neuron_ontology
Pierre Grenon
I've looked very quickly at the ontology
Here again, it is hard to say anything based on labels alone
My impression is that most of the subclasses of neuron_property
deputize for what should be parts of neurons or what should be
chemical entities of sorts which have a role of sort in relation to
neuron
Some lables suggest bfo:dependent:continuants
The should be few properties if this is intended to mean
bfo:dependent:continuant, mostly roles, e.g. receptor / transmitter
alternatively or complementarily there should be
it occurs to me that, in the neuron ontology, "neuron_property" might stand for:
the class which subsumes all classes of entities which have a relation
or another to neurons
is this the case?
pierre
On 4/16/07, Alan Ruttenberg <alanrut...@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
>
> -Alan
>
>
> >
>
Matthias Samwald
weeks now and I share your reservations about the class
"neuron_property". I guess we will remove during further development.
William Bug
Matthias Samwald
requested in the other discussion thread:
http://neuroweb.med.yale.edu/senselab/neuron_ontology-bfo-mapping.owl
As you can see, it is quite simple. Where I was uncertain whether to
map to "object" or "fiat part of object", i chose the more generic
superclass. We also plan to replace the property "has receptor" with
"has part" from the relation ontology.
cheers,
Matthias Samwald
William Bug
| Definition: A realizable entity the manifestation of which is an essentialy end-directed activity of a continuant entity in virtue of that continuant entity being a specific kind of entity in the kind or kinds of contexts that it is made for. |
| Examples: the function of a birth canal to enable transport, the function of the heart in the body to pump blood, the function of reproduction in the transmission of genetic material, the digestive function of the stomach to nutriate the body, the function of a hammer to drive in nails, the function of a computer program to compute mathematical equations, the function of an automobile to provide transportation, the function of a judge in a court of law |
Matthias Samwald
> Obviously, for the mapping to BFO to be really useful, you can't
> have classes sitting as siblings to the BFO root (bfo:entity).
Please take a closer look at the ontology. These classes HAVE been
mapped. However, they are also defined as subclasses of owl:Thing in
the original ontology, and that is why Protege (I assume you are using
it) displays them as both subclasses of certain BFO classes AND as
siblings of the BFO classes.
> The compartments are all
> fiatObjectParts of neuron types, given the way these compartments
> tend to be defined experimentally.
Apical, medial and proximal part are surely fiat parts. But soma, for
example? The soma of a neuron is quite an obvious and distinct
structure in itself.
I have to be honest: I find the distinction of "fiat" and "bona fide"
boundaries highly subjective and don't see how this is compatible with
the realist approach of BFO.
> Information entities such as publication CAN be more problematic,
> depending on what you actually mean by them, but there are examples
> of how to map these as well.
I did not map them, because these classes will probably be removed.
>
> 2) Currents:
> Have you considered mapping the currents (or conductances - NeuronDB
> is essentially modeling the transmitter- and voltage-dependent
> membrane conductances, even though they are given as currents. In
> fact, many of the papers used as evidence for the NeuronDB assertions
> will in fact document conductance change vs. voltage or transmitter
> or some other saline factor such as permeable ions or conductance
> modifier such as Zn++ with NMDA receptor channels). Here's the
> bfo:function defintion and examples:
> Definition: A realizable entity the manifestation of which is an
> essentialy end-directed activity of a continuant entity in virtue of
> that continuant entity being a specific kind of entity in the kind or
> kinds of contexts that it is made for.
I don't think we are are relating the 'currents' or their
corresponding channels to outside source (e.g. sequence databases) at
the moment, so I would prefer not to introduce such additional
constructs without a good reason.
William Bug
Obviously, for the mapping to BFO to be really useful, you can't have classes sitting as siblings to the BFO root (bfo:entity).Please take a closer look at the ontology. These classes HAVE been mapped. However, they are also defined as subclasses of owl:Thing in the original ontology, and that is why Protege (I assume you are using it) displays them as both subclasses of certain BFO classes AND as siblings of the BFO classes.
The compartments are all fiatObjectParts of neuron types, given the way these compartments tend to be defined experimentally.Apical, medial and proximal part are surely fiat parts. But soma, for example? The soma of a neuron is quite an obvious and distinct structure in itself.
I have to be honest: I find the distinction of "fiat" and "bona fide" boundaries highly subjective and don't see how this is compatible with the realist approach of BFO.
Information entities such as publication CAN be more problematic, depending on what you actually mean by them, but there are examples of how to map these as well.I did not map them, because these classes will probably be removed.
2) Currents:Have you considered mapping the currents (or conductances - NeuronDB is essentially modeling the transmitter- and voltage-dependent membrane conductances, even though they are given as currents. In fact, many of the papers used as evidence for the NeuronDB assertions will in fact document conductance change vs. voltage or transmitter or some other saline factor such as permeable ions or conductance modifier such as Zn++ with NMDA receptor channels). Here's the bfo:function defintion and examples:Definition: A realizable entity the manifestation of which is an essentialy end-directed activity of a continuant entity in virtue of that continuant entity being a specific kind of entity in the kind or kinds of contexts that it is made for.I don't think we are are relating the 'currents' or their corresponding channels to outside source (e.g. sequence databases) at the moment
, so I would prefer not to introduce such additional constructs without a good reason.
