Fair, Adam. (01)
I am curious, then, how would/should we be handling context in or
UBL-ontology, or, more precisely, in building an ontology for the
business domain (as "context" is an integral part of 'real' business)
-- we can't simply say we won't deal with it when we can't logically
represent it, can we? (02)
-ppy
-- (03)
Adam Pease wrote Mon, 23 Feb 2004 13:27:18 -0800: (04)
> Peter,
> I personally don't see how a discussion on context logic can do
> anything but sap energy from our nascent formalization effort. If
> there's a serious proposal, I'd suggest that it be offered to a peer
> reviewed conference, since any practical implementation of context logic
> and reasoning that could actually help us here would also be a major
> advance in the study of context logic. If a proposal can't get accepted
> or hasn't been formulated to the point of even being a mature submission
> to the context logic conference, it's unlikely to be of any practical
> benefit to this group. The primary context conference web site I'm
> aware of is <http://www.context.umcs.maine.edu/>
>
> Adam (05)
> At 12:47 PM 2/23/2004 -0800, Peter Yim wrote:
>
>> Adam,
>>
>> 1.
>> > ... But there's no agreed on solution for it, ... (and)
>> > ... we don't collectively have the background to even address the
>> > topic intelligently. ...
>>
>> I am in agreement, and hence my earlier suggestion to:
>>
>> >> that we don't pause our current attempt to formalize UBL v1.0 ...
>> >> ... just make sure it does not break up our project work.
>> (if you prefer the word "distract", that is fine too)
>>
>> >> start a new thread to discuss on how we want to deal with
>> >> "context"
>>
>> and,
>>
>> >> I humbly solicit input from those of you who can throw some light
>> >> onto the matter. Try bringing in people who aren't already in the
>> >> community to help too.
>>
>> 2. As for,
>>
>> > context is important ... But ... we're not going to solve it here.
>>
>> We will have to address this in due course, we might as well start the
>> separate discourse sooner rather than later (provided that it is not
>> hampering our progress with the more basic attempts.)
>> I think we'll all hope that you are wrong on this one. :-)
>>
>> -ppy
>> -- (06)
>> Adam Pease wrote Mon, 23 Feb 2004 11:41:49 -0800:
>>
>>> Peter,
>>> I agree that context is important. But there's no agreed on
>>> solution for it, and we're not going to solve it here. I'd venture
>>> to say we don't collectively have the background to even address the
>>> topic intelligently. People have made careers out of working on this
>>> topic.
>>> Most folks here don't even have a background in formal logic. This
>>> is another topic, like so many others, that can only distract us from
>>> the practical task at hand, that doesn't need content in order to
>>> make concrete progress.
>>> Adam (07)
>>> At 10:53 AM 2/23/2004 -0800, Peter Yim wrote:
>>>
>>>> Pat,
>>>>
>>>> > What point of Adam's did you agree with? Could you put it
>>>> > in your own words?
>>>>
>>>> I agree with Adam that "physical" and "abstract" are exhaustive
>>>> under "entity" as represented in SUMO now. I also agree that
>>>> "context" constitutes a whole separate area that is being studied
>>>> and pursued.
>>>>
>>>> I am neither agreeing nor disagreeing whether "context" belongs in
>>>> SUMO. I am, however, of the opinion that "whether or not" or "how"
>>>> context fits into a KIF based ontology (or into SUMO, in our case)
>>>> is beyond the scope of the ontolog-forum (and should belong to a
>>>> SUMO-forum or SUO-forum wherever they are.)
>>>>
>>>> I am not discounting the importance of the matter, though. In fact I
>>>> am positive that "context" will be a key area that needs to be
>>>> addressed if we really want to make a contribution to UBL and to the
>>>> development of business ontologies (I was among those who brought up
>>>> the need for a "context" driven example in our suite of use cases).
>>>>
>>>> I agree then, Pat, that framing it (the need for "context" in our
>>>> representation) in that context makes sense. Our situation is that:
>>>> Pat suggested one possible solution. Adam disagreed, and cited some
>>>> relevant work.
>>>>
>>>> We all want (and need) to move the project forward. May I, then,
>>>> suggest that we don't pause our current attempt to formalize UBL
>>>> v1.0, but to start a new thread to discuss on how we want to deal
>>>> with "context". I am sure the discourse is going to be useful and
>>>> interesting, but long and drawn out ... Let's just make sure it does
>>>> not break up our project work.
>>>>
>>>> I humbly solicit input from those of you who can throw some light
>>>> onto the matter. Try bringing in people who aren't already in the
>>>> community to help too.
>>>>
>>>> Fyi, ... I did a google on "guha mccarthy barwise situation
>>>> calculus", and it returned 279 hits (guha mccarthy situation
>>>> calculus). A lot of those hits will make interesting reading to some
>>>> of us, like myself.
>>>>
>>>> Regards. -ppy
>>>> -- (08)
>>>> Patrick Cassidy wrote Mon, 23 Feb 2004 01:00:03 -0500:
>>>>
>>>>> Peter --
>>>>> I'm a bit unclear as to what you mean by your note:
>>>>> [PY]
>>>>> > Good point, Adam.
