Statements in neurondb effectively say that membranes of certain compartments of certain neurons conduct various ionic currents. I'm having a little trouble parsing this into BFO.
it must be possible to draw an analogy with rivers, river banks and dams and river flows
my first reaction is to say:
- membranes et al are substantial of sort, so are the ions
- the ions move, I'd say what we are looking at here with 'ionic currents'' is a processual of sort, the movement of ions
in any event, my first though would be to say that nothing like a current distinct from either ions or their movements.
it is even more difficult considering there may be patterns of movement involved, but you should be fine saying that there are movements of sorts in which ions are involved in the relevant areas. the fine grained description of this would probably be too complex for the purpose
pierre
On 4/16/07, Alan Ruttenberg <alanruttenb...@gmail.com> wrote:
> Statements in neurondb effectively say that membranes of certain > compartments of certain neurons conduct various ionic currents. > I'm having a little trouble parsing this into BFO.
So, a related question. Is there a mechanism in BFO to associate particulates with a representation as mass? So - water-as-molecules vs water-as-substance? This seems relevant for the modelling of these neurons, as individual ions pass through individual channels in a particular region of membrane. When viewed in bulk, this movement gives rise to a current, which is itself a measure that only applies to a bulk, not a particulate.
A similar argument could be made for temperature - the particles have kinetic energy, and the substance has a temperature exactly under those conditions where the kinetic energies follow a poison distribution. The temperature is intimately associated with the moments of that distribution.
The particulate/substance distinction is important in experimental sciences. Otherwise you end up accidentally claiming that homeophathy realy could work in the physical world, or being utterly unable to account for the random motion of pollen in a petri-dish of water.
> it must be possible to draw an analogy with rivers, river banks and > dams and river flows
> my first reaction is to say:
> - membranes et al are substantial of sort, so are the ions
> - the ions move, I'd say what we are looking at here with 'ionic > currents'' is a processual of sort, the movement of ions
> in any event, my first though would be to say that nothing like a > current distinct from either ions or their movements.
> it is even more difficult considering there may be patterns of > movement involved, but you should be fine saying that there are > movements of sorts in which ions are involved in the relevant areas. > the fine grained description of this would probably be too complex for > the purpose
> So, a related question. Is there a mechanism in BFO to associate particulates > with a representation as mass? So - water-as-molecules vs water-as-substance? > This seems relevant for the modelling of these neurons, as individual ions > pass through individual channels in a particular region of membrane. When > viewed in bulk, this movement gives rise to a current, which is itself a > measure that only applies to a bulk, not a particulate.
> A similar argument could be made for temperature - the particles have kinetic > energy, and the substance has a temperature exactly under those conditions > where the kinetic energies follow a poison distribution. The temperature is > intimately associated with the moments of that distribution.
> The particulate/substance distinction is important in experimental sciences. > Otherwise you end up accidentally claiming that homeophathy realy could work > in the physical world, or being utterly unable to account for the random > motion of pollen in a petri-dish of water.
> Matthew
> On Monday 16 April 2007, Pierre Grenon wrote: > > aye, this sounds tricky
> > it must be possible to draw an analogy with rivers, river banks and > > dams and river flows
> > my first reaction is to say:
> > - membranes et al are substantial of sort, so are the ions
> > - the ions move, I'd say what we are looking at here with 'ionic > > currents'' is a processual of sort, the movement of ions
> > in any event, my first though would be to say that nothing like a > > current distinct from either ions or their movements.
> > it is even more difficult considering there may be patterns of > > movement involved, but you should be fine saying that there are > > movements of sorts in which ions are involved in the relevant areas. > > the fine grained description of this would probably be too complex for > > the purpose
My background is as a molecular neurophysiologist doing single ion channel, so I can vouch for why the description appears as it does in NeuronDB.
NeuronDB as it's name implies is a database of physiological models - models both at the molecular level (ion channel mechanics if you will) and on the cellular level - e.g., based on the composite ion channels present in the membranes of specific compartments in specific cells, those compartments' collective "ion currents" contribute to the overall spatio-temporal, emergent physiological excitable of the neuron to which they belong.
These models go back to the original experiments by Alan Hodgkin & Andrew Huxley in the mid-1930's, when the Squid Giant Axon membrane was studied in isolation. It was extruded like a toothpaste tube, salines on either side of the membrane (in the tube and surrounding the tube) were replaced with known ionic composition, and the propensity of current to flow as the voltage across the membrane was varied was tested under these specific ion conditions. The result was the characterization of specific Na+, K+, and Cl- conductance that all showed a particular time-course and voltage-dependence.
Soon after that, one of H&H's collaborators, Bernard Katz - and others - began to study the other major family of conductances found in neurons - the neurotransmitter-gated conductances found on the post-synaptic membrane.
Prior to the 1980s when it became possible to study single-ion channels in isolation, most all of the modeling was done at the level of membrane conductances - apart from a variety of noise analysis statistical techniques used to determine smaller units of conductance, such as the "quantum" of current generated when a SINGLE presynaptic transmitter vesicle fuses with the membrane and releases its contents into the synaptic cleft (work first performed by B Katz et al).
Essentially all the modeling of ion conductances in excitable membrane now-a-days cycles back-n-forth between modeling ion channel molecular events leading to individual ion penetration of the channel flowing down it's electrochemical gradient and modeling the emergent behavior of entire cells or portions of cells to macroscopic events such as depolarization or saline ion manipulation (and drug manipulation to study channel populations with specific drug sensitivities).
It's a standard modeling technique to characterize the membrane as essentially a capacitor, resistor, and transistor all wired in parallel. The lipid bilayer acts a capacitor (with a characteristic capacitance). There is typically a "passive", non-specific "leakage" current flowing down the resistance, and then there are all the ion channels, which essentially function as voltage-dependent (or transmitter-dependent) current sources (similar to a transistor). Typically, now that many characteristic measurements have been made on specific ion channel species, models seek to place specific populations of channels into specific compartments with the goal of matching the collective molecular-level conductances with the emergent compartment and cellular-level electrical activity.
That is the typical modeling approach, and that's where this assertion that "ion currents" are contained within the plasma membrane derives from.
If one want's to model ion channels as part of the "electrical excitability machinery" then they could treat them as continuants that are contained with the plasma membrane. Given the nature of the experimental work done presently, you'd want to be as specific as possible as to exactly WHERE across the cell the given ion conductance and/or ion channels have been identified.
If one wants to model ion channels as contributors to the "electrical excitability process", then you'd need to address the specifics of the role they play in that process. Simply to say they dwell in the membrane won't be sufficient, but dropping to the level of molecular events won't really suffice either. The "integrating" performed by the models which matches ions flowing down their electrochemical gradient through molecular channels to the emergent electrical behavior in compartments (very much defined as fiat boundaries) has a lot of unspecified details. It's certainly pretty meaningless to simply say neuron X has ion currents A,B, & C in its membrane - IF the goal is to effectively describe the electrical activity of the neuron - since SO MUCH of the behavior is based on the specific location, density, and local inter-relatedness of different channel types.
This is really a challenge.
I believe in the end, it will depend on what details one is trying to capture - and to what purpose - that may best guide the process of mapping neuronal ionic currents into BFO classes.
> it must be possible to draw an analogy with rivers, river banks and > dams and river flows
> my first reaction is to say:
> - membranes et al are substantial of sort, so are the ions
> - the ions move, I'd say what we are looking at here with 'ionic > currents'' is a processual of sort, the movement of ions
> in any event, my first though would be to say that nothing like a > current distinct from either ions or their movements.
> it is even more difficult considering there may be patterns of > movement involved, but you should be fine saying that there are > movements of sorts in which ions are involved in the relevant areas. > the fine grained description of this would probably be too complex for > the purpose
> pierre
> On 4/16/07, Alan Ruttenberg <alanruttenb...@gmail.com> wrote:
>> Statements in neurondb effectively say that membranes of certain >> compartments of certain neurons conduct various ionic currents. >> I'm having a little trouble parsing this into BFO.
Bill Bug Senior Research Analyst/Ontological Engineer
Laboratory for Bioimaging & Anatomical Informatics www.neuroterrain.org Department of Neurobiology & Anatomy Drexel University College of Medicine 2900 Queen Lane Philadelphia, PA 19129 215 991 8430 (ph) 610 457 0443 (mobile) 215 843 9367 (fax)
Please Note: I now have a new email - William....@DrexelMed.edu
I believe that neuroweb is talking about conductance, not conduction. It's not the movement of charge that is being talked about, but the relationship between voltage and resultant current. Or a description of how things can move, rather than the fact of their movement.
Phil
>>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
PG> aye, this sounds tricky
[...]
