@us.ibm.com> -- 916-549-4600
HG> "Horn, Graham" <graham.horn@aihw.gov.au <mailto:graham.horn@aihw.gov.au>
> on 10/17/2000 09:09:45 PM
HG> To: Douglas McDavid/Boulder/IBM@IBMUS
<mailto:McDavid/Boulder/IBM@IBMUS>
HG> cc: standard-upper-ontology@ieee.org
<mailto:standard-upper-ontology@ieee.org>
HG> Subject: RE: SUO: Maintenance - related issues
HG> Douglas,
HG> . Please bear in mind that, since I am a proponent of following
HG> ordinary English as closely as possible consistent with the rigour
HG> requirements of an ontology, that there will be NO rocket science in my
HG> responses. Generally I will just be applying home-spun common sense
aimed at
HG> appealing to the thinking of lay people with a command of the
traditional
HG> English. I say traditional, because, in the first instance I feel the
HG> structures developed, coded and refined over millennia provide a
significant
HG> resource at no cost.
HG> . That said, I had some thoughts on this very question last
night,
HG> and of course this can be used regardless of how closely the SUO is
based on
HG> English structures. Basically it came down to recalling that, * while
HG> adverbs modify verbs or adjectives, * adjectives qualify nouns.
HG> . Hence we can get some very clumsy labels, like:
HG> * "button" could become "button switch";
HG> * "mouse" could become "computer display remote position indicator
and
HG> actuator";
HG> * "head" could become "coded magnetic medium sensor",
HG> * "monitor" could become "video displayer";
HG> * "ram" could become "male sheep", to distinguish from
HG> * the device for pushing or battering, or
HG> * computer "RAM";
HG> * etc.
HG> . However, I think entities with highly structured technical
HG> specifications become too complicated for such handling (eg. consider
"Swiss
HG> bank", and "building society" for comparison), and we either require
another
HG> paradigm to get any convenient label. To accommodate this, we can define
HG> word groups, hence:
HG> * "button" could become the defined term "button switch";
HG> * "mouse" could become the defined term "computer mouse";
HG> * "head" could become the defined term "disk head", "tape head",
etc.,
HG> * "monitor" could become the defined term "video monitor";
HG> * "bank" could become the defined term "financial bank", along with
HG> defined variants including "UK bank", Swiss bank", and more broadly,
HG> "building society", "credit union", etc.
HG> . Alternatively, we can exploit the use of:
HG> * acronyms, and gain maximum leverage by respecting the tradition of
only
HG> capitalising first letters, eg.: RaDAR from Ra(dio) D(irection) A(nd)
HG> R(anging); and
HG> * made up words, possibly loosely adapted from others, eg.: Sonar,
HG> adapted from RaDAR;
HG> * slight spelling corruptions, such as: "disk", derived from
"computer
HG> disc"; but
HG> * more commonly done by advertisers, etc, as I said before, eg.
"birko"
HG> and "ramset", based on the trademarks; or
HG> * peoples' names, such as "biro", named after its invertor, and
HG> * borrowings from other languages, eg: soufflé.
HG> . Of course, we can also mandate recognised, but once again
HG> closely defined (possibly with slightly modified definitions):
HG> * prefixes, such as:
HG> * "ante-", meaning before,
HG> * "anti-" meaning opposing,
HG> * "dis-", meaning wrongly,
HG> * "non-", meaning not,
HG> * "pro-", meaning promoting the interests of,
HG> * "super-", meaning of greater magnitude or height than (meaning
HG> depends on whether meaning of word stem relates to height or not - note
HG> that
HG> we can implement more complex rules of this type, so long as they are
HG> consistent and precise),
HG> * "meta-", meaning beyond; and
HG> * suffixes, such as:
HG> * "-able", meaning able to have done to it whatever the word stem
HG> indicates,
HG> * "-ate", meaning to have done to it whatever the word stem
indicates,
HG> * "-er", indicating an object that performs whatever the word stem
HG> indicates,
HG> * "-or", indicating a person who performs whatever the word stem
HG> indicates.
HG> . The above is far from comprehensive, but I think I have given
HG> enough to demonstrate there are many ordinary possibilities to achieve
our HG> objectives.
HG> . As a matter of relevance, the point was also made on BBC radio
HG> last night, that one can already exploit the sort of approach I am
HG> suggesting, when one wishes to use the translation software available on
the
HG> Internet. The observation was made, very much in line with my proposal,
that
HG> many people learn to use an unambiguous subset of English in order to
get
HG> their messages clearly translated into other languages. It would be a
HG> relatively small step from there to achieving an ontology that is
rigorous,
HG> easily human and machine interpretable, and machine executable.
HG> Cheers Graham Horn
HG> Australian Institute of Health and Welfare
HG> ===============================================
HG> Phone: 02.6244.1094
HG> Fax: 02.6244.1199
HG> Email: Graham.Horn@aihw.gov.au <mailto:Graham.Horn@aihw.gov.au>
<mailto:graham.horn@aihw.gov.au <mailto:graham.horn@aihw.gov.au> >
HG> -----Original Message-----
HG> From: Douglas McDavid/Boulder/IBM [mailto:mcdavid@us.ibm.com]
<mailto:[mailto:mcdavid@us.ibm.com]>
HG> Sent: Tuesday, October 17, 2000 11:24 PM
HG> To: Horn, Graham
HG> Cc: standard-upper-ontology@ieee.org
<mailto:standard-upper-ontology@ieee.org>
HG> Subject: RE: SUO: Maintenance - related issues
HG> Graham-
HG> I am interested to know more about the way you envision handling this
issue:
HG> * there would be some new words fabricated to cover useful concepts
where
HG> these are either not currently covered by a single word, or, more
commonly,
HG> where the current word used is a metaphorical application of an existing
HG> word with another main meaning(eg. "button", "mouse", "head", "monitor",
HG> etc.); while
HG> Can you give us examples of fabricated words that would cover the newer
HG> (presumably less frequently used) senses of these words? I would also
HG> appreciate an example from a more fundamental example of polysemy, such
as HG> "bank".
HG> Doug McDavid
HG> Certified Executive Consultant
HG> Business Innovation Services - IBM, US
HG> Member of IBM Academy of Technology
HG> mcdavid@us.ibm.com <mailto:mcdavid@us.ibm.com> -- 916-549-4600
Best regards,
Leonid
mailto:leo@mmk.ru <mailto:leo@mmk.ru>
- Prev by Date:
SUO: STEP as an Ontology -- RE: RE: RE: Re: Re: A,A,E,I,M,M,M,P,P,S.......
- Next by Date:
SUO: Re: Abstraction, Analogy, Example, Icon, Metaphor, Model, Morphism, Paradigm, Prototype, Simulation
- Prev by thread:
SUO: RE: Full Text & WebSite Translators
- Next by thread:
SUO: Full Text & WebSite Translators
- Index(es):