Representing Stages and Process States
Mark Wood
In a somewhat similar vein, consider the process of placing a chicken in an oven for some prescribed amount of time. There are at least three distinct states or representations of the chicken: before, during, and after. The chicken may or may not be materially transformed by the process--the oven may or may not have been turned on. Again, there must be a material continuant that represents the chicken across these states. One could follow the example of the Common Core Ontologies, where each stage is an instance of a Stasis, which is ultimately a subclass of BFO's process. This breaks down, however, if it is important for the object at each stage to have its own identity---perhaps the chicken is subsequently carved in a subsequent process and I must associate a distinct material continuant with each slice, while retaining knowledge of its provenance; or perhaps I wish to go back in time to the egg from which the chicken came.
One could represent each stage (be that the stage of the being an ovum or the an uncooked chicken) as a quality, or set of qualities. However, this too would seem to break down if the process does result in a material transformation of the object. Is it appropriate to consider the existence of an independent continuant that spans specifically dependent continuants, and if so, what is the appropriate relationship between the stages and the unifying object?
The thread on process and artifact states in BFO seemed relevant but inconclusive.
Thank you for your thoughts and pointers.
Barry Smith
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/2ef898aa-7dcb-4d97-9b1b-ef16e9461381n%40googlegroups.com.
Chris Mungall
--
Cristian Cocos
Representing "stages" in the development of an object that undergoes a process is the province of the (relatively old) discipline of Systems Theory (aka "Dynamical Systems," "Control Engineering," "State-space representation" etc.), where a "stage" occurs as a state (or set of states) of a system. State of a process would be a derivative notion, namely the aggregate of the states of all the systems that participate in the process--hence we'd have one big system composed of all the participants in the process seen as subsystems of the big system. At any rate, my guess is that Systems Theory's propensity to carve the world into systems, subsystems, and hierarchies thereof, might not constitute a very exciting venue for BFO advocates. Long ago, yours truly attempted to sell a bunch of system-theoretic notions to the BFO community, but the enthusiasm was very low.
C
Mark Wood
Conceptually I do think of it as progressing through a series of states. Temporal part could work but as the cake gets slowly eaten I'm less sure.
Woland's Cat
Thanks for the comments.
In my application, I'm actually interested in inanimate, non-living objects, so perhaps I should not have used the examples I did; they just seemed to be the closest things I could find.
Let's suppose in the baking example we were making a cake. The batter that goes into the oven is an object aggregate. Depending on whether the oven is on or not, it may or may not come out materially transformed. Either way, it is in some sense the same object aggregate. The three phases (before, during, after) then do not correspond to life stages but states. It is helpful to be able to think of the object aggregate in each of the phases as a continuant in its own right, that can be acted on. Someone might stick a knife into the batter, or slice off a piece of cake.
Conceptually I do think of it as progressing through a series of states. Temporal part could work but as the cake gets slowly eaten I'm less sure.
Daniel Arista
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/54ade2f4-246d-c65f-87cd-aaac62b67e6e%40gmail.com.
Woland's Cat
- physics
- sub-atomic
- atomic
- chemistry
- molecular
- materials
- crystalline (in metals, salts)
- phase changes (solid, gas etc)
- biological
- cell organelle
- cellular
- ...
- organism
- ...
- society / community
- ...
Then any idea of 'stages' or 'states' I think has to follow the (correct version of the) above list. I.e. 'changes' from one stage to another occur within one (or more?) of the above levels. So a 'delta' between two stages e.g. pupa and moth is a change at the biological level; for the cake example, it is both chemical and material, and so on. I have never tried to do this analysis seriously, so I don't know how cleanly it would be.
Also, what might be considered life stages of an entity like a
cake are not all necessarily on the same level - rotting is a
different kind of change compared to cooking.
- thomas
Daniel Arista
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/50e3407c-8baf-3872-2654-3692a6f15738%40gmail.com.
Cristian Cocos
As to what the carrier of the system’s identity might be, BFO subscribes, as far as I recall, to an Aristotelian substantialist theory of identity, though what the exact Aristotelian substance this piece of dough instantiates escapes me at this point. Suffice to say that as long as this thing instantiates the same kind, it is the same kind, hence ceases to “exist” when it no longer instantiates the kind. In looser layman’s terms, a thing has, on the one hand, constitutive/individuative attributes, and, on the other, accidental or contingent attributes. The former are the object’s carriers of identity, while the latter constitute the state of the object.