PG> it is even more difficult considering there may be patterns of PG> movement involved, but you should be fine saying that there are PG> movements of sorts in which ions are involved in the relevant PG> areas. the fine grained description of this would probably be PG> too complex for the purpose
PG> pierre
PG> On 4/16/07, Alan Ruttenberg <alanruttenb...@gmail.com> wrote: >> >> Statements in neurondb effectively say that membranes of certain >> compartments of certain neurons conduct various ionic currents. >> I'm having a little trouble parsing this into BFO. >> >> Suggestions? >> >> http://neuroweb.med.yale.edu/senselab/neuron_ontology.owl
i haven't had time to look at the neuro ontology yet, in my reply I was trying to parse what I understood of the issue from a BFO perspective
the way BFO describes how things can move is by speaking of these things as participating in their movement, possibly by having among other thigs further specific categories of movements. Conceivably, moving things would have relevant property particulars (dependent continuants) which account for their movements being what they are... possibly you could apply this to a stream of ions as having a given property particular which would account for whichever physical property (not immediately in the BFO sense but such as the quantitative characterizations you find in physics) you are interested in here
i'll look at the file and see whether I can revise or precise this statement
pierre
On 4/16/07, Phillip Lord <phillip.l...@newcastle.ac.uk> wrote:
> I believe that neuroweb is talking about conductance, not > conduction. It's not the movement of charge that is being talked > about, but the relationship between voltage and resultant current. Or > a description of how things can move, rather than the fact of their > movement.
> Phil
> >>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
> PG> aye, this sounds tricky
> [...]
> PG> it is even more difficult considering there may be patterns of > PG> movement involved, but you should be fine saying that there are > PG> movements of sorts in which ions are involved in the relevant > PG> areas. the fine grained description of this would probably be > PG> too complex for the purpose
> PG> pierre
> PG> On 4/16/07, Alan Ruttenberg <alanruttenb...@gmail.com> wrote:
> >> Statements in neurondb effectively say that membranes of certain > >> compartments of certain neurons conduct various ionic currents. > >> I'm having a little trouble parsing this into BFO.
Well conductance of a membrane would be something which depends on both the membrane and the ion I would assume.
To give an analogy, the ratio between the speed and pressure for a fluid travelling down a pipe depends on the pipe (the narrower the bore, the smaller the ratio) and the fluid (treacle is thicker than water).
For water under reasonable conditions this relationship will approximate linear, although this will break down when you get turbulent flow, when it will probably become quadratic, cubic or some really weird power function that you only seem in fluid dynamics. I guess something similar is true with ionic currents as well.
Phil
>>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
PG> hear that...
PG> i haven't had time to look at the neuro ontology yet, in my PG> reply I was trying to parse what I understood of the issue from PG> a BFO perspective
PG> the way BFO describes how things can move is by speaking of PG> these things as participating in their movement, possibly by PG> having among other thigs further specific categories of PG> movements. Conceivably, moving things would have relevant PG> property particulars (dependent continuants) which account for PG> their movements being what they are... possibly you could apply PG> this to a stream of ions as having a given property particular PG> which would account for whichever physical property (not PG> immediately in the BFO sense but such as the quantitative PG> characterizations you find in physics) you are interested in PG> here
PG> i'll look at the file and see whether I can revise or precise PG> this statement
PG> pierre
PG> On 4/16/07, Phillip Lord <phillip.l...@newcastle.ac.uk> wrote: >> >> >> >> >> I believe that neuroweb is talking about conductance, not >> conduction. It's not the movement of charge that is being talked >> about, but the relationship between voltage and resultant >> current. Or a description of how things can move, rather than the >> fact of their movement. >> >> Phil >> >> >> >>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes: >> PG> aye, this sounds tricky >> >> [...] >> PG> it is even more difficult considering there may be patterns of PG> movement involved, but you should be fine saying that there are PG> movements of sorts in which ions are involved in the relevant PG> areas. the fine grained description of this would probably be PG> too complex for the purpose >> PG> pierre >> PG> On 4/16/07, Alan Ruttenberg <alanruttenb...@gmail.com> wrote: >> >> >> >> Statements in neurondb effectively say that membranes of >> >> certain compartments of certain neurons conduct various ionic >> >> currents. I'm having a little trouble parsing this into BFO. >> >> >> >> Suggestions? >> >> >> >> http://neuroweb.med.yale.edu/senselab/neuron_ontology.owl >> >> >> > >>
PG>
-- Phillip Lord, Phone: +44 (0) 191 222 7827 Lecturer in Bioinformatics, Email: phillip.l...@newcastle.ac.uk School of Computing Science, http://homepages.cs.ncl.ac.uk/phillip.lord Claremont Tower Room 909, skype: russet_apples Newcastle University, NE1 7RU
so conductance of a membrabe is a property particular which inheres in that membrane
I think this would suffice
you could type it as a conductance in relation to the class of ion say, in order to account for the fact that this is tied (dependent on) ions. The trick is that it is not specifically dependent on particular ions, right? It sounds then that conductance would fall under disposition or tendencies in BFO
Now, you could also attach to ions property particulars of a certain type which would account for the way they would behave in membranes of a given sort
- the pipe as a disposition to let fluid of a certain type travel through it at a certain speed
- the fluid as a disposition of travelling at a certain speed in pipes of a certain type
These properties are distinct although presumably dependent on various other properties (shape and rugosity of the pipe, etc)
with respect to the mathematical characterization of these, I gather this is not an issue specific to the sort of entities at end, anything which can be quantified and come with a scale will be more or less treated in teh same way, I gather. So, the treatment of the quantitaive aspects of conductance will be similar to that of length or weight say.
does that make sense?
pierre
On 4/16/07, Phillip Lord <phillip.l...@newcastle.ac.uk> wrote:
> Well conductance of a membrane would be something which depends on > both the membrane and the ion I would assume.
> To give an analogy, the ratio between the speed and pressure for a > fluid travelling down a pipe depends on the pipe (the narrower the > bore, the smaller the ratio) and the fluid (treacle is thicker than > water).
> For water under reasonable conditions this relationship will > approximate linear, although this will break down when you get > turbulent flow, when it will probably become quadratic, cubic or some > really weird power function that you only seem in fluid dynamics. I > guess something similar is true with ionic currents as well.
> Phil
> >>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
> PG> hear that...
> PG> i haven't had time to look at the neuro ontology yet, in my > PG> reply I was trying to parse what I understood of the issue from > PG> a BFO perspective
> PG> the way BFO describes how things can move is by speaking of > PG> these things as participating in their movement, possibly by > PG> having among other thigs further specific categories of > PG> movements. Conceivably, moving things would have relevant > PG> property particulars (dependent continuants) which account for > PG> their movements being what they are... possibly you could apply > PG> this to a stream of ions as having a given property particular > PG> which would account for whichever physical property (not > PG> immediately in the BFO sense but such as the quantitative > PG> characterizations you find in physics) you are interested in > PG> here
> PG> i'll look at the file and see whether I can revise or precise > PG> this statement
> PG> pierre
> PG> On 4/16/07, Phillip Lord <phillip.l...@newcastle.ac.uk> wrote:
> >> I believe that neuroweb is talking about conductance, not > >> conduction. It's not the movement of charge that is being talked > >> about, but the relationship between voltage and resultant > >> current. Or a description of how things can move, rather than the > >> fact of their movement.
> >> Phil
> >> >>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
> PG> aye, this sounds tricky
> >> [...]
> PG> it is even more difficult considering there may be patterns of > PG> movement involved, but you should be fine saying that there are > PG> movements of sorts in which ions are involved in the relevant > PG> areas. the fine grained description of this would probably be > PG> too complex for the purpose
> PG> pierre
> PG> On 4/16/07, Alan Ruttenberg <alanruttenb...@gmail.com> wrote:
> >> >> Statements in neurondb effectively say that membranes of > >> >> certain compartments of certain neurons conduct various ionic > >> >> currents. I'm having a little trouble parsing this into BFO.
>>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
PG> so conductance of a membrabe is a property particular which PG> inheres in that membrane
PG> I think this would suffice
PG> you could type it as a conductance in relation to the class of PG> ion say, in order to account for the fact that this is tied PG> (dependent on) ions. The trick is that it is not specifically PG> dependent on particular ions, right?
No, I don't think so. You should expect a membrane with for example Chloride channels to have a very different conductance to chloride ions than you would to phosphotase ions.
PG> It sounds then that conductance would fall under disposition or PG> tendencies in BFO
PG> Now, you could also attach to ions property particulars of a PG> certain type which would account for the way they would behave PG> in membranes of a given sort
Why would you not attach this to the membrane rather than the ion? It seems to me that the conductance of a membrane is a mathematical relationship between voltage and current for a given type of ion over a given type of membrane. Conductance is a property of two things not one.
PG> So, the treatment of the quantitaive aspects of conductance will PG> be similar to that of length or weight say.
PG> does that make sense?
Not entirely. Mass can be specified as force over acceleration -- at least when the observer has a static frame of reference, although not in general. This is independent of the entity providing the force. Likewise length which is dependent only on the entity (although again not in general).
Conductance is more similar to force producing a velocity against a friction. It's dependent on the entity being pushed and the entity causing the friction. And as I said previously, this is all dependent on the velocity of the entity being pushed -- in general, these measurements are taken at the linear phase, but it breaks down eventually if you go fast enough.