William Bug
Pierre Grenon
I would say it's not like you want to kill it altogether but trimming
it and considering whether some subclasses ougt to be migrated and
where seems indeed necessary.
thanks for following up!
p
Pierre Grenon
>
> Here is the mapping of the SenseLab NeuronDB OWL ontology to BFO, as
> requested in the other discussion thread:
> http://neuroweb.med.yale.edu/senselab/neuron_ontology-bfo-mapping.owl
>
> As you can see, it is quite simple. Where I was uncertain whether to
> map to "object" or "fiat part of object", i chose the more generic
> superclass.
good thinking
> We also plan to replace the property "has receptor" with
> "has part" from the relation ontology.
why not keeping it and making it a subrelation?
pierre
> cheers,
> Matthias Samwald
>
>
> >
>
Pierre Grenon
outrageously ignoring the question he brought on
I've trimmed the file too
> <?xml version="1.0"?>
> <rdf:RDF
...
> <owl:Ontology rdf:about="">
> >A mapping of the SenseLab NeuronDB OWL ontology to BFO. Work in
> progress.</rdfs:comment>
...
> </owl:Ontology>
> <rdf:Description
> rdf:about="http://neuroweb.med.yale.edu/senselab/neuron_ontology.owl#Neuron">
> <rdfs:subClassOf
> rdf:resource="http://www.ifomis.org/bfo/1.0/snap#Object"/>
This will mean that there are no objects which are parts of a neuron
nor any objects of which a neuron is a part, if you are confortable
with this, fine otherwise your strategy of mapping to
independant:continuant might be a safeguard here, if only
preliminarily, you would not really loose anything provided you have
the right relations at hand and could use the one from RO in
particular
> <owl:disjointWith
> rdf:resource="http://www.ifomis.org/bfo/1.0/snap#Object"/>
This seems like an error, if you think this is a subclass of object
how would it be disjoint with it, did you mean to assert disjointness
with another subclass of Independent:Continuant?
Note that
if you keep Neuron as a specialization of bfo:Object, you don't need
such disjoint assertions
if you make Neuron a self standing specialization of
bfo:independent:continuant, it would make more sense to assert
disjointness with other speicalizations of bfo:independent:continuant
in your ontology, the details might depend on how you have categorzied
this, in some cases bfo could be able to take part of this load
> </rdf:Description>
> <rdf:Description
> rdf:about="http://neuroweb.med.yale.edu/senselab/neuron_ontology.owl#Neuron_Receptor">
> <owl:disjointWith
> rdf:resource="http://www.ifomis.org/bfo/1.0/snap#Object"/>
> </rdf:Description>
So, is what you want something like :
<rdfs:subClassOf
rdfs:resource="http://www.ifomis.org/bfo/1.0/snap#IndependentContinuant"/>
Is this the class of parts of neuron which have a receptor role in
some processual entity of sort?
> <rdf:Description
> rdf:about="http://neuroweb.med.yale.edu/senselab/neuron_ontology.owl#BrainRegion">
> <owl:disjointWith
> rdf:resource="http://www.ifomis.org/bfo/1.0/snap#IndependentContinuant"/>
> </rdf:Description>
Are you sure you mean owl:disjointWith, I am not sure what instances
of this class are supposed to be, parts of brain? sites within the
brain? spatial regions at which parts of the brain are located or
spatial regions at which sites in the brain are located?
Only in the case that you intend spatial regions would you want the
disjointness but you would get it from BFO ad a rdfs:subClassOf
assertion
> <rdf:Description
> rdf:about="http://neuroweb.med.yale.edu/senselab/neuron_ontology.owl#Neuron_Transmitter">
> <owl:disjointWith
> rdf:resource="http://www.ifomis.org/bfo/1.0/snap#Object"/>
> </rdf:Description>
Am I right to suggest that the same comments as concerning
Neuron_Receptor might apply here?
> <rdf:Description
> rdf:about="http://neuroweb.med.yale.edu/senselab/neuron_ontology.owl#Compartment">
> <owl:disjointWith
> rdf:resource="http://www.ifomis.org/bfo/1.0/snap#IndependentContinuant"/>
> </rdf:Description>
I don't know what compartments are, possibly, comments similar to
those on BrainRegion apply here
> <rdf:Description
> rdf:about="http://neuroweb.med.yale.edu/senselab/neuron_ontology.owl#Neuron_Current">
> <owl:disjointWith
> rdf:resource="http://www.ifomis.org/bfo/1.0/span#ProcessualEntity"/>
> </rdf:Description>
> </rdf:RDF>
OK, so this says Neuron_Current are either
- regions of time
- regions of spacetime
or
- continuants
Are these supposed to be the places in the neuron or some part of the
neurological system where ions or other things move or are these
supposed to be ions movements?
> > > WB The compartments are all fiatObjectParts of neuron types, given the way
> > > WB these compartments tend to be defined experimentally.
>
> > MS Apical, medial and proximal part are surely fiat parts. But soma, for
> > example? The soma of a neuron is quite an obvious and distinct structure in
> > itself.
Apply yoour better safe than sorry intuition here, make these all
bfo:independent:continuant
You have asserted for the moment that Neuron is a subclass of
bfo:Object, to be consistent with your claim here you have to make any
class of parts subclasses of bfo:FiatPartOfObject
If you are not confortable with this, at least in some cases map to
the generic class and use the RO relations or the mereological
relation of BFO
but it would seem better to be consistent
> WB Reasonable people can differ on this issue. I know what you are
...
> WB 95% - and for many types of neuron it would be much less than that - it is
> WB still a fiat boundary from my point of view.
having whatever is discussed here (I do not have the fiantest of idea
of what it is) as subclass of the lowest most generic class possible
(i like these tricks) would make the man happy, you could still have
variant specific of your ontology in which things get categorized more
specifically while remaining safe and sound during development
> > MS I have to be honest: I find the distinction of "fiat" and "bona fide"
> > MS boundaries highly subjective and don't see how this is compatible with the
> > MS realist approach of BFO.