C
From: Daniel Arista
Sent: Monday, October 5, 2020 9:55
To: bfo-d...@googlegroups.com
Subject: Re: [bfo-discuss] Representing Stages and Process States
@wolandscat
Thomas,
ref, " if we agree that chemical reactions, protein modifications (due to butter, eggs doing their thing during the cooking of the cake); if not, then 'chemical reactions' and 'biological cell modifications in response to environmental conditions' are two further potential kinds of 'state' or 'stage' change over time."
...could it be that the 'reaction' or 'modification' is (necessarily) the transformation of some continuant as describable as some process? It would follow that then the matter being transformed, when in the process of transformation, is undergoing a state change; it is in a state of transformation.
If the question is, when does a transformation constitute the end of an entity's existence?; then I think it's important to leverage dispositions and functions.
Batter has the disposition to turn into cake when baked. These dispositions can be tracked down to the subatomic particles .. To ascribe a function, we have to remember it's in the teleological sense per BFO.
There is no cake w/ batter ( don't know what that "cake in a mug" out of the microwave business is :). This imposes a necessary condition for cake (iff). Cake was once batter. Neither cake nor batter is a process, they are both continuants. Processes can induce transformations in continuants (e.g batter to cake). Temporally demarcating (slicing) these at some (potentially fiat) states is all we're doing.
These types of necessary and sufficient conditions, etc. need to come from the axioms of the domain of interest.
Cake has a life cycle, as it has the disposition to rot. If digested by bacteria, it will go through a sufficient transformation where it will no longer be cake. I eat some cake, it may still be cake in my stomach, but not for long. How long do I follow the cake through a lifecycle?? It seems this may be a matter of semantics, and should be formalized in the axioms of domain-specific science. BFO, as a top level ontology, won't tell you where to place these demarcations, it only provides the means to describe them. So if the "essential nature of the organism or system remains the same" then I think focusing on the level of description where things are changing and if/how it changes that continuant's dispositions makes sense. I think of things like the situation calculus or event calculus potentially if you wanted to build these axioms in a sorted FOL.
-Dan
On Sat, Oct 3, 2020 at 12:43 PM Woland's Cat <wolan...@gmail.com> wrote:
I've had some discussions with Barry on this subject before, and one thing we got caught on was what we mean by 'state'. From my (engineering / complex systems) perspective, a 'state' is a bundle of values of particular attributes (BFO qualities) of particular entities making up the system - which could be a patient, some industrial plant etc - at a particular time. So the 'system' - the organism or machine progresses through states in time - this gives you a state-space picture of the entity. (I still want a better way of representing this in BFO, because it's what's needed in most domains to track normal progress of patient vital signs, the operation of an industrial system etc). In all these cases, the essential nature of the organism or system remains the same - it's operating in 'normal mode'.
On the other hand, life 'stages' or 'phases' as between caterpillar and butterfly, egg and blastocyst and so on are something else, and within each stage, the above notion of 'state' will apply. In this notion of state/stage, there are major changes or additions to the entity phenotype, via new morphogenesis or some other mechanism.
Thirdly, we have the physics idea of phase changes, i.e. ice -> water -> vapour etc. I'd say the cake mixture -> cooked cake is similar to this, if we agree that chemical reactions, protein modifications (due to butter, eggs doing their thing during the cooking of the cake); if not, then 'chemical reactions' and 'biological cell modifications in response to environmental conditions' are two further potential kinds of 'state' or 'stage' change over time.
I'd suggest we need to get the language sorted out as well as the ontological distinctions we wish to make, such as the above.
- thomas
On 03/10/2020 15:10, Mark Wood wrote:
Thanks for the comments.
In my application, I'm actually interested in inanimate, non-living objects, so perhaps I should not have used the examples I did; they just seemed to be the closest things I could find.
Let's suppose in the baking example we were making a cake. The batter that goes into the oven is an object aggregate. Depending on whether the oven is on or not, it may or may not come out materially transformed. Either way, it is in some sense the same object aggregate. The three phases (before, during, after) then do not correspond to life stages but states. It is helpful to be able to think of the object aggregate in each of the phases as a continuant in its own right, that can be acted on. Someone might stick a knife into the batter, or slice off a piece of cake.