>>>>> >
>>>>> > May I suggest that discussions specific to Upper Ontology be
>>>>> considered
>>>>> > as being outside of our scope here. We should, predominantly, be
>>>>> dealing
>>>>> > in the "business" domian ontology. Is that ok with both of you,
>>>>
>>>>
>>>> > > Pat & Adam?
>>>>
>>>>> >
>>>>> > Discussions like that should probably be done elsewhere, say, for
>>>>> > example, at the IEEE-SUO list.
>>>>> >
>>>>
>>>>
>>>>
>>>>> What point of Adam's did you agree with? Could you put it
>>>>> in your own words?
>>>>
>>>>
>>>>
>>>>> The additional content I have been suggesting is in most cases
>>>>> *directly* related to the business Ontology, including the concept of
>>>>> a "Context", which is used within the UBL specification. More
>>>>> detail will have to be added to "Context" to capture the different
>>>>> types of business context that they feel are relevant.
>>>>
>>>>
>>>>
>>>>> I will be quite happy to pursue discussions one-on-one
>>>>> with anyone who has any questions, objections, or
>>>>> suggestions for change in anything I have suggested.
>>>>> However, I will simply ignore any suggestions to junk
>>>>> everything and start all over, as Adam has recommended.
>>>>
>>>>
>>>>
>>>>> I plan to suggest additional content as time permits,
>>>>> and will be happy to discuss specifics about this content
>>>>> in any forum. I will also make, as I have already, comments
>>>>> and suggestions about what others recommend.
>>>>
>>>>
>>>>
>>>>> I have bent over backwards to avoid making any changes
>>>>> in the existing SUMO/MILO, focusing on additions, with
>>>>> only a few renamings where there was a logical error
>>>>> (such as the use of "contract" in different senses)
>>>>> or ambiguity was a threat.
>>>>
>>>>
>>>>
>>>>> I have sent another note explaining why I think that
>>>>> restricting additions of content from members (except
>>>>> for logical inconsistency) is a really really **bad**
>>>>> idea, and why addition of content only peripherally
>>>>> related to the core business concepts can be helpful.
>>>>> Whether or not a suggested business-related concept
>>>>> captures the reality of business practice, or is
>>>>> consistent with the UBL version, are important issues
>>>>> for us. Discussing whether a peripheral concept happens to
>>>>> be essential at this particular time is likely to be a great
>>>>> time-waster, with no possible objective resolution.
>>>>
>>>>
>>>>
>>>>> Does anyone have any constructive comments on the
>>>>> business concepts I have suggested? Anyone?
>>>>> Pat
>>>>> ============================================= (09)
>>>> Adam Pease wrote Sat, 21 Feb 2004 08:26:09 -0800:
>>>>
>>>> > Pat,
>>>> >
>>>> > In general, while of course it's fair to have new content that is
>>>> > in progress, and only partially defined, you are suggesting very
>>>> > substantial additions at a high level of the ontology. The
>>>> > standard must be quite different there. If someone adds a new
>>>> > class JoineryInvoice, and makes it a subclass of Invoice, no harm
>>>> > done. You have however proposed, among other things, adding a
>>>> > new concept right at the top (under the top node of Entity),
>>>> > called "Context", and then a whole new tree of concepts which
>>>> > include Situation, Event etc.
>>>>
>>>> > Now, people spend large parts of their research careers (John
>>>> > McCarthy, Guha etc) working on context logic. There are yearly
>>>> > conferences on the topic. Context usually involves a context logic
>>>> > that is not first order. And yet, you've inserted this notion
>>>> > without definition. In addition, since the first level concepts
>>>> > of Physical and Abstract already form a partition (are exhaustive),
>>>> > it's not even clear in the most basic sense what "Context" means,
>>>> > nor how it would be used in inference, that in most prior
>>>> > research has involved speculative reasoning facilities that
>>>> > don't even exist yet.
>>>>
>>>> > From there the problem just gets worse since you seem to be
>>>> > proposing a whole new scheme for representing situations for SUMO.
>>>> > Again, this is an area that has received a vast amount of research
>>>> > effort, and you've provided only a taxonomy. New content at this
>>>> > level simply has to be justified with axioms even to be a coherent
>>>> > proposal. Otherwise we don't have any basis for a coherent
>>>> > discussion about its merits.
>>>> >
>>>> > Adam
>>>> --- (010)
>>>> > At 12:56 AM 2/21/2004 -0500, Patrick Cassidy wrote:
>>>> >
>>>> >> Adam -
>>>> >> Thanks for your comment. Since the issues you discuss
>>>> >> are relevant to much of the work we will be doing, I am
>>>> >> sending this reply to the whole Ontolog group. The bottom
>>>> >> line of this message is that I think we need to vote
>>>> >> on the issue, so I hope all Ontolog members will
>>>> >> slog through this and provide their own feedback. ...[snip]... (011)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Unsubscribe/Config:
http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (012)
|