Dear Pat,
I've been struggling a bit to understand what it is you are saying, but I
think I may have it at last.
I think what you are saying is that you can state facts in a neutral way
that would allow them to be imported and reasoned over by either an ontology
that was endurantist or perdurantist, and this gives you an easy way to
federate a group of ontologies. You'd obviously have to have shared names
for the types of things you were talking about, but as long as you leave any
axioms in the exporting and importing ontologies, and don't try to impose
them across the exchange, you should be just fine, and as long as the axioms
are valid within their own framework you should be able to do useful work.
Is that right?
I certainly think that works for a lot of the time, but I think there are
edges where you might have to work a bit harder. The obvious example would
be the need to recognise that the endurant object and its life were the same
thing when translated into a perdurant ontology and vice versa. (01)
Regards (02)
Matthew West
http://www.matthew-west.org.uk
+44 750 338 5279 (03)
-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Pat Hayes
Sent: 31 March 2015 16:23
To: William Frank
Cc: [ontolog-forum]
Subject: Re: [ontolog-forum] Endurantism and Perdurantism - Re: Some
Comments on Descriptive vs. Prescriptive Ontologies (04)
On Mar 31, 2015, at 9:27 AM, William Frank <williamf.frank@xxxxxxxxx> wrote: (05)
> Mathew,
>
> I would be very interested to know how you eliminate the boundaries
between these categories. I think they must be eliminable, not to say,
though, that things can't be categorized in one of these ways, depending on
what aspect of a situation we want to look at.
>
> To me, philosophers and too many engineers seem moved to consider whether
there 'really are' processes, objects, and events, and whether they are
'really' different. I note in particular as a description of a paper
>
> "argues that processes are like objects, and distinct from events. "
>
> What is clear is that these are distinctions that **we are able to make**,
as many do. We can give examples, even though the lines between things we
might choose to classify one way or the other, as with most things, is
fuzzy.
>
> The question for the engineer is, is it **useful* to make these
distinctions?
>
> If one makes it critical to success to have 'correctly' classified
everything, into its one true category, and uses a language and a logic that
makes us have to repeat information in each category, then the answer is
that however useful it **might be** to make the distinctions, we are doing
it in a manner that is awkward.
>
> On the other hand, if we treat what we are doing when we classify things
as *casting* them, rather than saying what they 'really are', and have a
method by which things in one category can be *recast* in another, and
information not repeated, then, what practical person, working say in
business process design, would object <330.gif> being able reference
events, processes, and objects. And, anything we would want to talk about,
don't there need to be types for all those things, as well as individuals?
>
> So, if we can categorize things, we can recategorize them, as need be.
Is there a reference in this thread as to how to do that that I missed, that
is not alot of math about time series, or is that it? (06)
That is exactly what I was trying to explain at the start of this thread.
Yes, there is such a way, if we can use the syntactic freedom available in
ISO Common Logic. Suppose we are talking about a relation R between two
things A and B: (R A B) in the CLIF dialect of CL, which uses a LISP-like
prefix syntax. (Or R(A, B) in a perhaps more familiar notation. I will stick
to the CLIF style to keep things coherent.) But things change with time, so
this relationship may be true at one time but not at another, or have a
temporal parameter, or be thought of as holding between temporal 'parts' of
the objects. Respectively: (07)
1 (ist (R A B) T) where 'ist' is the modal 'is true at' operator. (08)
To be strict, this cannot be expressed directly in CL but requires the IKL
extension, and should be written using the IKL 'that' operator to make the
embedded atomic sentence into a term: (ist (that (R A B)) T) See
http://www.ihmc.us/users/phayes/IKL/GUIDE/GUIDE.html for details, especially
#ContextsModalities.) (09)
2 (R A B T)
3 (R (A T)(B T)) (010)
2 treats the objects as continuants: they exist through time, retaining
their identity (so they can be referred to by simple names even in a
temporal framework) but their relations are time-dependent fluents (using
McCarthy's old terminology), so we see a ubiquitious time-argument in most
relational atomic sentences, typically by convention the last argument, as
here.