I think that is the point of the distinction though (accounting for
some degree of subjectivity in a realist way)
> > WB 2) Currents:
> WB I'd be really interested to hear what Pierre, Matthew, Fabian, and Phil have
> WB to say on this issue.
The issue concerning yurrent is not at all clear to me, waiting to see
more on this...
cheers
pierre
PS: Bill, would you mind switching off html?
Pierre Grenon
> P.S.: Thinking a little more about it, I'm really not certain which of the
> following three possible ways to represent plasma membrane ionic
> currents/conductances in BFO would be preferred:
> a) bfo:Function
Why function? sounds like there's something closer to a disposition, imhu
> b) bfo:ObjectAggregate
> c) bfo:Process
The most credible BFO story in a nutshell would be something along these lines:
There is a membrane (indep:cont) which has the disposition (dep:cont)
of allowing for translocation (process) of ions (indep:cont), this
disposition is realized when translocation of a ion (object?) or a
group of ions (aggregate) occurs
> I don't really know what is both more fitting to the intended use of BFO
> classes, which is the most "real" representation of the underlying entities
> and events, and which will turn out to be the maximally complete while
> minimally complex way to do this in BFO. One would hope the former two -
> preferred BFO use & most "real" representation - would in fact collapse to
> the same thing.
>
> In the end, there will probably be a need to represent all three.
>
> For instance, the I.K+-delayed_rectifier conductance/current in a somatic
> neuronal plasma membrane could be represented as follows:
> a) The function
disposition, sounds like membrane are membrane first and second
membrane which allow for certain processus to occur
> of the somatic plasma membrane to allow K+ to flow down its
> electrochemical gradient in a time-varying manner under conditions where the
> membrane potential is depolarized beyond a threshold level
you forgot mentioning the headhache...
> b) The collection of I.K+-delayed_rectifier ion channels
channels are?
> in the somatic
> plasma membrane (function then being specified for the ion channel object)
> c) The time-varying process whereby K+ ions flow down their electrochemical
> gradient under conditions where the somatic plasma membrane potential is
> depolarized to beyond a threshold level
i'm wondering if it would be the same headhache as above or another headhache...
cheers
pierre
> Cheers,
> Bill
>
>
> On Apr 17, 2007, at 9:40 PM, William Bug wrote:
> Hi Matthias,
>
> Just to be clear, I'm not trying to be argumentative or contrary. I'm just
> trying to help us get clear about what is actually meant by the universal
> definitions in neuron_ontology.owl, because I really want to use them.
>
> Please see below for brief comments.
>
> Cheers,
> Bill
>
>
> On Apr 17, 2007, at 8:59 PM, Matthias Samwald wrote:
>
> Obviously, for the mapping to BFO to be really useful, you can't have
> classes sitting as siblings to the BFO root (bfo:entity).
>
> Please take a closer look at the ontology. These classes HAVE been mapped.
> However, they are also defined as subclasses of owl:Thing in the original
> ontology, and that is why Protege (I assume you are using it) displays them
> as both subclasses of certain BFO classes AND as siblings of the BFO
> classes.Using SWOOP. For some reason, the only neuron_ontology.owl entities
> itself.Reasonable people can differ on this issue. I know what you are
> saying in a textbook sense, but - as I assume you have done - when you are
> about to do physiological recording, looking at cell in an inverted
> microscope using Nomarksy optics (as is the case for many of the articles
> cited by the SenseLab curators to create the database), if you were asked to
> place the electrode at the boundary between axon hillock and soma, with what
> certainty would you be able to place it accurately. Even if the answer were
> 95% - and for many types of neuron it would be much less than that - it is
> still a fiat boundary from my point of view.
>
>
>
> I have to be honest: I find the distinction of "fiat" and "bona fide"
> boundaries highly subjective and don't see how this is compatible with the
> realist approach of BFO.It took me a few times through several of the
> articles on fiat boundry before it started to feel like it made sense. I
> agree it can appear a bit opaque, and only clear examples can make it clear.
> Not ALL examples are easily interpreted in this way.
>
>
>
> Information entities such as publication CAN be more problematic, depending
> on what you actually mean by them, but there are examples of how to map
> these as well.
>
> I did not map them, because these classes will probably be removed.Good idea
Smith, Barry
>I have to be honest: I find the distinction of "fiat" and "bona fide"
>boundaries highly subjective and don't see how this is compatible with
>the realist approach of BFO.
A realist ontology needs to do justice to the fact that there are
some boundaries in physical reality -- for example the surface of
your skin -- and some boundaries which exist only because subjects
put them there, e.g. the boundary of Poland.
Do you accept that there are these two kinds of boundary (and of
course a variety of mixed cases including elements of each)?
BS
Smith, Barry
>boundaries highly subjective and don't see how this is compatible with
>the realist approach of BFO.
A realist ontology needs to do justice to the fact that there are
Cristian Cocos
As far as I know, the laws of Physics are indifferent to our carving
reality into nicely separated "bone fide" bodies--or, as they say,
this distinction is "transparent" from a basic scientific pov. The
idea of a physical system is "blind" to such a distinction: There is
tremenduous liberty in considering physical systems, and the laws of
Physics can be reformulated so as to take into consideration the
wackiest system configurations one might fancy. Now, agreed, whatever
laws one ends up writing following this "scrambling up" may be pretty
funky-looking--just as much as they could turn out, on the contrary,
to be better looking and simpler than the "originals"--however, I
claim, they are no less legitimate that the original ones.