Both membranes and ions have properties but they do not have the same property,
or rather
there is one type of properties for membranes (the property of alloywing for this and that wrt ions of a certain sort) and there is one type of properties for ions (the property of doing this or that wrt membranes of a certain sort)
Moreover, membranes can have properties of one type in relation to ions of a given type and properties of another type in relation to ions of another type, I meant this too
More in line below...
On 4/16/07, Phillip Lord <phillip.l...@newcastle.ac.uk> wrote:
> >>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
> PG> so conductance of a membrabe is a property particular which > PG> inheres in that membrane
> PG> I think this would suffice
> PG> you could type it as a conductance in relation to the class of > PG> ion say, in order to account for the fact that this is tied > PG> (dependent on) ions. The trick is that it is not specifically > PG> dependent on particular ions, right?
> No, I don't think so. You should expect a membrane with for example > Chloride channels to have a very different conductance to chloride > ions than you would to phosphotase ions.
This is not what I meant. I didn't ask whether the property in question of the membrane is the same for all ions, I agree with you that it could be the most specific properties in questions are likely or could be different depending on the kind of ions. Although, they would all be determinates (more specific) of a same determinable (the family of properties if you want).
I meant, the property of the membrane is not dependent on ion1, ion2, ion3 etc where these are particular ions (not kinds), but is all things being equal the same which ever particular ions of a specific kind (chloride, etc) happen to traverse it.
> PG> It sounds then that conductance would fall under disposition or > PG> tendencies in BFO
> PG> Now, you could also attach to ions property particulars of a > PG> certain type which would account for the way they would behave > PG> in membranes of a given sort
> Why would you not attach this to the membrane rather than the ion? It > seems to me that the conductance of a membrane is a mathematical > relationship between voltage and current for a given type of ion over > a given type of membrane. Conductance is a property of two things not > one.
I'm saying two things have distinct properties, but there is no one property of two things. They each have these properties in relation to the other, of course.
> PG> So, the treatment of the quantitaive aspects of conductance will > PG> be similar to that of length or weight say.
> PG> does that make sense?
> Not entirely. Mass can be specified as force over acceleration -- at > least when the observer has a static frame of reference, although not > in general. This is independent of the entity providing the > force. Likewise length which is dependent only on the entity (although > again not in general).
> Conductance is more similar to force producing a velocity against a > friction. It's dependent on the entity being pushed and the entity > causing the friction. And as I said previously, this is all dependent > on the velocity of the entity being pushed -- in general, these > measurements are taken at the linear phase, but it breaks down > eventually if you go fast enough.
I'm not talking about what produces, but what accounts for. It might be the same thing which does both, but it is another issue.
>>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
PG> Moreover, membranes can have properties of one type in relation PG> to ions of a given type and properties of another type in PG> relation to ions of another type, I meant this too
Well, that's one way of looking at it. It seems to me a lot clearer and more straightforward to consider it to be something which is a feature of both the entities that are interacting. You can have conductance without a media and you can't have it without a charged entity.
PG> you could type it as a conductance in relation to the class of PG> ion say, in order to account for the fact that this is tied PG> (dependent on) ions. The trick is that it is not specifically PG> dependent on particular ions, right? >> >> >> No, I don't think so. You should expect a membrane with for >> example Chloride channels to have a very different conductance to >> chloride ions than you would to phosphotase ions.
PG> I meant, the property of the membrane is not dependent on ion1, PG> ion2, ion3 etc where these are particular ions (not kinds), but PG> is all things being equal the same which ever particular ions of PG> a specific kind (chloride, etc) happen to traverse it.
Difficult to say till you have tested it. But yes, as a first approximation, this would make sense. Of course, it's not dependent on a particular membrane either.
>> Why would you not attach this to the membrane rather than the >> ion? It seems to me that the conductance of a membrane is a >> mathematical relationship between voltage and current for a given >> type of ion over a given type of membrane. Conductance is a >> property of two things not one.
PG> I'm saying two things have distinct properties, but there is no PG> one property of two things. They each have these properties in PG> relation to the other, of course.
Are there not? Mass, length, passage through time, colour. All these are dependant at least two different things.
>> Not entirely. Mass can be specified as force over acceleration -- >> at least when the observer has a static frame of reference, >> although not in general. This is independent of the entity >> providing the force. Likewise length which is dependent only on >> the entity (although again not in general). >> >> Conductance is more similar to force producing a velocity against >> a friction. It's dependent on the entity being pushed and the >> entity causing the friction. And as I said previously, this is >> all dependent on the velocity of the entity being pushed -- in >> general, these measurements are taken at the linear phase, but it >> breaks down eventually if you go fast enough.
PG> I'm not talking about what produces, but what accounts for. It PG> might be the same thing which does both, but it is another PG> issue.
I am not sure that I understand. Conductance is a description of a behaviour as is friction. It seems to me to be dependent on the two entities involved equally and equivalently.
The only difference between two entities in this case (an membrane and an ion) is that in one case we could distinguish between two particular membranes and there is no way of distinguishing between two particular ions. This distinction disappears with friction, at least in general.
So it seems to me that conductance and friction are properties of the system, not the individual members of it.
On 4/16/07, Phillip Lord <phillip.l...@newcastle.ac.uk> wrote:
> >>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
> PG> Moreover, membranes can have properties of one type in relation > PG> to ions of a given type and properties of another type in > PG> relation to ions of another type, I meant this too
> Well, that's one way of looking at it. It seems to me a lot clearer > and more straightforward to consider it to be something which is a > feature of both the entities that are interacting. You can have > conductance without a media and you can't have it without a charged > entity.
There is indeed interaction and the properties in question are realized in the interaction.
in short and rough approximation probably, you need to dissociate the scientific concept from what is happening
what is happening is that a fluid goes through a pipe. That happening is actually a processual (occurrent). That entity involves (at least) two things: fluid and pipe.
In that occurrence, in this process, are realized two things:
the tendency/disposition/property-of-sort of the pipe to allow in a certain way for the fluid in question to travel thorugh it (correlated to type of fluid)
the tendency/... of the fluid (some token thereof) to travel in a ceratin way through the pipe in question (correlated to type of pipe)
Both these properties are relational, but the property of the pipe is not a property of the fluid and the property of the fluid is not a property of the pipe.
They are however correlated.
I think it could be possible to make properties of this sort dependent on specific particular (portions of) fluid and speak of a given membrane having numerically distinct but similar properties in relation to different portions of fluid of the same type. This, in order to achieve the same effect as saying that the membrane has a single property which is in fact 'generically' dependent (in the sense that they dependent on the type of entities but not any particular instance) on the type of fluid, or on the fluid taken as an object disseminated through teh universe (a property such that any portion of the fluid going through the pipe would do this and that). The approach in term of multiple properties of the pipe each tied to distinct portions of fluid seems very complicated to me, but I admit simplicity is not necessarily the best criterion here.
> PG> you could type it as a conductance in relation to the class of > PG> ion say, in order to account for the fact that this is tied > PG> (dependent on) ions. The trick is that it is not specifically > PG> dependent on particular ions, right?
> >> No, I don't think so. You should expect a membrane with for > >> example Chloride channels to have a very different conductance to > >> chloride ions than you would to phosphotase ions.
> PG> I meant, the property of the membrane is not dependent on ion1, > PG> ion2, ion3 etc where these are particular ions (not kinds), but > PG> is all things being equal the same which ever particular ions of > PG> a specific kind (chloride, etc) happen to traverse it.
> Difficult to say till you have tested it. But yes, as a first > approximation, this would make sense. Of course, it's not dependent on > a particular membrane either.
The property of a given membrane is dependent on this membrane (as it inheres in it). But it is of a type such that there can be other instances of it which inhere and are dependent on other membranes and not the first one.
> >> Why would you not attach this to the membrane rather than the > >> ion? It seems to me that the conductance of a membrane is a > >> mathematical relationship between voltage and current for a given > >> type of ion over a given type of membrane. Conductance is a > >> property of two things not one.
> PG> I'm saying two things have distinct properties, but there is no > PG> one property of two things. They each have these properties in > PG> relation to the other, of course.
> Are there not? Mass, length, passage through time, colour. All these > are dependant at least two different things.
Yes, possibly, but they do not inhere in a multiplicity of instances.
My take on relation property is that: - they inhere in the entity they are the property of - they are dependent on some other entity (in which, however, they do not inhere)
> >> Not entirely. Mass can be specified as force over acceleration -- > >> at least when the observer has a static frame of reference, > >> although not in general. This is independent of the entity > >> providing the force. Likewise length which is dependent only on > >> the entity (although again not in general).
> >> Conductance is more similar to force producing a velocity against > >> a friction. It's dependent on the entity being pushed and the > >> entity causing the friction. And as I said previously, this is > >> all dependent on the velocity of the entity being pushed -- in > >> general, these measurements are taken at the linear phase, but it > >> breaks down eventually if you go fast enough.
> PG> I'm not talking about what produces, but what accounts for. It > PG> might be the same thing which does both, but it is another > PG> issue.
> I am not sure that I understand. Conductance is a description of a > behaviour as is friction. It seems to me to be dependent on the two > entities involved equally and equivalently.
I think conductance is in the end a property of a conductor.
