There are potentially an infinite number of definitions of what a river is or is not based on the context in which the question is posed. While I am not passing judgment on any work which may capture a small yet adequate subset of “river” it should certainly never be deemed “all inclusive” or even purport to have contemplated all possible contexts. The subset of definitions is only limited by what was considered or imagined by those defining it.
I agree it would be helpful in providing a good basis for understanding. I have no idea what a “numerical precise” definition can include in terms of components (whether aggregate or composite relationships). Does saline concentration have a certain threshhold when once crossed, the stream of water is no longer considered a river? Is there a minimum speed? A viscosity index rating? A minimal vertical drop?
Again – I am not passing judgment on the work, just posing the possibility that it would be impossible to state that you have captured the definition of a river for all possible use cases.
Duane
On 24/02/09 6:17 PM, "Lieberman Joshua" <jlieberman@xxxxxxxxxxxxxxxxxxxxxxxx> wrote:
I think what Carl was getting at was that geographic feature ontologies provide a particularly clear example of different overlapping / conflicting ways to chop up the world, because the physical world is literally continuous, while features represent in numerically precise ways what does and does not belong to something in it which we wish to recognize. There may be a merriam-webster definition of a river, but there are numerous different and precise definitions of what is river and what is not river. This sort of diversity is the target of mediation efforts such as MMI.
Looking at the m-w definition again, I would also have to say that if one did not already have a pretty clear idea of what a river was, it would be remarkably unhelpful in conveying the idea of river "from scratch".
Josh Lieberman
On Feb 23, 2009, at 2:34 PM, Александр Шкотин wrote:
Carl,
thank you for interesting urls. But I did not find "river" definition from scratch.
and today benchmark is from m-w.com <http://m-w.com> :
http://www.merriam-webster.com/dictionary/river
"1 a: a natural stream of water of usually considerable volume"
if you have url for better definition, please send me one.
[CR]
The real issue is to figure out how to map between these the various ontologies to foster data sharing greater interoperability between and among communities. This, for example, is why the OGC community and the hydrology community have initiated more intense interactions.
<AS>
the simple way to interoperability (especially between formal ontologies and professional community;) is to convert all ontologies and RDB to controled natural language, understandable by both people and formal reasoners.
They positioned RDF as "RDF--the Resource Description Framework <http://www.w3.org/TR/REC-rdf-syntax> --is a universal format for data on the Web." (http://www.w3.org/RDF/FAQ#What <http://www.w3.org/RDF/FAQ> ).
Well, CNL is a universal format for knowledge [not only] on the Web;)
Alex
----- Original Message -----
From: Ali Hashemi <mailto:ali.hashemi+ontolog@xxxxxxxxxxx>
To: [ontolog-forum] <mailto:ontolog-forum@xxxxxxxxxxxxxxxx>
Sent: Monday, February 16, 2009 1:08 AM
Subject: Re: [ontolog-forum] a skill of definition - "river"
@ John,
I appreciate Waismann's point. I don't think what I said is in opposition to it, as far as i can tell, it embraces it. Though I like to get at his insight by thinking that once specified, any system begins to decay. Suggesting then simply that maintenance / upkeep / revision is a necessary part of any specification framework.
@ Matthew
Oops. Thanks for clarifying, I was confusing senses of extension :P.
or whether one particular ontic category hierarchy is appropriate for all, I think efforts would be more fruitful in explicating / generating mappings between what various peoples find useful. I.E. take the IDEAS hierarchy and compare it to DOLCE or SUMO -- what's being reused?
[MW] Wrong question. More important is how do you map from one to the other.
I think those questions (reuse) are part of the process of understanding and developing such mappings.
[MW] Well the way this normally plays out is that those who take an intensional approach do not necessarily think they have something different when the membership of a set changes. So If I ask "How many cars are there". They will give a certain answer, and if I ask the same question a year later, they will give a different number, and will be quite happy that the membership of the set has changed. An extensionalist, on the other hand, will insist that these are actually two different sets: Cars-at-time-1 and Cars-at-time-2, and a 4D extensionalist will say that the set of all cars, is all the cars that have existed and will exist.
So each ontology interprets the notion of set differently, which affects the notion of a definition.
Forgive me if what I say is obvious, but for posterity, in the above we have a statement, "set of all cars" which corresponds to three distinct sets, depending on what framework one employs.
Let's call them:
Sint - intensional
Sext - extensional
S4Dex - 4D extensional
which are unique.
So in an intensional framework, depending on when a query is executed (a question is asked), Sint= Sext@Tquery.
In a 4D extensional framework, set S4Dex = U (forall i) Sext@Ti
where @Tx is the unique name of each set in the extensional perspective at time x.
Seemingly, the above suggests that if we want mappings to work, while each group may choose their own framework, if they intend to interoperate, we need to know what pieces of information we need to track (though perhaps not ontologically commit to), to enable such mappings. Thus if using an intensional framework, with an eye on translating to an extensional one, we'd need to track when extensions are generated, etc.
Is this a semi-accurate catch-up to where people thinking about this issue are?
Ali
On Mon, Feb 16, 2009 at 2:19 AM, John F. Sowa <sowa@xxxxxxxxxxx> wrote:
Ali, Mitch, and Frank,
AH> If you've given a definition of River, and it is inadequate
> when you encounter something as the Okavango River, ought it
> not indicate that you only need to update your definition of
> river? Isn't the whole point of defining something trying to
> abstract the generalizable qualities / properties of the
> object/entity under consideration? You can still have a
> monotonic logic, you just need smart revision policies...
It's always possible to legislate a definition and to revise
the definition whenever you encounter an exception. But the
point of Waismann's notion of 'open texture' is that there is
no stopping point. If you arbitrarily choose a stopping point,
you will inevitably exclude unanticipated cases that are just
as reasonable as the ones you do include.
That is a serious problem for any legal system. Any system
of laws has inevitable exceptions and borderline cases that
require a judge and jury to decide.
_________________________________________________________________
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
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
--
**********************************************************************
Duane Nickull, Vancouver, BC Canada
Senior Technical Evangelist - Adobe LiveCycle ES and Enterprise
Duane's World TV Show - http://tv.adobe.com/#pg+1537
Blog - http://technoracle.blogspot.com
Twitter - http://twitter.com/duanechaos
Community Open Source Music - http://www.mixmatchmusic.com
My Band - http://www.myspace.com/22ndcentury
**********************************************************************
_________________________________________________________________
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
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (01)
|