It depends, then, on what one takes physical reality to stand for,
i.e. what sort of reality indicators you adhere to. If one relies on
Physics for this job (let us go, for now, with standard macroscopical
Physics, like standard Newtonian Dynamics, or standard
phenomenological Thermodynamics, where the notion of a physical system
is quite prominent), i.e. if one takes whatever Physics describes to
be THE reality (which is quite a common view), then the distinction
fiat/bona fide evaporates.
CC
Matthias Samwald
> Using SWOOP. For some reason, the only neuron_ontology.owl entities
> I see mapped into bfo classes are the neuron_ontology:Neuron class
> and its children. I see several neuron_ontology.owl classes declared
> as disjointWith certain bfo clases, but that's it.
Good god. After taking a closer look myself (...) I saw that I
uploaded a bogus version of the mapping yesterday. The disjoint
statements should in fact have been subclass statements. I uploaded
the right version now. I'm truly sorry for wasting your time with
that!
William Bug
The most credible BFO story in a nutshell would be something along these lines:
There is a membrane (indep:cont) which has the disposition (dep:cont)
of allowing for translocation (process) of ions (indep:cont), this
disposition is realized when translocation of a ion (object?) or a
group of ions (aggregate) occurs
b) The collection of I.K+-delayed_rectifier ion channelschannels are?
Ingvar Johansson
> On 4/18/07, Smith, Barry <phis...@buffalo.edu> wrote:
>
>> At 03:31 PM 4/18/2007, Matthias Samwald wrote:
>>
>>> I have to be honest: I find the distinction of "fiat" and "bona fide"
>>> boundaries highly subjective and don't see how this is compatible with
>>> the realist approach of BFO.
>>>
>> A realist ontology needs to do justice to the fact that there are
>> some boundaries in physical reality -- for example the surface of
>> your skin -- and some boundaries which exist only because subjects
>> put them there, e.g. the boundary of Poland.
>>
>> Do you accept that there are these two kinds of boundary (and of
>> course a variety of mixed cases including elements of each)?
>>
>
> As far as I know, the laws of Physics are indifferent to our carving
> reality into nicely separated "bone fide" bodies--or, as they say,
> this distinction is "transparent" from a basic scientific pov. The
> idea of a physical system is "blind" to such a distinction:
The distinction between, e.g., gravitational forces and electromagnetic
forces is a bona fide distinction. The distinction between forces that
are weaker and stronger than 1,67 Newton is fiat. Examples can be
multiplied.
> There is
> tremenduous liberty in considering physical systems, and the laws of
> Physics can be reformulated so as to take into consideration the
> wackiest system configurations one might fancy. Now, agreed, whatever
> laws one ends up writing following this "scrambling up" may be pretty
> funky-looking--just as much as they could turn out, on the contrary,
> to be better looking and simpler than the "originals"--however, I
> claim, they are no less legitimate that the original ones.
>
> It depends, then, on what one takes physical reality to stand for,
> i.e. what sort of reality indicators you adhere to. If one relies on
> Physics for this job (let us go, for now, with standard macroscopical
> Physics, like standard Newtonian Dynamics, or standard
> phenomenological Thermodynamics, where the notion of a physical system
> is quite prominent), i.e. if one takes whatever Physics describes to
> be THE reality (which is quite a common view), then the distinction
> fiat/bona fide evaporates.
>
Each and every physical theory contains bona fide distinctions; and fiat
distinctions can easily be introduced. The historical fact that in
physics one theory (e.g. Newton's) may be superseded by others (e.g.,
relativity theory and quantum mechanics) does not alter the fact first
mentioned.
best,
Ingvar
> CC
>
> >
--
Ingvar Johansson
IFOMIS, Saarland University
home site: http://ifomis.org/
personal home site:
http://hem.passagen.se/ijohansson/index.html
Phillip Lord
IJ> The distinction between, e.g., gravitational forces and
IJ> electromagnetic forces is a bona fide distinction. The
IJ> distinction between forces that are weaker and stronger than
IJ> 1,67 Newton is fiat. Examples can be multiplied.
Well my physics is a bit dodgy, but I think the distinction between
gravitational and electromagnetic is bona fide in some theories and
fiat in the others.
To use an example, easier for myself, the distinction between space
and time is bona fide in Newtonian and fiat in Relativistic
mechanics.
IJ> Each and every physical theory contains bona fide distinctions;
IJ> and fiat distinctions can easily be introduced. The historical
IJ> fact that in physics one theory (e.g. Newton's) may be
IJ> superseded by others (e.g., relativity theory and quantum
IJ> mechanics) does not alter the fact first mentioned.
But Newtonian mechanics hasn't been superseded by Relativistic. We use
both, cause Newtonian mechanics is computational and mathematically
more tractable. And quantum mechanics and relativistic mechanics
contradict each other.
How do we determine which theory we use to determine whether a
boundary is fiat or real?
Phil
Ingvar Johansson
>>>>>> "IJ" == Ingvar Johansson <ingvar.j...@ifomis.uni-saarland.de> writes:
>>>>>>
>
> IJ> The distinction between, e.g., gravitational forces and
> IJ> electromagnetic forces is a bona fide distinction. The
> IJ> distinction between forces that are weaker and stronger than
> IJ> 1,67 Newton is fiat. Examples can be multiplied.
>
> Well my physics is a bit dodgy, but I think the distinction between
> gravitational and electromagnetic is bona fide in some theories and
> fiat in the others.