The 'flowability' of a pipe is a property of the pipe. That the pipe may have different "flowability" for different (types of) fluids, just means that it has different properties. Not that these are properties of the fluid as well, although yes, they do depend in one sense or another on the fluids.
> The only difference between two entities in this case (an membrane and > an ion) is that in one case we could distinguish between two > particular membranes and there is no way of distinguishing between two > particular ions. This distinction disappears with friction, at least > in general.
> So it seems to me that conductance and friction are properties of the > system, not the individual members of it.
What about saying they are systemic properties of one member of the system?
But being a property of a system is not what you haev suggested before. You have suggested that conductance was a property of all of the elements.
If the system is something above more than its member, it is not the same to say that a property is one which belongs to the system and to say that it is one which belongs to all the elements of the system.
Actually, if you take systems as objects of sort in their own rights and wish to say that you are speaking of a property as one of the system's properties, I might not argue against this. I'm not too sure what I think of it but it is not as problematic as speaking of a shared property of all members.
One of the difficulty here is that the members of the system change and arguably the system as well. Ions once they've travelled one membrane might not be the same which will travel through it at a later time. But the property you are lookig at would remain the same. The membrane would remain the same. For this reason, it seems to me to make more sense of my suggestion.
As you can imagine I am following your discussion with great interest, since it is directly connected to our recurring discussion about the existence of dependent entities that inhere in two (or more) other entities. I tend to agree with Phillip, but I try to understand your position. Thus two short question:
> Both these properties are relational, but the property of the pipe is > not a property of the fluid and the property of the fluid is not a > property of the pipe.
[...]
> My take on relation property is that: > - they inhere in the entity they are the property of > - they are dependent on some other entity (in which, however, they do > not inhere)
I use "relational property" as a name for properties that inhere in two or more entities -- but that's obviously not intended. Now I try to understand what you mean 'relational property'. Does the following capture it? A relational property is a property that inheres on one entity and depends on at least one other entity?
I am surprised that you believe that the property of the pipe depends on the fluid. Assume you have two fluids A and B which behave exactly the same with respect to flowing through pipes, but are different in some other respect. (E.g. assume that A is water with molecule mass of 19 where the additional neutron is attached to one hydrogen atom, while B is water with molecule mass of 19 where the additional neutron is attached to the oxygen atom.) It seems to me that according to your position if the pipe should have only one tendency/disposition, namely the disposition to interact with ALL fluids that have some distinct properties (a certain mass, a certain chemical structure, ...) in a given way. (By the way, these properties are not realizable entities themselves.) So my second question is: Do you believe that there are two properties that inhere in the pipe --one tendency/disposition to allow in a certain way for fluid A to travel through and another for fluid B or is there only one tendency/disposition that applies to all fluids that are similar with respect to 'flow properties'? In the latter case I don't see how the property of the pipe can depend on a particular fluid.
> On 4/16/07, Phillip Lord <phillip.l...@newcastle.ac.uk> wrote:
>>>>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
>> PG> Moreover, membranes can have properties of one type in relation >> PG> to ions of a given type and properties of another type in >> PG> relation to ions of another type, I meant this too
>> Well, that's one way of looking at it. It seems to me a lot clearer >> and more straightforward to consider it to be something which is a >> feature of both the entities that are interacting. You can have >> conductance without a media and you can't have it without a charged >> entity.
> There is indeed interaction and the properties in question are > realized in the interaction.
> in short and rough approximation probably, you need to dissociate the > scientific concept from what is happening
> what is happening is that a fluid goes through a pipe. That happening > is actually a processual (occurrent). That entity involves (at least) > two things: fluid and pipe.
> In that occurrence, in this process, are realized two things:
> the tendency/disposition/property-of-sort of the pipe to allow in a > certain way for the fluid in question to travel thorugh it (correlated > to type of fluid)
> the tendency/... of the fluid (some token thereof) to travel in a > ceratin way through the pipe in question (correlated to type of pipe)
> Both these properties are relational, but the property of the pipe is > not a property of the fluid and the property of the fluid is not a > property of the pipe.
> They are however correlated.
> I think it could be possible to make properties of this sort dependent > on specific particular (portions of) fluid and speak of a given > membrane having numerically distinct but similar properties in > relation to different portions of fluid of the same type. This, in > order to achieve the same effect as saying that the membrane has a > single property which is in fact 'generically' dependent (in the sense > that they dependent on the type of entities but not any particular > instance) on the type of fluid, or on the fluid taken as an object > disseminated through teh universe (a property such that any portion of > the fluid going through the pipe would do this and that). The approach > in term of multiple properties of the pipe each tied to distinct > portions of fluid seems very complicated to me, but I admit simplicity > is not necessarily the best criterion here.
>> PG> you could type it as a conductance in relation to the class of >> PG> ion say, in order to account for the fact that this is tied >> PG> (dependent on) ions. The trick is that it is not specifically >> PG> dependent on particular ions, right?
>>>> No, I don't think so. You should expect a membrane with for >>>> example Chloride channels to have a very different conductance to >>>> chloride ions than you would to phosphotase ions.
>> PG> I meant, the property of the membrane is not dependent on ion1, >> PG> ion2, ion3 etc where these are particular ions (not kinds), but >> PG> is all things being equal the same which ever particular ions of >> PG> a specific kind (chloride, etc) happen to traverse it.
>> Difficult to say till you have tested it. But yes, as a first >> approximation, this would make sense. Of course, it's not dependent on >> a particular membrane either.
> The property of a given membrane is dependent on this membrane (as it > inheres in it). But it is of a type such that there can be other > instances of it which inhere and are dependent on other membranes and > not the first one.
>>>> Why would you not attach this to the membrane rather than the >>>> ion? It seems to me that the conductance of a membrane is a >>>> mathematical relationship between voltage and current for a given >>>> type of ion over a given type of membrane. Conductance is a >>>> property of two things not one.
>> PG> I'm saying two things have distinct properties, but there is no >> PG> one property of two things. They each have these properties in >> PG> relation to the other, of course.
>> Are there not? Mass, length, passage through time, colour. All these >> are dependant at least two different things.
> Yes, possibly, but they do not inhere in a multiplicity of instances.
> My take on relation property is that: > - they inhere in the entity they are the property of > - they are dependent on some other entity (in which, however, they do > not inhere)
>>>> Not entirely. Mass can be specified as force over acceleration -- >>>> at least when the observer has a static frame of reference, >>>> although not in general. This is independent of the entity >>>> providing the force. Likewise length which is dependent only on >>>> the entity (although again not in general).
>>>> Conductance is more similar to force producing a velocity against >>>> a friction. It's dependent on the entity being pushed and the >>>> entity causing the friction. And as I said previously, this is >>>> all dependent on the velocity of the entity being pushed -- in >>>> general, these measurements are taken at the linear phase, but it >>>> breaks down eventually if you go fast enough.
>> PG> I'm not talking about what produces, but what accounts for. It >> PG> might be the same thing which does both, but it is another >> PG> issue.
>> I am not sure that I understand. Conductance is a description of a >> behaviour as is friction. It seems to me to be dependent on the two >> entities involved equally and equivalently.
> I think conductance is in the end a property of a conductor.
> The 'flowability' of a pipe is a property of the pipe. That the pipe > may have different "flowability" for different (types of) fluids, just > means that it has different properties. Not that these are properties > of the fluid as well, although yes, they do depend in one sense or > another on the fluids.
>> The only difference between two entities in this case (an membrane and >> an ion) is that in one case we could distinguish between two >> particular membranes and there is no way of distinguishing between two >> particular ions. This distinction disappears with friction, at least >> in general.
>> So it seems to me that conductance and friction are properties of the >> system, not the individual members of it.
> What about saying they are systemic properties of one member of the > system?
> But being a property of a system is not what you haev suggested > before. You have suggested that conductance was a property of all of > the elements.
> If the system is something above more than its member, it is not the > same to say that a property is one which belongs to the system and to > say that it is one which belongs to all the elements of the system.
> Actually, if you take systems as objects of sort in their own rights > and wish to say that you are speaking of a property as one of the > system's properties, I might not argue against this. I'm not too sure > what I think of it but it is not as problematic as speaking of a > shared property of all members.
> One of the difficulty here is that the members of the system change > and arguably the system as well. Ions once they've travelled one > membrane might not be the same which will travel through it at a later > time. But the property you are lookig at would remain the same. The > membrane would remain the same. For this reason, it seems to me to > make more sense of my suggestion.
Folks might want to look back at my post to this thread.
I'm assuming since there was no response to my post, either it was not considered useful or it was incomprehensible. I'd be glad to follow-up with anyone who has a specific comment or question.
The discussion that's transpired over the last day is very helpful and gradually converging on a way to use BFO to improve the current NeuronDB.owl representation.
Having said that, my post gives a description of where the NeuronDB content derives from and can help to better inform the analysis of this issue. In the end, regardless of the current state of the NeuronDB OWL representation - which may or may not be either an optimal representation of the underlying NeuronDB info in OWL - or an optimal mapping to BFO - this representation has as its starting point the model data in NeuronDB. That is the content the OWL file must reflect, as that content is NOT going to change the way it represents the relevant continuants and occurents in any appreciable manner.