3. treats the objects as occurrents: they have temporal parts, indicated
here by treating the names as functions from times to temporal parts, and
relationships between temporal sections of entities are asserted timelessly. (011)
So, there are various ways to do it. But my original point was that one can
COMPLETELY IGNORE all philosophical speculation about the metaphysical
nature of these things, and simply treat all these various options as purely
syntactic variations on how to say a fact. There is ONE SINGLE FACT being
expressed here: that R holds between the things A and B at time T. That is
really all that matters, and arguments about whether A and B are continuants
(and so must not be spoken of in style 3, because the holy texts assert that
continuants do not have temporal parts) or are occurrents (and so should not
be spoken of in style 2) is basically just noise. You can view them either
way, if it suits your way of thinking. Nothing of ontological importance
turns on that decision. You can also completely ignore the question, and the
metaphysical distinction it presumes. All it boils down to is where you like
to see your temporal parameters. Do you want to see them attached to
abitrary sentences? (Use a modal language or IKL, see style 1.) Or to
relations in atomic sentences? (Style 2) Or to referring terms? (Style 3)
There really are no other options in first-order logical syntax, so you have
to choose one (or more) of these. So go ahead, choose whichever one makes
you comfortable. It is easy to translate between these variations, given a
certain minimal discipline about where temporal parameters are placed. In
ISO CL, one can even write axioms which will do the translation, along the
lines of (012)
(forall (r x y (T time))(iff (r x y T)(r (x T)(y T)) )) (013)
although it takes a few more lines to do this for every possible number of
arguments; and in any case, you might want to be more picky about where you
put the temporal parameters in some cases. I do not recommend actually using
axioms to make the translation, but the fact that is is possible, and even
easy, surely suggests that the distinctions are not as deep as many have
assumed. (014)
But to return to your question, such translation axioms are exactly the
'recategorization' to which you refer. x and y here are treated as
continuants on the LHS of the iff and as occurrents on the RHS. And yet they
are both identically the same thing on both sides of the equivalence. (015)
Pat Hayes (016)
>
> Tx
>
> Wm
>
>
>
>
> On Tue, Mar 31, 2015 at 9:59 AM, Obrst, Leo J. <lobrst@xxxxxxxxx> wrote:
> Matthew,
>
>
>
> Yes, this does not break down the distinction between continuant and
occurrent, but instead argues that processes are like objects, and distinct
from events. They are following and building on [1], and of course other
papers in that vein, including other papers of Galton.
>
>
>
> Thanks,
>
> Leo
>
>
>
> [1] Stout, R. (1997). Processes. Philosophy, 72, 19-27.
>
>
>
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
> [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Matthew
> West
> Sent: Tuesday, March 31, 2015 2:27 AM
> To: '[ontolog-forum] '
> Subject: Re: [ontolog-forum] Endurantism and Perdurantism - Re: Some
> Comments on Descriptive vs. Prescriptive Ontologies
>
>
>
> Dear Leo,
>
> Whilst this is a paper that does a good job of showing how interdependent
continuants and occurrents are. It still maintains the dichotomy of
continuant and occurrent, and even explicitly states that this means a
duplication of an occurrent and its life. So it has not actually taken the
step of breaking down the barriers between them and picking up that
interdependence might point to something common underlying them both.
>
>
>
> Regards
>
>
>
> Matthew West
>
> http://www.matthew-west.org.uk
>
> +44 750 338 5279
>
>
>
>
>
>
>
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Obrst, Leo J.
> Sent: 25 March 2015 19:45
> To: [ontolog-forum]
> Subject: Re: [ontolog-forum] Endurantism and Perdurantism - Re: Some
> Comments on Descriptive vs. Prescriptive Ontologies
>
>
>
> One interpretation (process, event) that Robert may be referring to is the
"waterfall" paper:
>
>
>
> Galton, Antony, and Riichiro Mizoguchi. 2009. The water falls but the
waterfall does not fall: New perspectives on objects, processes and events.