Conceptually I do think of it as progressing through a series of states. Temporal part could work but as the cake gets slowly eaten I'm less sure.
--
Thomas Beale
Health IT blog | Culture blog | The Objective Stance
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/54ade2f4-246d-c65f-87cd-aaac62b67e6e%40gmail.com.
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/CAGNnAedGm3hnDUO2Wk8oCEUxatJAOSfJpTjs9_T6%2BnTOVg_Fvg%40mail.gmail.com.
robert.bur...@gmail.com
robert.bur...@gmail.com
Pierre Grenon
it is an interesting query that has been discussed in the past yet perhaps can always use clarification or revisiting.
As I understand, you are seeking to represent stages, phases or states of entities where this means an entity that for some period of time is endowed with a number of properties that makes it stand out in its own right and contrast with some sort of temporally more broadly encompassing main entity. The short answer is that, on the side of continuants in BFO, there are no such entities. Continuants endure by being one and the same thing, in their own right, throughout their life. There is no further time-wise decomposition of continuants. This does not mean that what you aim at representing is not representable in BFO. It only means that the representation does not respect the surface syntax of your way of speaking and perhaps it does not respect your intuitions at all.
On the question of finding good examples, the problem with examples is that their validity and relevance is often an empirical matter, a matter of convention or perhaps just language. I believe that you do not have to find an example where people usually use the phrase 'stage' or 'state' or that sort of language in order to raise your question, however. Take any entity, say the Earth, take any arbitrary period of time. The question arises whether there is the Earth, the Earth through year 2019, 2020 or January 2020 or 1 January 2020 etc. Well, in first analysis, it is sorta up to you (it's empirical, conventional, linguistic). What BFO says is that whichever the case with the example of a story of a continuant that extends in time you pick, it is either of two cases:
Case 1. The continuant is an endurant and there is just one entity. There is the Earth and it continues to exist throughout these periods of time.
Case 2. There is a multiplicity of continuants involved. There are temporally secluded Earth things that in their succession amount pretty much to a similar history. In BFO, you would say that these are all entities in their own right and the successive ones are in a relation of genidentity (this captures the idea that one comes from another, you can go on about it but, in effect, here, this is what gives you the capture of your intuition that they are the same thing over time, but there is no extra thing that they would each in turn stand for). They themselves are endurants. There is no 'super temporal Earth'.
These cases are exclusive, there is no BFO case in which there is both a whole entity and its successive parts (whatever that would mean). Indeed BFO has no concept of what that would mean for continuants. Entities that would be like this are persisting in time as perdurants do, for example, processes in BFO and continuants are not perdurants.
To illustrate case 1, we say: there is a chicken (I assume you mean what's really one part of a chicken cadaver, headless, gutless etc). It is a continuant, it has a history. It was in your fridge, it went on your worktop, it went in your oven, it went on your dinner table...
To illustrate case 2, we say: there is a scattered aggregate of ingredients, there is a mixed aggregate of ingredients, there is a cooking aggregate of ingredients and then baked aggregate of ingredients. There is no single entity throughout whatever period of time we picked, contrary to what was the case with the chicken story. Let's say this is because mixing and then baking both trigger substantial changes, it is one entity and then there is a change into another entity and so on. Then there is a succession of continuants that, in their own lifetime, are all like the chicken above. They relate serially in time through genidentity. Perhaps it fits intuition that no cake springs into existence but rather comes from the batter that has undergone a process that changed it as a substance.
- there is a chicken,
- the chicken participates in a chicken-cooking process
- and, during this process, the temperature of the chicken is in the warm range.
When we say 'the chicken before cooking was cold', we say:
- there is the chicken,
- there is a process of, I don't know, thawning or just preparation that occurs outside the oven (and before the cooking process),
- and, during this process, the chicken has a temperature in the ambient range.
Before that, there may have been a refrigeration process, the chicken was being refrigerated (participant), the chicken during this process has a temperature in the cold range. And you can add more along the same lines.
Let's assume we use a few short-cuts in representation, the way you would represent would be along the following lines.