>>>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
> PG> so conductance of a membrabe is a property particular which > PG> inheres in that membrane
> PG> I think this would suffice
> PG> you could type it as a conductance in relation to the class of > PG> ion say, in order to account for the fact that this is tied > PG> (dependent on) ions. The trick is that it is not specifically > PG> dependent on particular ions, right?
> No, I don't think so. You should expect a membrane with for example > Chloride channels to have a very different conductance to chloride > ions than you would to phosphotase ions.
> PG> It sounds then that conductance would fall under disposition or > PG> tendencies in BFO
> PG> Now, you could also attach to ions property particulars of a > PG> certain type which would account for the way they would behave > PG> in membranes of a given sort
> Why would you not attach this to the membrane rather than the ion? It > seems to me that the conductance of a membrane is a mathematical > relationship between voltage and current for a given type of ion over > a given type of membrane. Conductance is a property of two things not > one.
> PG> So, the treatment of the quantitaive aspects of conductance will > PG> be similar to that of length or weight say.
> PG> does that make sense?
> Not entirely. Mass can be specified as force over acceleration -- at > least when the observer has a static frame of reference, although not > in general. This is independent of the entity providing the > force. Likewise length which is dependent only on the entity (although > again not in general).
> Conductance is more similar to force producing a velocity against a > friction. It's dependent on the entity being pushed and the entity > causing the friction. And as I said previously, this is all dependent > on the velocity of the entity being pushed -- in general, these > measurements are taken at the linear phase, but it breaks down > eventually if you go fast enough.
> Phil
Bill Bug Senior Research Analyst/Ontological Engineer
Laboratory for Bioimaging & Anatomical Informatics www.neuroterrain.org Department of Neurobiology & Anatomy Drexel University College of Medicine 2900 Queen Lane Philadelphia, PA 19129 215 991 8430 (ph) 610 457 0443 (mobile) 215 843 9367 (fax)
Please Note: I now have a new email - William....@DrexelMed.edu
> bulk --> group/population of object --> aggregate
> (and, although not specified by BFO but if needed to tidy up things: a > specialzation of parthood (member in group/population))
> not enought for this?
Perhaps this is sufficient. What would the relationship be between water (the substance) and water (a mass of molecules)? At some point, water stops behaving (within observational error) like a substance and starts behaving like it is made up of discrete particles. It is not as if the water molecules are water-molecule-sized bits of water-substance.
We can say lots of usefull things about water in terms of viewing it as a substance, and lots of questions that would be intractable at the level of a particulate model are tractable at the substance level - for example, fluid dynamics using equations rather than tracing flows of billions of individual molecules in a discrete-time-state simulator. But, the view of water as a substance is only valid within certain scale limits, hence homeopathy.
It's not as if substance and particle are the only two models of water as physical stuff. There's also the atomic/electron high-school view of water as an oxygen with two hydrogens and the ability to hydrogen-bond. Or there's the quantum-soup version where we can describe fancy chemistry. I guess what I'm trying to get at in my groping way is to ask how we can talk about water in each of these ways and state how these different ways are related to each other, and what the limitations of each kind of description of water are, so that we don't apply a way of talking about water to an inappropriate situation and therefore avoid reaching some classes of wrong conclusions. To be more general, what kinds of relationships and concepts would we need to capture this kind of knowledge?
>>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
PG> what is happening is that a fluid goes through a pipe. That PG> happening is actually a processual (occurrent). That entity PG> involves (at least) two things: fluid and pipe.
PG> In that occurrence, in this process, are realized two things:
PG> the tendency/disposition/property-of-sort of the pipe to allow PG> in a certain way for the fluid in question to travel thorugh it PG> (correlated to type of fluid)
PG> the tendency/... of the fluid (some token thereof) to travel in PG> a ceratin way through the pipe in question (correlated to type PG> of pipe)
PG> Both these properties are relational, but the property of the PG> pipe is not a property of the fluid and the property of the PG> fluid is not a property of the pipe.
Again, I don't think that this makes any sense. Let me recap.
- the ability of a (kind of) pipe to let a (kind of) fluid through it is the property of a pipe - the ability of a (kind of) fluid to flow through a (kind of) pipe is a property of the fluid - these two properties are different - but always numerically equal.
PG> I meant, the property of the membrane is not dependent on ion1, PG> ion2, ion3 etc where these are particular ions (not kinds), but PG> is all things being equal the same which ever particular ions of PG> a specific kind (chloride, etc) happen to traverse it. >> >> Difficult to say till you have tested it. But yes, as a first >> approximation, this would make sense. Of course, it's not >> dependent on a particular membrane either.
PG> The property of a given membrane is dependent on this membrane PG> (as it inheres in it). But it is of a type such that there can PG> be other instances of it which inhere and are dependent on other PG> membranes and not the first one.
Maybe. I think that you are using your conclusions to justify themselves. I don't think that conductance inheres in the membrane and is not a property of the membrane alone.
PG> I'm saying two things have distinct properties, but there is no PG> one property of two things. They each have these properties in PG> relation to the other, of course. >> >> Are there not? Mass, length, passage through time, colour. All >> these are dependant at least two different things.
PG> Yes, possibly, but they do not inhere in a multiplicity of PG> instances.
Yes, they do. The mass (length, colour etc) of an entity inheres in the entity and the thing that is observing it.
PG> My take on relation property is that: PG> - they inhere in the entity they are the property of PG> - they are dependent on some other entity (in which, however, PG> they do PG> not inhere)
Okay, so in which case you need a new kind of property which is symmetric, or a principled mechanism for determining which way the asymmetry works.
>> I am not sure that I understand. Conductance is a description of >> a behaviour as is friction. It seems to me to be dependent on the >> two entities involved equally and equivalently.
PG> I think conductance is in the end a property of a conductor
I think that it is not. Conductance has no meaning without a thing conducted. You can abstract away from this by having a standard entity conducted. This is how it works for electrical components -- a 200ohm resistor has 200ohm resistance to electrons. Things would be different if you tried to conductor chloride ions down it.
>> So it seems to me that conductance and friction are properties of >> the system, not the individual members of it.
PG> What about saying they are systemic properties of one member of PG> the system?
PG> But being a property of a system is not what you haev suggested PG> before. You have suggested that conductance was a property of PG> all of the elements.
PG> If the system is something above more than its member, it is not PG> the same to say that a property is one which belongs to the PG> system and to say that it is one which belongs to all the PG> elements of the system.
Ah, my apologies. Perhaps my English was not as clear as it should be. I said "conductance is a property of two things not one". By which I meant, two things considered together at the same time. I was probably not clear enough in distinguishing between the two interpretations that you've pointed out.
PG> One of the difficulty here is that the members of the system PG> change and arguably the system as well. Ions once they've PG> travelled one membrane might not be the same which will travel PG> through it at a later time.
As I said before, you can't tell. How can you tell between one ion and another?
PG> But the property you are lookig at would remain the same. The PG> membrane would remain the same. For this reason, it seems to me PG> to make more sense of my suggestion.
You can flow the same piece of water through multiple pipes. I am not sure what difference this makes.
FN> So my second question is: Do you believe that there are two FN> properties that inhere in the pipe --one tendency/disposition to FN> allow in a certain way for fluid A to travel through and another FN> for fluid B or is there only one tendency/disposition that FN> applies to all fluids that are similar with respect to 'flow FN> properties'? In the latter case I don't see how the property of FN> the pipe can depend on a particular fluid.
It's worse than this. The pipe will behave differently at different temperatures and pressures. So in the end, to describe the pipe it would have a multitude of properties for any possible different fluid (or indeed solid) and at lots of different temperatures.
Alternatively, you can consider fluid conductance to be a property of the system. Then the system has one property, the pipe and fluid none (with respect to conductance). Course, you have a potentially large number of systems, but it still seems simpler to me.
sorry for the lag, it took me an enormous amount of energy to refrain from cribbling the reply with anti german slur,
> > Both these properties are relational, but the property of the pipe is > > not a property of the fluid and the property of the fluid is not a > > property of the pipe.
> [...]
> > My take on relation property is that: > > - they inhere in the entity they are the property of > > - they are dependent on some other entity (in which, however, they do > > not inhere)
> I use "relational property" as a name for properties that inhere in two > or more entities -- but that's obviously not intended.
Right you are, I don't mean multiply inhering
> Now I try to > understand what you mean 'relational property'. Does the following > capture it? A relational property is a property that inheres on one > entity and depends on at least one other entity?
Yes, that is just what I meant.
> I am surprised that you believe that the property of the pipe depends > on the fluid.
Hold on... I'm confused by the type token ambiguity in your sentence my friend.
> Assume you have two fluids A and B which behave exactly > the same with respect to flowing through pipes, but are different in > some other respect. (E.g. assume that A is water with molecule mass of > 19 where the additional neutron is attached to one hydrogen atom, while > B is water with molecule mass of 19 where the additional neutron is > attached to the oxygen atom.)