>
It would be easier to answer if you took the effort to try to remember
the theories that are hovering in your head.
> To use an example, easier for myself, the distinction between space
> and time is bona fide in Newtonian and fiat in Relativistic
> mechanics.
>
This is wrong. In the special theory of relativity every inertial frame
of reference has a bona fide distinction between space and time. This
fact is by no means cancelled because, to speak a bit loosely, what is a
spatial distance in one frame may be a temporal distance in another frame.
>
> IJ> Each and every physical theory contains bona fide distinctions;
> IJ> and fiat distinctions can easily be introduced. The historical
> IJ> fact that in physics one theory (e.g. Newton's) may be
> IJ> superseded by others (e.g., relativity theory and quantum
> IJ> mechanics) does not alter the fact first mentioned.
>
>
> But Newtonian mechanics hasn't been superseded by Relativistic. We use
> both, cause Newtonian mechanics is computational and mathematically
> more tractable.
We use Newtonian mechanics today only when we know that it gives
*approximately* the same results as relativity theory.
> And quantum mechanics and relativistic mechanics
> contradict each other.
>
Originally - yes. Today - only parts of QM.
> How do we determine which theory we use to determine whether a
> boundary is fiat or real?
We trust the physicists' choice. This is an epistemological question. I
merely sid that whatever theory you opt for you will get some bona fide
distinctions. And you have to rely on some theory or other.
Ingvar
>
>
>
> Phil
Cristian Cocos
> of reference has a bona fide distinction between space and time. This
> fact is by no means cancelled because, to speak a bit loosely, what is
> a spatial distance in one frame may be a temporal distance in another
> frame.
The point is that time gets absolutely no privileged treatment in
Minkowsky's space (special relativity (SR)), it's just one of the
dimensions. No principled distinction between time and the other dimensions
can be made *within* SR. Any such distinction can only be made if one sort
of "jumps" outside SR (jumps where? I don't know. That's the point, there's
nowhere to jump, hence no objective distinction can be made). The bona fide
distinction that you mention cannot captured from within SR, nor can it, for
that reason, be captured from within GR or any theory that contains SR.
> We trust the physicists' choice. This is an epistemological question. I
> merely sid that whatever theory you opt for you will get some bona fide
> distinctions. And you have to rely on some theory or other.
Hence fiat/bona fide distinctions are theory-relative. Take two different
(but equivalent(!)) axiomatizations of the same domain, and you get
competing distinctions. What is bona fide in one appears as fiat in the
other and vice versa. I certainly agree with that, but I am not sure that
proponents of the fiat/bona-fide distinction would find this comforting. I
do not think they would like to see this trivialized in this manner, so as
to depend on the whim of some physicist or another (whoever yells louder, or
has the most financial resources to make his voice heard and stifle the
competition). For them, I take it, *a certain* fiat/bona-fide distinction
has an air of inevitability, an aura of absolute, where all physical
theories (formulated at the *same* granularity level), and all sane
physicists, should eventually converge. For them, I take it, there is such a
thing as THE ultimate reality carving-up. (Again, all mentions of
perspectivalism I have come across so far in seem to be involving appeal to
different granularity levels, which is, of course trivial. My concern is
theories formulated at the *same* granularity level, whatever that may be.)
CC
Matthew Pocock
> > But Newtonian mechanics hasn't been superseded by Relativistic. We use
> > both, cause Newtonian mechanics is computational and mathematically
> > more tractable.
>
> We use Newtonian mechanics today only when we know that it gives
> *approximately* the same results as relativity theory.
I beg to differ. We use Newtonian mechanics because a) it matches our
experience of the motions of and allows the predictions of masses under
forces, within the accuracy we require for every-day speeds, masses and
distances; and b) it is mathematically tractable and understandable.
That it gives *approximately* the same results as relativity under the range
of circumstances that we day-to-day encounter is a different issue, and
arrises at least in part from the historical derivations of the two
mechanics, and that they are accounting for similar observations (masses
changing velocity under the action of forces) under the constraint that these
two mechanics must be, within some acceptable error, a good description of,
or a good predictor of what we actually observe. To put it another way, both
newtonian mechanics and relativistic mechanics are artefactually
*aproximately* the same in the results they give because they are accounting
for these same results.
> > How do we determine which theory we use to determine whether a
> > boundary is fiat or real?
>
> We trust the physicists' choice. This is an epistemological question. I
> merely sid that whatever theory you opt for you will get some bona fide
> distinctions. And you have to rely on some theory or other.
Surely at this point, we are describing the theories that we 'opt for', and
not reality except indirectly in as much as those theories are theories about
reality? The object of enquiry has shifted from reality to conceptualisations
and theories which because we rely upon 'some theory or other' we assume have
a basis in reality.
>
> Ingvar
Matthew
Alan Ruttenberg
Alan Ruttenberg
equally describe it as a function of the membrane to conduct ions?
The only sense I have about disposition versus function (and this is
really not founded on anything but a feeling, unfortunately) is
something about whether, in the process, the bearer is active or
passive. But even with that distinction membranes can both actively
(via embedded channels) or passively (by diffusion) conduct ions?
Certainly, even the controversial "beneficial" requirement for
functions is satisfied by the membrane conducting ions.