Applied Ontology 4 (2009), pp. 71-107, DOI 10.3233/AO-2009-0067, IOS Press.
>
>
>
> Thanks,
>
> Leo
>
>
>
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
> [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Matthew
> West
> Sent: Monday, March 23, 2015 3:12 PM
> To: '[ontolog-forum] '
> Subject: Re: [ontolog-forum] Endurantism and Perdurantism - Re: Some
> Comments on Descriptive vs. Prescriptive Ontologies
>
>
>
> Dear Robert,
>
> If you get something going with this, I'll be interested in the outcome.
>
>
>
> Regards
>
>
>
> Matthew West
>
> http://www.matthew-west.org.uk
>
> +44 750 338 5279
>
>
>
>
>
>
>
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
> [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of
> rrovetto@xxxxxxxxxxx
> Sent: 23 March 2015 17:21
> To: [ontolog-forum]
> Subject: Re: [ontolog-forum] Endurantism and Perdurantism - Re: Some
> Comments on Descriptive vs. Prescriptive Ontologies
>
>
>
>
>
> On Mon, Mar 23, 2015 at 5:21 AM, Matthew West <dr.matthew.west@xxxxxxxxx>
wrote:
>
> Dear Robert,
>
>
>
> Replies below after [RR] (= Robert Rovetto)...
>
> [MW>] I think it is more useful to think of being a process or a
> physical object are different views on things, rather than being
> entirely different things or just eliminating the physical object view
>
> [RR] Agreed. A colleague not long ago mention this vis-a-vis bfo: that it
should return to the idea that the continuant-occurent (or snap-span)
distinction are two perspectives on the world.
>
> [MW>] The key is whether it is one object with two perspectives, or one
object per perspective with the perspectives being mutually exclusive (as
BFO requires). Changing between these two might seem simple, but it is a
change to a core commitment of the ontology. I'd rather not be around when
you suggested it to Barry Smith.
>
>
>
> For sure I think the perspectives are ours and can be many. I question,
and others should as well, whether the mutual exclusivity (I think some have
here) of any given two or more perspectives--speaking of any ontology,
now--should be. Right, changing b/w them for that particular ontology would
be an issue, but no one should have any consternation to suggesting it, not
if the intention is to contribute, help and ensure that the ontology or
system in question itself helps the communities it serves. And if people
have consternation then something is wrong.
>
>
>
> It is also more help, and perhaps less stressful, to think of them as
perspectives rather than one or the other being metaphysically (objectively)
reflective of reality.
>
> [MW>] I would not see these different perspectives as not being reflective
of reality, just reflective of different aspects of it.
>
> Some have metaphysical views, and there may be a fact of the matter, but
when it comes to solving real-world problems (to the extent that applied
ontologies can even do so!) taking the perspective approach appears better
since, again, the goal is to solve problems.
>
> [MW>] The question is, if there is one object, with a physical object and
process perspective, what sort of thing is it that allows those two
perspectives? I suppose you could just say that they are particulars without
saying anything more. My answer would be chunks of space-time (or
spatio-temporal extents). A chunk of space time might be the spatio-temporal
extent of a person, or it might be the spatio-temporal extent of a meeting,
or of a person whistling.
>
>
>
> My intuition is that a given physical object is more than s-t extent/c,
but since this question delves into space-time, it might be wise to consult
physicists. A number of interesting issues come up here with the question:
implicit (perhaps outdated/naive-physics sense) conceptions of space and
time (e.g. container view), s-t boundaries of, say, a meeting (event), etc.
>
>
>
>
>
> Some have held that processes (but not events) endure.
>
> [MW>] I'm not sure what you mean by this last statement.
>
> [RR] What I meant was this. In the traditional sense...