I assume
A subclass_of Fluid B subclass_of Fluid not (A = B)
> It seems to me that according to your > position if the pipe should have only one tendency/disposition, namely > the disposition to interact with ALL fluids that have some distinct > properties (a certain mass, a certain chemical structure, ...) in a > given way. (By the way, these properties are not realizable entities > themselves.)
Carefull here
All I said rested on the assumption that the two types of fluids were distincts. I assume this notion. Whether A and B have similar flowing properties, if they are distinct then it seems to me there ought to be distinct even if somewhat similar properties of the pipe.
The risk here is of going from the assumed distinction between A and B as two different types of fluid to an indistinction of A and B as having similar "fluidity properties" for lack of better term... your emphasis on such similar is close to suggests that you are now defining types of fluids based on their "fluidity property"
Actually, I tried merely to describe a position according to which you could have:
(1) not (Disp001 = Disp002) (2) Disp001 inst_of Disposition (3) Disp001 inst_of Disposition (4) Disp001 inheres_in Pipe001 (5) Disp002 inheres_in Pipe001 (6) Disp001 depends_on A (7) Disp002 depends_on B
where you interpret (6) and (7) in a way or another
and I also meant to suggest that it is credible that
I'm not sure but I like to think I tried to leave open the possiblity which you describe, more likely I meant to say it is credible that the position you describe is false, i.e. in general it is more credible that for truly distinct kinds of fluids there will be truly distinct properties in the same pipe.
Moreover, I meant that it is more credible that for two instances of A, a1 and a2, there are no two properties of the pipe such that p1 depends on a1 and p2 depends on a2 but just one.
> So my second question is: Do you believe that there are > two properties that inhere in the pipe --one tendency/disposition to > allow in a certain way for fluid A to travel through and another for > fluid B or is there only one tendency/disposition that applies to all > fluids that are similar with respect to 'flow properties'?
I'm just suggesting one possible take on this which seems to be the simplest and most consistent with BFO of the ways of approaching the issue.
It is more credible that pipes have different properties in relation to different true types of fluids.
You describe two fluids which have similar flowing properties, but you do not suggest that there is a single property which is that of both fluids. The same goes for the pipe, it has two similar properties but they are numerically distinct and even do not belmong to the same type, as long as the type of fluids are genuinely distinct.
> In the > latter case I don't see how the property of the pipe can depend on a > particular fluid.
Well, if we assume that A and B are two distinct types of fluids, despite all their similarity, if we index dispositions of pipes to a type of fluid, then the pipe has two disposition, one with resepct to each fluid, although these might be similar.
If we define a type of fluid as having a particular fluidity property and allegeldy distinct A and B have that property (os similar properties) then it sounds more like there is one type of fluid only, call it AuB, in which case the pipe has only one property which depends on AuB.
> > On 4/16/07, Phillip Lord <phillip.l...@newcastle.ac.uk> wrote:
> >>>>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
> >> PG> Moreover, membranes can have properties of one type in relation > >> PG> to ions of a given type and properties of another type in > >> PG> relation to ions of another type, I meant this too
> >> Well, that's one way of looking at it. It seems to me a lot clearer > >> and more straightforward to consider it to be something which is a > >> feature of both the entities that are interacting. You can have > >> conductance without a media and you can't have it without a charged > >> entity.
> > There is indeed interaction and the properties in question are > > realized in the interaction.
> > in short and rough approximation probably, you need to dissociate the > > scientific concept from what is happening
> > what is happening is that a fluid goes through a pipe. That happening > > is actually a processual (occurrent). That entity involves (at least) > > two things: fluid and pipe.
> > In that occurrence, in this process, are realized two things:
> > the tendency/disposition/property-of-sort of the pipe to allow in a > > certain way for the fluid in question to travel thorugh it (correlated > > to type of fluid)
> > the tendency/... of the fluid (some token thereof) to travel in a > > ceratin way through the pipe in question (correlated to type of pipe)
> > Both these properties are relational, but the property of the pipe is > > not a property of the fluid and the property of the fluid is not a > > property of the pipe.
> > They are however correlated.
> > I think it could be possible to make properties of this sort dependent > > on specific particular (portions of) fluid and speak of a given > > membrane having numerically distinct but similar properties in > > relation to different portions of fluid of the same type. This, in > > order to achieve the same effect as saying that the membrane has a > > single property which is in fact 'generically' dependent (in the sense > > that they dependent on the type of entities but not any particular > > instance) on the type of fluid, or on the fluid taken as an object > > disseminated through teh universe (a property such that any portion of > > the fluid going through the pipe would do this and that). The approach > > in term of multiple properties of the pipe each tied to distinct > > portions of fluid seems very complicated to me, but I admit simplicity > > is not necessarily the best criterion here.
> >> PG> you could type it as a conductance in relation to the class of > >> PG> ion say, in order to account for the fact that this is tied > >> PG> (dependent on) ions. The trick is that it is not specifically > >> PG> dependent on particular ions, right?
> >>>> No, I don't think so. You should expect a membrane with for > >>>> example Chloride channels to have a very different conductance to > >>>> chloride ions than you would to phosphotase ions.
> >> PG> I meant, the property of the membrane is not dependent on ion1, > >> PG> ion2, ion3 etc where these are particular ions (not kinds), but > >> PG> is all things being equal the same which ever particular ions of > >> PG> a specific kind (chloride, etc) happen to traverse it.
> >> Difficult to say till you have tested it. But yes, as a first > >> approximation, this would make sense. Of course, it's not dependent on > >> a particular membrane either.
> > The property of a given membrane is dependent on this membrane (as it > > inheres in it). But it is of a type such that there can be other > > instances of it which inhere and are dependent on other membranes and > > not the first one.
> >>>> Why would you not attach this to the membrane rather than the > >>>> ion? It seems to me that the conductance of a membrane is a > >>>> mathematical relationship between voltage and current for a given > >>>> type of ion over a given type of membrane. Conductance is a > >>>> property of two things not one.
> >> PG> I'm saying two things have distinct properties, but there is no > >> PG> one property of two things. They each have these properties in > >> PG> relation to the other, of course.
> >> Are there not? Mass, length, passage through time, colour. All these > >> are dependant at least two different things.
> > Yes, possibly, but they do not inhere in a multiplicity of instances.
> > My take on relation property is that: > > - they inhere in the entity they are the property of > > - they are dependent on some other entity (in which, however, they do > > not inhere)
> >>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
> PG> what is happening is that a fluid goes through a pipe. That > PG> happening is actually a processual (occurrent). That entity > PG> involves (at least) two things: fluid and pipe.
> PG> In that occurrence, in this process, are realized two things:
> PG> the tendency/disposition/property-of-sort of the pipe to allow > PG> in a certain way for the fluid in question to travel thorugh it > PG> (correlated to type of fluid)
> PG> the tendency/... of the fluid (some token thereof) to travel in > PG> a ceratin way through the pipe in question (correlated to type > PG> of pipe)
> PG> Both these properties are relational, but the property of the > PG> pipe is not a property of the fluid and the property of the > PG> fluid is not a property of the pipe.
> Again, I don't think that this makes any sense. Let me recap.
> - the ability of a (kind of) pipe to let a (kind of) fluid through it > is the property of a pipe > - the ability of a (kind of) fluid to flow through a (kind of) pipe > is a property of the fluid > - these two properties are different > - but always numerically equal.
I'm not sure I understand "numerically equal"
Suppose we have:
a has_value "Aunit" b has_value "Bunit"
by "a and b are numerically equal", do you mean:
(1) Aunit = Bunit
or
(2) a = b ?
I use the phrase "a and b are numerically equal" to mean (2). And the last clause you state is definitely nothing I meant to suggest, on the contrary.
If you mean (1), I am not sure the last clause makes sense either, as it is credible that if a and b are the properties in question and they are quantifiable they might not even be given the same units. Also I don't think I have suggested so far that the properties in question had the same value, because I haven't assumed that if they were quantifiable they would be comparable, if they exist, they are correlated and so are their value, that's one thing, but they are probably not numerically identical in neither of the (1) and (2) senses above.
> PG> I meant, the property of the membrane is not dependent on ion1, > PG> ion2, ion3 etc where these are particular ions (not kinds), but > PG> is all things being equal the same which ever particular ions of > PG> a specific kind (chloride, etc) happen to traverse it.
> >> Difficult to say till you have tested it. But yes, as a first > >> approximation, this would make sense. Of course, it's not > >> dependent on a particular membrane either.
> PG> The property of a given membrane is dependent on this membrane > PG> (as it inheres in it). But it is of a type such that there can > PG> be other instances of it which inhere and are dependent on other > PG> membranes and not the first one.
> Maybe. I think that you are using your conclusions to justify > themselves. I don't think that conductance inheres in the membrane and > is not a property of the membrane alone.
wait, I am trying to *describe* a way in which to parse the example in BFO, but OK, it's possible I misunderstood the example or your own understanding of the example
The property of a membrane is dependent on the membrane is what BFO says
I'm suggesting a representation according to which the conductance of a membrane is a property of the membrane
You disagree with this view but I am merely telling you that in BFO there is no property which is a property of more than one thing -- I am not considering the recent addition of so-called generically dependent continuants although to my understanding a property of more than one thing in the sense you are aiming at would not be a generically dependent continuant of BFO
> PG> I'm saying two things have distinct properties, but there is no > PG> one property of two things. They each have these properties in > PG> relation to the other, of course.