-Alan
On Apr 18, 2007, at 4:55 AM, Pierre Grenon wrote:
>> neuron_ontology.owl#BrainRegion">
>> <owl:disjointWith
>> rdf:resource="http://www.ifomis.org/bfo/1.0/
>> snap#IndependentContinuant"/>
>> </rdf:Description>
>> <rdf:Description
>> rdf:about="http://neuroweb.med.yale.edu/senselab/
>> neuron_ontology.owl#Neuron_Transmitter">
>> <owl:disjointWith
>> rdf:resource="http://www.ifomis.org/bfo/1.0/snap#Object"/>
>> </rdf:Description>
>> <rdf:Description
>> rdf:about="http://neuroweb.med.yale.edu/senselab/
>> neuron_ontology.owl#Compartment">
>> <owl:disjointWith
>> rdf:resource="http://www.ifomis.org/bfo/1.0/
>> snap#IndependentContinuant"/>
>> </rdf:Description>
>> <rdf:Description
>> rdf:about="http://neuroweb.med.yale.edu/senselab/
Alan Ruttenberg
On Apr 18, 2007, at 3:47 AM, Pierre Grenon wrote:
>
> On 4/18/07, Matthias Samwald <sam...@gmx.at> wrote:
>>
>> Here is the mapping of the SenseLab NeuronDB OWL ontology to BFO, as
>> requested in the other discussion thread:
>> http://neuroweb.med.yale.edu/senselab/neuron_ontology-bfo-mapping.owl
>>
>> As you can see, it is quite simple. Where I was uncertain whether to
>> map to "object" or "fiat part of object", i chose the more generic
>> superclass.
>
> good thinking
OTOH, they aren't boundaries, and assignment to the more general
level would allow that.
>
>> We also plan to replace the property "has receptor" with
>> "has part" from the relation ontology.
>
> why not keeping it and making it a subrelation?
Could, but should then probably say:
subClassOf(restriction(hasReceptor some protein) restriction(hasPart
some (protein and restriction(has_function some receptor_activity)))
or maybe equivalentClass?
-Alan
> pierre
>> cheers,
>> Matthias Samwald
>>
>>
>>>
>>
>
> >
Ingvar Johansson
> On Wednesday 18 April 2007, Ingvar Johansson wrote:
>
>>> But Newtonian mechanics hasn't been superseded by Relativistic. We use
>>> both, cause Newtonian mechanics is computational and mathematically
>>> more tractable.
>>>
>> We use Newtonian mechanics today only when we know that it gives
>> *approximately* the same results as relativity theory.
>>
>
> I beg to differ. We use Newtonian mechanics because a) it matches our
> experience of the motions of and allows the predictions of masses under
> forces, within the accuracy we require for every-day speeds, masses and
> distances; and b) it is mathematically tractable and understandable.
>
I stated only a necessary condition for the application of Newtonian
mechanics. Your reasons apply as well.
> That it gives *approximately* the same results as relativity under the range
> of circumstances that we day-to-day encounter is a different issue, and
> arrises at least in part from the historical derivations of the two
> mechanics, and that they are accounting for similar observations (masses
> changing velocity under the action of forces) under the constraint that these
> two mechanics must be, within some acceptable error, a good description of,
> or a good predictor of what we actually observe.
> To put it another way, both
> newtonian mechanics and relativistic mechanics are artefactually
> *aproximately* the same in the results they give because they are accounting
> for these same results.
>
But when the velocities become close to that of light the theories are
NOT approximately the same. Then they differ dramatically.
>
>>> How do we determine which theory we use to determine whether a
>>> boundary is fiat or real?
>>>
>> We trust the physicists' choice. This is an epistemological question. I
>> merely sid that whatever theory you opt for you will get some bona fide
>> distinctions. And you have to rely on some theory or other.
>>
>
> Surely at this point, we are describing the theories that we 'opt for', and
> not reality except indirectly in as much as those theories are theories about
> reality? The object of enquiry has shifted from reality to conceptualisations
> and theories which because we rely upon 'some theory or other' we assume have
> a basis in reality.
>
The object of enquiry has NOT shifted from reality to conceptualisations
merely because we have become aware of the fact that even scientific
theories are fallible.
Ingvar
>
>> Ingvar
>>
>
> Matthew
Ingvar Johansson
>> This is wrong. In the special theory of relativity every inertial frame
>> of reference has a bona fide distinction between space and time. This
>> fact is by no means cancelled because, to speak a bit loosely, what is
>> a spatial distance in one frame may be a temporal distance in another
>> frame.
>>
>
> The point is that time gets absolutely no privileged treatment in
> Minkowsky's space (special relativity (SR)), it's just one of the
> dimensions.
First point: but this does not mean that time can be reduced to the
spatial dimensions, or that there is only a fiat distinction between the
temporal dimension and the spatial dimensions. SR postulates three
spatial dimensions and one temporal. If the theory is true, the world
is like that - bona fide. If you think that SR is merely a social
construction (like a novel), and can't be judged as being either true or
false, then I am prepared to discuss this issue and defend external
world realism (see my last comment to this mail).
> No principled distinction between time and the other dimensions
> can be made *within* SR.
Second point: time is represented by the variable t, and (normally) the
three spatial dimensions are represented by the three variables x, y,
and z. When space and time are 'added', the time variable is multiplied
by i (the sqaure of minus 1) in order to keep it distinct from the
spatial dimensions.
> Any such distinction can only be made if one sort
> of "jumps" outside SR (jumps where? I don't know. That's the point, there's
> nowhere to jump, hence no objective distinction can be made). The bona fide
> distinction that you mention cannot captured from within SR,
As explained above: completely false!
> nor can it, for
> that reason, be captured from within GR or any theory that contains SR.
>
>
>> We trust the physicists' choice. This is an epistemological question. I
>> merely sid that whatever theory you opt for you will get some bona fide
>> distinctions. And you have to rely on some theory or other.