Chicken001 is Continuant
There's just one continuant and a number of processes, say 3 that are salient for our purpose.
Refrigeration001 is Process
Preparation001 is Process
Cooking001 is Process
Chicken001 participates-in Refrigeration001
Chicken001 participates-in Preparation001
Chicken001 participates-in Cooking001
during Refrigeration001, Chicken001 hasTemp Cold
during Refrigeration001, Chicken001 hasTemp Ambiant
during Refrigeration001, Chicken001 hasTemp Warm
When we talk about the batter and the cake, again the batter and the cake will be each like the chicken above but then we link them using some trans-temporal relation like genidentity. There is no super entity of which they are constituents of sort. Instead there is a series of processes and we use these processes that can be sliced up according to any way we want, time-wise or entirely arbitrarily, because these processes have temporal parts or phases, stages --- whatever you want to call that.
Batter001 is Continuant
BakingCakeBatter001 is Continuant
BakedCake001 is Continuant
BakingCakeBatter001 genid Batter001
BakedCake001 genid BakingCakeBatter001
Batter001 hasLife Life002
BakingCakeBatter001 hasLife Life003
Cake001 hasLife Life004
Now let's pick some processes:
Baking002 isPartOf Life003
Cooling001 isPartOf Life004
Serving002 isPartOf Life004
during Life002, Batter001 hasTemp Ambient
during Baking002, BakingCakeBatter001 hasTemp Warm
during Cooling001, Cake001 hasTemp Warm
during Serving002, Cake001 hasTemp Ambient
Well, first of all, if we know when the processes (say, cooking) happen, for some purposes, we can ignore them and come up with purely temporal ascriptions of properties ('it is warm at time t'). But that seems besides the point of interest if we intend to represent, say, the chicken as it is cooking --- in that 'state'. If we follow the above, however, we are led back to the process as the fact that the chicken is in a given 'cooking' state (at that time) is represented by the fact that there is the cooking process.
In the above, Cake001 is like Chicken001 --- there are times during its lifetime when it has different properties. Now, I don't know if cakes are like this. If you disagree, then the alternative is to say that Cake001 is in fact a couple of cakes, one warm that comes first then and one cold that comes next. This is like saying that the first cake and the second cake are analogous to the batter and the 'baking cake batter' duo, and then you say they are genidentical.
Usually, or often, in using or just illustrating BFO, things are handled as Chicken001 unless there are good (in domains, domain expert motivated) non-arbitrary reasons to recognise them to be like the Batter-BakingCakeBatter-Cake trinity. So, just being in different locations or existing at successive times is not quite enough to warrant a substantial change. Chicken (carcasses) can be moved around and undergo a number of processes without really changing what they are, instead they undergo positional or qualitative changes or any number of such changes sometimes at the same time.
The same would apply to a light bulb that is either attached or not to a fixture or that, when it is screwed, is either on or off. In each case it is one light bulb --- not phases, stages or whatnots existing in their own right. We use processes, some which may appear extremely banal, that are parts of the life of the light bulb in order to pinpoint the periods of time during which the light bulb exhibits certain properties of interest. The same for the oven that is on or off.
The banality of these processes is a different problem. What process is it that the chicken participates in when it is sitting on your dinner table? Is there a sitting on the dinner table process? What kind of process is that? I think, I may be wrong, that the appeal to state talk here is a shorthand and the nature of the processes or rather the choice of processes and their identification is again empirical, perhaps conventional and warped by language. There's a lot of processes involving the chicken, it is exposed to its ambient environment in many ways, perhaps it undergoes internal processes, but the idea in terms of BFO is that, whatever these are, they are in any case parts of the sum of all those processes (which is trivial) --- called the life or history of the chicken and so for many purpose we use that because that is often enough and also the best we can do.
I hope that the long email provides a few more concrete comments and perhaps clarifies Barry's suggestion if it needed to be. It would be interesting if you wanted to get back with any difficulty you meet while working on your particular cases, either as rebuttals to the proposed approach or as further elaborations. Also, there were a number of comments in response to your query that I couldn't entirely follow and these may have interesting complementary or valid alternative views --- I haven't commented on any of them.
With many thanks and kind regards,
Pierre
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/0e8f0dfb-78b0-4807-bd6b-13da461ff066n%40googlegroups.com.