> >> Are there not? Mass, length, passage through time, colour. All > >> these are dependant at least two different things.
> PG> Yes, possibly, but they do not inhere in a multiplicity of > PG> instances.
> Yes, they do. The mass (length, colour etc) of an entity inheres in > the entity and the thing that is observing it.
wait, the major assumption here is that this is absurd, there might be things inhering in an observer but the length of of a rod is something which exists whether observed or not, so it is not even dependent on an observer
by dependent, I mean "existentially dependent", I mean to say that the rod has a property which is an entity,
I do not mean to say anything about the the quantity which can be associated with this entity, it is credible that the measurement of a rod is something that depends on (i.e. is functionally related to) various conditions of measurement, but the length of a rod (in other, though not quite synonymous, words, the fact that the rod has a length) is not existentially dependent on these conditions...
in both case, it is actually not the same kind of dependence we are talking about (might be arguable, but it would be in the sense that something that shows in measurement is an entity which is dependent in the way the length is, not in the sense that the length as an entity is dependent in the way that measurement values or what have you aredependent on conditions or observers)
are we clear about this assumption?
> PG> My take on relation property is that: > PG> - they inhere in the entity they are the property of > PG> - they are dependent on some other entity (in which, however, > PG> they do > PG> not inhere)
> Okay, so in which case you need a new kind of property which is > symmetric, or a principled mechanism for determining which way the > asymmetry works.
not sure I follow
> >> I am not sure that I understand. Conductance is a description of > >> a behaviour as is friction. It seems to me to be dependent on the > >> two entities involved equally and equivalently.
> PG> I think conductance is in the end a property of a conductor
> I think that it is not. Conductance has no meaning without a thing > conducted. You can abstract away from this by having a standard > entity conducted. This is how it works for electrical components -- a > 200ohm resistor has 200ohm resistance to electrons. Things would be > different if you tried to conductor chloride ions down it.
Did I really say anything contradicting what you are saying here?
> >> So it seems to me that conductance and friction are properties of > >> the system, not the individual members of it.
> PG> What about saying they are systemic properties of one member of > PG> the system?
> PG> But being a property of a system is not what you haev suggested > PG> before. You have suggested that conductance was a property of > PG> all of the elements.
> PG> If the system is something above more than its member, it is not > PG> the same to say that a property is one which belongs to the > PG> system and to say that it is one which belongs to all the > PG> elements of the system.
> Ah, my apologies. Perhaps my English was not as clear as it should > be.
Thanks for your understanding...
> I said "conductance is a property of two things not one".
And I said this is not BFO. I tried to offer you two ways:
1) conductance is a property of the conductor which it has in relation to the conductee or kind of conductee.
2) conductance is the property of the conductor-conductee system, but of neither the conductor nor the conductee alone
I think 1) is more credible, but I am just stating two possible approaches and if I have seemed to suggest that I was holding the truth about which is right, maybe I should just back off and suspend any judgement.
The above "So it seems to me that conductance and friction are properties of the system, not the individual members of it.. " seems to square with 2)
> By which > I meant, two things considered together at the same time.
I do not understand this I am afraid. I tought first you were speaking of a property of two things, and second of a property of a thrid thing which is composed of the first two.
> I was > probably not clear enough in distinguishing between the two > interpretations that you've pointed out.
I didn't mean to berate you if that's the worry
> PG> One of the difficulty here is that the members of the system > PG> change and arguably the system as well. Ions once they've > PG> travelled one membrane might not be the same which will travel > PG> through it at a later time.
> As I said before, you can't tell. How can you tell between one ion and > another?
The point is that I don't have to, though
> PG> But the property you are lookig at would remain the same. The > PG> membrane would remain the same. For this reason, it seems to me > PG> to make more sense of my suggestion.
> You can flow the same piece of water through multiple pipes. I am not > sure what difference this makes.
me neither,
I meant to say that the property is dependent on the type or on every piece of water if you will, but not on any specific piece of water
> At some point, water stops > behaving (within observational error) like a substance and starts behaving > like it is made up of discrete particles. It is not as if the water molecules > are water-molecule-sized bits of water-substance.
> We can say lots of usefull things about water in terms of viewing it as a > substance, and lots of questions that would be intractable at the level of a > particulate model are tractable at the substance level - for example, fluid > dynamics using equations rather than tracing flows of billions of individual > molecules in a discrete-time-state simulator. But, the view of water as a > substance is only valid within certain scale limits, hence homeopathy.
> It's not as if substance and particle are the only two models of water as > physical stuff. There's also the atomic/electron high-school view of water as > an oxygen with two hydrogens and the ability to hydrogen-bond. Or there's the > quantum-soup version where we can describe fancy chemistry. I guess what I'm > trying to get at in my groping way is to ask how we can talk about water in > each of these ways and state how these different ways are related to each > other, and what the limitations of each kind of description of water are, so > that we don't apply a way of talking about water to an inappropriate > situation and therefore avoid reaching some classes of wrong conclusions. To > be more general, what kinds of relationships and concepts would we need to > capture this kind of knowledge?
I would suggest you look at the OpenCyc ontology
in particluar, look for
ObjectType (unfortunately, there is no link to StuffType which might have been helpful) ExistingObjectType ExistingStuffType granuleOfSpatialStuff
> On 4/17/07, Phillip Lord <phillip.l...@newcastle.ac.uk> wrote:
>> PG> I think conductance is in the end a property of a conductor
>> PL> I think that it is not. Conductance has no meaning without a thing >> conducted. You can abstract away from this by having a standard >> entity conducted. This is how it works for electrical components -- a >> 200ohm resistor has 200ohm resistance to electrons. Things would be >> different if you tried to conductor chloride ions down it.
> PG> Did I really say anything contradicting what you are saying here?
First, electrical conductance (G) is the inverse of electrical resistance (R), i.e., G = 1/R. Resistance is measured by 'ohm', and conductance is measured by 'siemens'. Let me now talk about resistance. The fact that there is a *relational* and *dispositional description* of a property of certain resistor T is quite consistent with the fact *what is described* inheres in T. Here is a possible dispositional description:
"if an electric current (a stream of electrons) passes through T, then T makes resistance (of 200 ohm)".
But this can be regarded as being due to the microstructure of T that is a property of T only.
>>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
>> Again, I don't think that this makes any sense. Let me recap. >> >> - the ability of a (kind of) pipe to let a (kind of) fluid >> through it >> is the property of a pipe >> - the ability of a (kind of) fluid to flow through a (kind of) >> pipe >> is a property of the fluid >> - these two properties are different >> - but always numerically equal. >>
PG> I'm not sure I understand "numerically equal"
PG> Suppose we have:
PG> a has_value "Aunit" b has_value "Bunit"
PG> by "a and b are numerically equal", do you mean:
PG> (1) Aunit = Bunit
PG> or
PG> (2) a = b ?
PG> if they exist, they are correlated and so are their value, PG> that's one thing, but they are probably not numerically PG> identical in neither of the (1) and (2) senses above.
The ability of a particular pipe to allow a particular fluid through it, seems to me, to be the same (or the inverse of) as the ability of that particular fluid to pass through that particular pipe.
But, to me, it seems that you are saying the former is the property of a pipe, the latter the property of the fluid.
>> Maybe. I think that you are using your conclusions to justify >> themselves. I don't think that conductance inheres in the >> membrane and is not a property of the membrane alone.
PG> wait, I am trying to *describe* a way in which to parse the PG> example in BFO, but OK, it's possible I misunderstood the PG> example or your own understanding of the example
PG> The property of a membrane is dependent on the membrane is what PG> BFO says
PG> I'm suggesting a representation according to which the PG> conductance of a membrane is a property of the membrane
PG> You disagree with this view but I am merely telling you that in PG> BFO there is no property which is a property of more than one PG> thing
But conductance clearly is a property of two things. From which I conclude that either
- BFO cannot represent conductance - BFO can represent conductance but we have to make an arbitrary decision about which of two entities it depends on it inheres in. - BFO should be extended.
PG> Yes, possibly, but they do not inhere in a multiplicity of PG> instances. >> >> Yes, they do. The mass (length, colour etc) of an entity inheres >> in the entity and the thing that is observing it.
PG> wait, the major assumption here is that this is absurd, there PG> might be things inhering in an observer but the length of of a PG> rod is something which exists whether observed or not, so it is PG> not even dependent on an observer
Ah, okay, I was indeed thinking of the value of the length.
Now, obviously, by definition we can't know about the length of a rod (either it's value or it's existence) when there is no observer, so there's probably no point discussing that. But I would agree that all observers would believe the rod to have a length, at least if they could work out a way of distinguishing a length from a width.