>>
>
> Hence fiat/bona fide distinctions are theory-relative. Take two different
> (but equivalent(!)) axiomatizations of the same domain, and you get
> competing distinctions. What is bona fide in one appears as fiat in the
> other and vice versa. I certainly agree with that, but I am not sure that
> proponents of the fiat/bona-fide distinction would find this comforting. I
> do not think they would like to see this trivialized in this manner, so as
> to depend on the whim of some physicist or another (whoever yells louder, or
> has the most financial resources to make his voice heard and stifle the
> competition).
Why do you think physicists are making experiments in order to test
their theories? Do you think the theories that made the atom bomb
possible are mere "whims" of some physicists and have nothing with a
mind-independent reality to do?
> For them, I take it, *a certain* fiat/bona-fide distinction
> has an air of inevitability,
I think that the natural sciences so far have been able to create
theories with an increasing degree of truthlikeness, and hope that they
will continue to try to do so. This means that I do *not* believe that
today's sciences have got hold of the ultimate and specific bona fide
boundaries in the world, but since all theories we know of postulate
some such boundaries, it seems safe to say that the fiat/bona fide
distinction will never be cancelled.
> an aura of absolute, where all physical
> theories (formulated at the *same* granularity level), and all sane
> physicists, should eventually converge. For them, I take it, there is such a
> thing as THE ultimate reality carving-up.
Don't you believe that your last sentence is true? Don't you believe,
for instance, in THE ultimate reality of your two kids? Don't you think
they have a real existence that is independent of other peoples' minds,
and that their bodies are structured in a very specific way? Do you
think they are *mere* social constructions?
Ingvar
> (Again, all mentions of
> perspectivalism I have come across so far in seem to be involving appeal to
> different granularity levels, which is, of course trivial. My concern is
> theories formulated at the *same* granularity level, whatever that may be.)
>
> CC
>
>
> >
Alan Ruttenberg
> I would say it's not like you want to kill it altogether but trimming
> it and considering whether some subclasses ougt to be migrated and
> where seems indeed necessary.
Any good reason to keep it, other than it not needing to be killed?
-Alan
Matthew Pocock
> Matthew Pocock schrieb:
>
> > That it gives *approximately* the same results as relativity under the
> > range of circumstances that we day-to-day encounter is a different issue,
> > and arrises at least in part from the historical derivations of the two
> > mechanics, and that they are accounting for similar observations (masses
> > changing velocity under the action of forces) under the constraint that
> > these two mechanics must be, within some acceptable error, a good
> > description of, or a good predictor of what we actually observe.
> > To put it another way, both
> > newtonian mechanics and relativistic mechanics are artefactually
> > *aproximately* the same in the results they give because they are
> > accounting for these same results.
>
> But when the velocities become close to that of light the theories are
> NOT approximately the same. Then they differ dramatically.
We all agree on this, hence my hedge about day-to-day situations.
> >> We trust the physicists' choice. This is an epistemological question. I
> >> merely sid that whatever theory you opt for you will get some bona fide
> >> distinctions. And you have to rely on some theory or other.
> >
> > Surely at this point, we are describing the theories that we 'opt for',
> > and not reality except indirectly in as much as those theories are
> > theories about reality? The object of enquiry has shifted from reality to
> > conceptualisations and theories which because we rely upon 'some theory
> > or other' we assume have a basis in reality.
>
> The object of enquiry has NOT shifted from reality to conceptualisations
> merely because we have become aware of the fact that even scientific
> theories are fallible.
Falliability of theories is neither here nor there. The issue at hand is that
if we accept that there is some underlying reality and also recognise that
there are theories about it with varying abilities to account for various
portions of the reality, such as the different mechanics accounting for
motions under the action of forces, then we have a plain choice. Either we
attempt to describe reality directly, or we describe it in terms of one or
more of these theories. If we are actually doing the latter while claiming to
do (or even believing that we are doing) the former then a subtle shift in
the object of conceptualisation has taken place. More incideously, if we are
doing the latter, and recognising no distinction between the theory about
reality and the reality itself, we will be reaching very wrong conclusions
about the world about us.
So, in an ontology discussing mechanics, for the sake of getting some physics
done we could agree that in reality there is some 'massyness' of a physical
object, and that this is responsible for how 'resistive' that object is to
altering velocity in response to an external 'force'. A particular mechanics,
e.g. newtonian, may postulate that newtonian interpretations of objects have
a particular newtonian interpretation of mass. However, this 'newtonian mass'
bears a complex relationship to 'massyness' and is clearly theory-laden. To
claim that 'newtonian mass' == 'massyness' is to invite desaster.
The 'massyness' itself is not theory-laden, as it is a direct property of the
real physical world by virtue of how you can interact with it. Regardless of
what theory of mechanics you are using, the physical world will carry on
working the way it does (at least, that's how things appear to be working to
me).
Matthew
>
> Ingvar
>
> >> Ingvar
> >
> > Matthew
Matthew Pocock
On Thursday 19 April 2007, Ingvar Johansson wrote:
> Matthew Pocock schrieb:
> > then we have a plain choice. Either we
> > attempt to describe reality directly, or we describe it in terms of one
> > or more of these theories.
>
> There is no choice! There is no way to describe reality *directly*! We
> have to use concepts; and with the concepts come more or less elaborated
> theories.
>
> Ingvar
If this is true, then we should not be claiming to be describing reality. We
should be claiming to be describing theories. We chose to restrict ourselves
to describing those theories relevant to some aspect of reality.