> Endurance is the form of persistance attributed of endurants/continuants
(objects).
> Perdurance is the form of persistance attributed of perdurants/occurrents
(processes, events).= temporal extension, temporal parts, etc. Some in
philosophy and applied onto hold that processes are different from events,
the former enduring (no temporal parts), with the latter perduring (having
temporal parts).
>
> [MW>] So an event has zero thickness in time? That would be how I use the
term, as a temporal boundary. But many use it as a synonym for activity or
process.
>
>
>
> [MW>] No, I disagree here. I (and I think Pat) would consider that having
an upper level ontology like that of OBO is better than no upper level at
all, even though neither of us would want to use it ourselves. It is
important that different parts of an ontology are consistent, or you just
end up in all kinds of mess.
>
> I've said it before, but it's worth repeating. The problems arise in
> the constraints that an ontology imposes. You need to be very critical
> of any constraint that is proposed. Leave it out unless you are
> certain it is one that always applies, no exceptions ever
>
> [RR] I did not say no upper level at all. I said ontology projects, such
as obo foundry, should not have *as a rule/requirement* the adoption of this
or that particular top-level. Certainly not for projects dealing with
socially critical data and subject matters such as health. There are too
many risks.
>
> [MW>] The risks however, compare to the certainty that if you use
different upper level ontologies, you will have work to do to get them to
interoperate.
>
> We want the models to be helpful, match the helpful domain
conceptualizations and domain knowledge, and we do not want the ontological
systems or ontologists to impost this is or that philosophical view that may
in fact distort the domain knowledge, or worse distort the way domain
scientists think! (an interesting research project, yes. Interested parties,
please contact me). Some in the foundry have agreed on no rule/requirement,
even prior to me even thinking about it (independent of knowing their
sharing this point).
>
> [MW>] So how are you going to determine your upper level ontology, i.e.
the top level commitments and constraints you are going to accept?
>
>
>
> You're right. But I think you might be missing the point. I did not
say--nor did I mean to imply--that multiple top-level ontologies should be
used for a given domain ontology or for a project like obo. I said, again,
it should NOT be a rule or requirement (of potential ontology members of the
project or of ontologies being subsumed) to use this or that particular
ontology. That's all. The point is that in making it a rule, you take steps
toward monopolization, and greater risk of what some people in this thread
have concerns about also, e.g., imposition, forcing, constraints, syntax
issues, etc. The goal is to solve real-world problems, and if there happens
to be a different upper-level that can help accomplish them (or even better
captures the domain), then such a rule would stand in that way. This does
not mean change top-levels haphazardly. In fact I would hope that any
top-level has checks and balances in place to ensure those risks are not
realized and that it is open to change in the light of discovery and
error-finding. The concern is largely ensuring that the system solves
real-world problems it's intended to (to the extent ontologies will/have
even prove useful to do that!), and that the domain science be represented
faithfully without any distorting affects on domain scientists thinking
(assuming their thinking is rational).
>
>
>
> Regards
>
>
>
> Matthew West
>
> http://www.matthew-west.org.uk
>
> +44 750 338 5279
>
>
>
>
>
> [MW] I've said it before, but it's worth repeating. The problems arise in
the constraints that an ontology imposes. You need to be very critical of
any constraint that is proposed.
>
> [RR] I agree.
>
>
>
> Best,
>
> Rob
>
>
>
> On Sat, Mar 21, 2015 at 2:57 PM, Matthew West <dr.matthew.west@xxxxxxxxx>
wrote:
>
> Dear Roberto
>
> <snip>
>
> Two good points:
> 1a) Questioning the mutual exclusivity of the object-process
endurant-perdurant distinction.
>
> There is, at least to me, something odd about conceptualizing a process as
distinct from the participant, at least in any objective or metaphysical
sense. In reality whatever we call 'process' and their 'participant' (or
'object') are mutually interrelated. The distinction, the separation, may at
most be an artificial one. The question is, what are symbolisms or
representations that better capture that?