So, if we have two observers looking at one rod, does the rod have two lengths or one? If you are stating that the rod has an existential relationship to it's length, then I guess that it can have many different lengths at once.
PG> My take on relation property is that: PG> - they inhere in the entity they are the property of PG> - they are dependent on some other entity (in which, however, PG> they do not inhere) >> >> Okay, so in which case you need a new kind of property which is >> symmetric, or a principled mechanism for determining which way >> the asymmetry works.
PG> not sure I follow
As we have said, conductance depends on two things. So, how do we work out which one it inheres in?
>> >> I am not sure that I understand. Conductance is a description >> >> of a behaviour as is friction. It seems to me to be dependent >> >> on the two entities involved equally and equivalently. >> PG> I think conductance is in the end a property of a conductor >> >> I think that it is not. Conductance has no meaning without a >> thing conducted. You can abstract away from this by having a >> standard entity conducted. This is how it works for electrical >> components -- a 200ohm resistor has 200ohm resistance to >> electrons. Things would be different if you tried to conductor >> chloride ions down it. >>
PG> Did I really say anything contradicting what you are saying PG> here?
I think so. You are claiming that conductance is a property of a conductor. Therefore it inheres in the conductor. And, therefore, I think, it does not inhere in the thing conducted. Which I think contradicts my statement.
>> I said "conductance is a property of two things not one".
PG> And I said this is not BFO. I tried to offer you two ways:
PG> 1) conductance is a property of the conductor which it has in PG> relation PG> to the conductee or kind of conductee.
PG> 2) conductance is the property of the conductor-conductee PG> system, but PG> of neither the conductor nor the conductee alone
PG> I think 1) is more credible, but I am just stating two possible PG> approaches and if I have seemed to suggest that I was holding PG> the truth about which is right, maybe I should just back off and PG> suspend any judgement.
PG> The above "So it seems to me that conductance and friction are PG> properties of the system, not the individual members of it.. " PG> seems to square with 2)
Okay, perhaps we are getting into an English problem again here.
I think that "conductance is a property of two things not one" is the same as "conductance in the property of the conductor-conductee system".
But if it is possible to represent this in BFO, it seems the way to go to me.
> >>>>> "PG" == Pierre Grenon <pierregre...@gmail.com> writes:
> >> Again, I don't think that this makes any sense. Let me recap.
> >> - the ability of a (kind of) pipe to let a (kind of) fluid > >> through it > >> is the property of a pipe > >> - the ability of a (kind of) fluid to flow through a (kind of) > >> pipe > >> is a property of the fluid > >> - these two properties are different > >> - but always numerically equal.
> PG> I'm not sure I understand "numerically equal"
> PG> Suppose we have:
> PG> a has_value "Aunit" b has_value "Bunit"
> PG> by "a and b are numerically equal", do you mean:
> PG> (1) Aunit = Bunit
> PG> or
> PG> (2) a = b ?
> PG> if they exist, they are correlated and so are their value, > PG> that's one thing, but they are probably not numerically > PG> identical in neither of the (1) and (2) senses above.
> The ability of a particular pipe to allow a particular fluid through > it, seems to me, to be the same (or the inverse of) as the ability of > that particular fluid to pass through that particular pipe.
I think you still don't get the distintion I am trying to draw here.
Suppose you have a tunnel between France and England T Suppose T has a uniform circular section which measures 2 meters
Suppose the French want to invade England for some obscure reason and commission their German expatriate engineers to design a special vehicle to travel through T, they are working on vehicle of class V
One of the engineering requirement on V is that the height of V vehicles be less than 2 meters
T has the property of allowing for vehicle of height inferior to 2 meters to travel through it, call it p1
if v is a V, v has the property of being able to travel through tunnels which allow for travelling of vehicles of maximal height 2 meters, call it p2
Are you willing to say that there is a single property which is shared by v (and all V class vehicles) and T?
> But, to me, it seems that you are saying the former is the property of > a pipe, the latter the property of the fluid.
Yes,
I am saying that p1 is a property of T which depends on some class to which Vs belong
I am saying that p2 is a property of a v (all Vs have a similar property more accurately but do not let this bother you here) which depends on some class of which T is an instance
> >> Maybe. I think that you are using your conclusions to justify > >> themselves. I don't think that conductance inheres in the > >> membrane and is not a property of the membrane alone.
> PG> wait, I am trying to *describe* a way in which to parse the > PG> example in BFO, but OK, it's possible I misunderstood the > PG> example or your own understanding of the example
> PG> The property of a membrane is dependent on the membrane is what > PG> BFO says
> PG> I'm suggesting a representation according to which the > PG> conductance of a membrane is a property of the membrane
> PG> You disagree with this view but I am merely telling you that in > PG> BFO there is no property which is a property of more than one > PG> thing
> But conductance clearly is a property of two things. From which I > conclude that either
before concluding anything, let's see if conductance is the property of two things
> - BFO cannot represent conductance > - BFO can represent conductance but we have to make an arbitrary > decision about which of two entities it depends on it inheres in. > - BFO should be extended.
> >> Yes, they do. The mass (length, colour etc) of an entity inheres > >> in the entity and the thing that is observing it.
> PG> wait, the major assumption here is that this is absurd, there > PG> might be things inhering in an observer but the length of of a > PG> rod is something which exists whether observed or not, so it is > PG> not even dependent on an observer
> Ah, okay, I was indeed thinking of the value of the length.
ok, can you apply this to the above as well?
> So, if we have two observers looking at one rod, does the rod have two > lengths or one? If you are stating that the rod has an existential > relationship to it's length, then I guess that it can have many > different lengths at once.
The rod has one length and that is all I am interested in, whether people can disagree on the measurement is an entirely different issue.
> PG> My take on relation property is that: > PG> - they inhere in the entity they are the property of > PG> - they are dependent on some other entity (in which, however, > PG> they do not inhere)
> >> Okay, so in which case you need a new kind of property which is > >> symmetric, or a principled mechanism for determining which way > >> the asymmetry works.
> PG> not sure I follow
> As we have said, conductance depends on two things. So, how do we work > out which one it inheres in?
Well, if conductance is the property of a conductor, it should be easy to figure out which is which, conductors do not travel through ions, say?
> >> >> I am not sure that I understand. Conductance is a description > >> >> of a behaviour as is friction. It seems to me to be dependent > >> >> on the two entities involved equally and equivalently.
> PG> I think conductance is in the end a property of a conductor
> >> I think that it is not. Conductance has no meaning without a > >> thing conducted. You can abstract away from this by having a > >> standard entity conducted. This is how it works for electrical > >> components -- a 200ohm resistor has 200ohm resistance to > >> electrons. Things would be different if you tried to conductor > >> chloride ions down it.
> PG> Did I really say anything contradicting what you are saying > PG> here?
> I think so. You are claiming that conductance is a property of a > conductor. Therefore it inheres in the conductor. And, therefore, I > think, it does not inhere in the thing conducted. Which I think > contradicts my statement.
> >> I said "conductance is a property of two things not one".
> PG> And I said this is not BFO. I tried to offer you two ways:
> PG> 1) conductance is a property of the conductor which it has in > PG> relation > PG> to the conductee or kind of conductee.
> PG> 2) conductance is the property of the conductor-conductee > PG> system, but > PG> of neither the conductor nor the conductee alone
> PG> I think 1) is more credible, but I am just stating two possible > PG> approaches and if I have seemed to suggest that I was holding > PG> the truth about which is right, maybe I should just back off and > PG> suspend any judgement.
> PG> The above "So it seems to me that conductance and friction are > PG> properties of the system, not the individual members of it.. " > PG> seems to square with 2)
> Okay, perhaps we are getting into an English problem again here.
I don't think this is a problem with language of the level you seem to have in mind, i rather have the feeling to put it simply that we do not use the word property in the same way.
What I am talking about here is not measurements or values, it is the thing which is measured if you want. That is a property, let's say. All you say here seems to indicate to me that you think of the value as the property.
> I think that "conductance is a property of two things not one" is the > same as "conductance in the property of the conductor-conductee > system".
so for you the phrase "the conductor-conductee system" refers to two things and not one? that is very odd isn't it?
>>>>> "IJ" == Ingvar Johansson <ingvar.johans...@ifomis.uni-saarland.de> writes:
IJ> First, electrical conductance (G) is the inverse of electrical IJ> resistance (R), i.e., G = 1/R. Resistance is measured by 'ohm', IJ> and conductance is measured by 'siemens'. Let me now talk about IJ> resistance.
Yes.
IJ> The fact that there is a *relational* and *dispositional IJ> description* of a property of certain resistor T is quite IJ> consistent with the fact *what is described* inheres in IJ> T.
I'm afraid this paragraph doesn't mean anything to me.
IJ> Here is a possible dispositional description:
IJ> "if an electric current (a stream of electrons) passes through IJ> T, then T makes resistance (of 200 ohm)".
An electric current is not a stream of electrons. It's one way of making a current. Conductance is defined for current not electrons.
IJ> But this can be regarded as being due to the microstructure of T IJ> that is a property of T only.
I think not. It's a property of the microstructure of T, the nature of an electron, and how the two interact.
