>
> On 21 Jul 2015, at 18:52, John F Sowa <sowa@xxxxxxxxxxx> wrote:
>
> On 7/21/2015 12:02 PM, David Price wrote:
>> There are many domains where terms are very well-defined, specifically
>> many engineering disciplines exhibit that characteristic … if not
>> planes would be falling from the sky every day.
>
> Some critical questions: What do you mean by 'domain'? What do you
> mean by 'well-defined'? And what do you mean by 'term’? (01)
I mean there are engineering manuals that define things used in places like
Boeing and NASA and Statoil that those often define a discipline/domain very
well. There are also standard textbooks (and even models) used in many
engineering disciplines that describe how materials behave, how fluids flow,
how oil platforms are broken down functionally, etc. There are often even
standards in these areas. (02)
>
> For example, the Boeing 777 -- the first airplane that was completely
> specified by computer -- was precisely defined. But it used a large
> number of *mutually inconsistent* domains -- i.e., different and
> inconsistent approximations for different theories used for different
> components and aspects of the same airplane. (03)
Yep - there are many engineering disciplines, but they are not all mutually
inconsistent. There is also a good overlap in many of those where things are
commonly defined or where one builds on the other (e.g. finite element
analysis, shading, cable/pipe layout, interference analysis, etc. over geometry
and topology). (04)
>
> From http://www.jfsowa.com/pubs/crystal.htm
>> As these examples illustrate, physics is not a monolithic subject,
>> but a loose collection of subfields, each characterized by what it
>> ignores or approximates. The totality forms an inconsistent knowledge
>> soup, where one scientist ignores features that are the focus of
>> attention for another.
>
> David
>> My experience is seeing enterprises waste a lot of money doing
>> ontology work because they are like Wall Street-driven CEOs only
>> thinking about the short term when they should have a much longer
>> term vision.
>
> Certainly. But that "vision" is an extremely underspecified
> theory stated only in natural language. Even an attempt to make
> it "precise" in the details would a very bad idea, typically
> known as "micromanaging”. (05)
The EPIM vision is a standardised, shareable knowledge base across a variety of
disciplines and they make that precise by saying that wrt ontologies the vision
is executed using ISO 15926. If you want to call forcing a specific set of
metaphysical decisions on modelers “micromanaging” then so be it. Their
view is that the benefit outweighs the cost. (06)
>
>> The idea of everyone in an industry or enterprise using the same
>> framework/approach links up nicely with the certified accountancy
>> guidelines analogy thread that appeared in the last week or two.
>
> The words 'framework', 'approach', and 'guidelines' are a sign
> that we're talking about a broad, general, underspecified theory
> stated *only* in a natural language. Each application will have
> a different special case that is also stated in a natural language. (07)
Nope. I used those words as generalisations of the specific “use ISO
15926”, and in doing so I simply mean that you can choose any other theory or
metaphysics as long as you choose something. (08)
>
>> Think continents of agreement rather than islands of isolation.
>
> I agree with all the points you're making. But all those agreements
> are stated in ordinary language. There is no way that they could
> be translated automatically to logic or any other computable form. (09)
Those agreements are specified using ontologies. I’m not sure why you think
otherwise, as it’s clear that’s what I mean. I said nothing about anything
being translated automatically, creating these ontologies is hard work. (010)
>
>>> You can standardize terminology, but in every branch of science,
>>> technology, business, and the arts, the definitions of those
>>> terms and the theories about them are always in constant flux.
>>
>> They are not all in constant flux.
>
> Again, we're talking about different levels of detail. I agree
> that many terms in science have been given precise definitions:
> NaCL, for example. But if you talk about 'salt', even if you
> mean 'sodium chloride', there is a huge difference between what
> you sprinkle on your food and what you sprinkle on an icy highway. (011)
There are always counter examples but that does not disprove my point. I’m
saying is that there are plenty of examples where there is no confusion, and in
those areas the use of small islands of ontologies is counter-productive. I say
that because I see this happening regularly and am tired of seeing so much time
and money wasted by folks following the path you seem to describe. (012)
>
>> it’s continents that should be the objective.
>
> We need agreements at every level, and different people in different
> roles need different terminology and details. For example:
>
> http://www.jfsowa.com/pubs/sowazach.pdf
>
> See Figure 6 on pp. 600-601. That table shows people in five
> different roles who each have six different perspectives on the
> same system -- for a total of 30 different descriptions.
>
> And those are just 30 broad-level views. There is much
> more to be said at the detailed levels. (013)
But all those views are not interesting wrt the subject of my email. I’m
talking about the development of ontologies as the basis for software
applications for data integration and interchange (i.e. the Data column) and
about useful approaches to making those more reusable and interoperable across
disciplines. I’m also saying that their are disciplines that are very well
defined and that ontologies can be built based on that already agreed
understanding. (014)
Ed B say that every modeler should be allowed complete freedom to do things
however they see fit and you say that nobody can agree on anything or that
everything is in flux. You also claimed that everyone agreed with your
assessment of the state of the art and I am pointing out that’s not true.
I’m suggesting that there are well-defined disciplines/domains where there is
no flux, and that a do whatever you want approach wastes time and money. (015)
Cheers,
David (016)
>
> John
>
> _________________________________________________________________
> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
> Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
> Shared Files: http://ontolog.cim3.net/file/
> Community Wiki: http://ontolog.cim3.net/wiki/
> To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (017)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (018)
|