>
> [MW>] The way that I do it is that an activity/process consists of its
participants, where a participant is the state of a particular whilst it
participates in the activity/process. This works both for things like a
banana ripening, where there is only one participant, and for things like
meetings, or a game of football, where there are multiple participants.
>
> 1b) And opening the door to other conceptualizations of these categories.
>
> We also read: "I see no strong or principled difference between things
undergoing change and processes of change in things"
> This intuition is shared by others and should be explored and formalized.
But it need not mean that things are processes in the traditional
perdurantist sense.
>
> [MW>] I think it is more useful to think of being a process or a physical
object are different views on things, rather than being entirely different
things or just eliminating the physical object view.
>
> Some have held that processes (but not events) endure.
>
> [MW>] I'm not sure what you mean by this last statement.
>
> 2) Questioning and preventing the formalization (or the
> symbolism/logic) from distorting or misrepresenting the world (or the
> conceptualization of it we want to formalize)
>
> - "axiom-bloat"
> - "I meant decisions such as whether to treat a concept as a relation or a
function or an individual, where to locate the temporal parameters, whether
or not one uses a discipline to keep differently typed parameters distinct,
and if so what it is, and so on. There are many alternative ways to express
a given set of facts in a given formal language"
>
> A question to ask is how much do philosophical theories/views affect the
treatment of the concepts and the symbolism.
> For example, the concern about forcing the distinction or requiring a
specific syntax--a concern I've expressed elsewhere--is important. The obo
foundry and other similar projects should not have as a rule/requirement a
particular upper-level ontology. This might seem contrary to the goal of
interoperability in the domain, but it is simply to ensure that the forcing
does not take place, that monopolies are avoided, and that alternative
representations that might better serve the biomedical community are sought
and available/open to be sought and created.
>
> [MW>] No, I disagree here. I (and I think Pat) would consider that having
an upper level ontology like that of OBO is better than no upper level at
all, even though neither of us would want to use it ourselves. It is
important that different parts of an ontology are consistent, or you just
end up in all kinds of mess.
>
> I've said it before, but it's worth repeating. The problems arise in the
constraints that an ontology imposes. You need to be very critical of any
constraint that is proposed. Leave it out unless you are certain it is one
that always applies, no exceptions ever.
>
> The goal is (should be) *the solving of real-world problems*, and health,
biomedicine, privacy, etc. are most certainly domains where we should keep
that in mind. The particular upper-level (or otherwise) views and symbolisms
should not hinder that goal. The point about the limits of owl is also worth
repeating.
>
> Finally, I find what Avril S. said interesting. But there may be mistake
in: "a particular at one time is called an occurrent; a sequence of two or
more particulars at two or more consecutive times is called a continuant."
> In the traditional endur-perd/contin-occur sense, a partiular *at a time*
would be a continuant, i.e., a wholly-present persisting entity. If parts of
occurrents are particulars, then it could be a temporal part (slice) of an
occurrent, but not the whole occurrent. And I think a particular over a time
interval would be an occurrent.
>
> [MW>] Traditional occurrents don't have temporal parts of course, and at
each time it exists you have all of it. Probably better to use another name
if you mean something different.
>
>
>
> Regards
>
>
>
> Matthew West
>
> http://www.matthew-west.org.uk
>
> +44 750 338 5279
>
>
>
> _________________________________________________________________
> 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
>
>
>
>
>
>
> _________________________________________________________________
> 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
>
>
>
>
>
>
> _________________________________________________________________
> 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
>
>
>
> _________________________________________________________________
> 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)
------------------------------------------------------------
IHMC (850)434 8903 home
40 South Alcaniz St. (850)202 4416 office
Pensacola (850)202 4440 fax
FL 32502 (850)291 0667 mobile (preferred)
phayes@xxxxxxx http://www.ihmc.us/users/phayes (018)
_________________________________________________________________
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 (019)
_________________________________________________________________
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 (020)
|