Matthew
Phillip Lord
>>>>> "IJ" == Ingvar Johansson <ingvar.j...@ifomis.uni-saarland.de> writes:
>> Well my physics is a bit dodgy, but I think the distinction
>> between gravitational and electromagnetic is bona fide in some
>> theories and fiat in the others.
IJ> It would be easier to answer if you took the effort to try to
IJ> remember the theories that are hovering in your head.
As I said, my physics is a bit dodgy. The absence of names stems from
ignorance not memory loss. It what happens when a biologist tries to
struggle with such issues.
>> To use an example, easier for myself, the distinction between
>> space and time is bona fide in Newtonian and fiat in Relativistic
>> mechanics.
>>
IJ> This is wrong. In the special theory of relativity every
IJ> inertial frame of reference has a bona fide distinction between
IJ> space and time. This fact is by no means cancelled because, to
IJ> speak a bit loosely, what is a spatial distance in one frame may
IJ> be a temporal distance in another frame.
Ah, okay. Again forgive my physics, would acceleration and gravity be
a fiat boundary?
BTW, I am getting a little confused. The distinction between space and
time seems to me to be, well, a distinction rather than a
boundary. Are they the same thing?
>> But Newtonian mechanics hasn't been superseded by
>> Relativistic. We use both, cause Newtonian mechanics is
>> computational and mathematically more tractable.
IJ> We use Newtonian mechanics today only when we know that it gives
IJ> *approximately* the same results as relativity theory.
Yes, I agree. And we use Newtonian mechanics because it's
easier. If they had the same complexity, we'd always use relativistic
mechanics.
>> And quantum mechanics and relativistic mechanics contradict each
>> other.
>>
IJ> Originally - yes. Today - only parts of QM.
>> How do we determine which theory we use to determine whether a
>> boundary is fiat or real?
IJ> We trust the physicists' choice. This is an epistemological
IJ> question.
Yes. I am just worried about what we do when they haven't chosen.
IJ> I merely sid that whatever theory you opt for you will get some
IJ> bona fide distinctions. And you have to rely on some theory or
IJ> other.
Ah, okay, so where they haven't chosen, you would choose to represent
one theory or the other?
Phil
Ingvar Johansson
Why not? If I am saying 'Today the sun is shining in Saarbrücken', then
I am not trying to describe any theory/concepts. I am using everyday
language and common sense theories in order to describe a part and an
aspect of reality. Furthermore, it happens to be true. You can come and
look. But you should look at the weather, not at some theory or some
concepts. If you want to continue this discussion, I would like you to
read my paper "Bioinformatics and Biological Reality" (linked to section
0 on my home site).
best,
Ingvar Johansson
> Ah, okay. Again forgive my physics, would acceleration and gravity be
> a fiat boundary?
>
Acceleration is one thing, and gravitational forces is quite another
thing. In one sense there is no boundary between them; they are simply
in and of themselves completely different. However, I allowed myself to
say that there is then a *bona fide distinction* between them.
> BTW, I am getting a little confused. The distinction between space and
> time seems to me to be, well, a distinction rather than a
> boundary. Are they the same thing?
O.K., I understand your confusion, but I hope my remark above make the
thing clear.
Ingvar
Cristian Cocos
On 4/19/07, Ingvar Johansson <ingvar.j...@ifomis.uni-saarland.de> wrote:
>
> First point: but this does not mean that time can be reduced to the
> spatial dimensions, or that there is only a fiat distinction between the
> temporal dimension and the spatial dimensions. SR postulates three
> spatial dimensions and one temporal. If the theory is true, the world
> is like that - bona fide.
SR postulates the fact that the Universe is a Minkowsky universe. A
Minkowsky universe has four dimensions, i.e. it is "composed" of
so-called quadrivectors. An orthonormal basis (reference frame) is,
hence, made up of four unit orthogonal vectors, say x1, x2, x3, x4.
Its metric is given by dR = dx1^2 + dx2^2 + dx3^2 + dx4^2. So far time
(the temporal dimension) has nowhere been mentioned. In fact we could
stop here, without ever mentioning "time" or "spatial dimensions." It
should suffice. Time, or "spatial dimensions" are obsolete terms from
the pov of SR. All these books that present SR do nothing but
establish the relation between a faint approximation of reality
(described by Newtonian mechanics, with its nice clear "bona-fide"
distinctions between time on the one hand, and spatial dimensions on
the other), and reality itself (described by SR (let us, for the
moment, assume that SR is true)), where no such separation occurs. Now
*why* do they go to the trouble of presenting this relation between
the two theories, when they could simply describe the Minkowsky
universe? It is because this approximation happens to have a huge
relevance for us, "obsolete" humans, whose life has been (so far!)
confined to a world where relativistic notions have very little day to
day impact. *That* is the only reason why you still see the word
"time" mentioned in SR books. In fact, you could very well imagine
that our civilization discovered SR directly, without going first
through the long and arduous road of discovering Newtonian mechanics
first. Moreover, in this scenario, Newtonian mechanics has never been
discovered. There is no word "time" or "spatial dimension," just x1,
x2, x3, x4.
"Time" and "space" are the reflexes of a familiar, but unfortunately
obsolete, world. It is an ordinary psychological fact that we are
inclined to "find" familiar structures into new territory, however
arcane this might turn out to be.
There is something essential about time and spatial dimensions that
happens in Newtonian mechanics, but does *not* happen in SR, which
could, indeed, count as a bona-fide demarcation in Newtonian
mechanics. And that is the fact that in Newtonian mechanics the metric
of the space does not depend on time (the universe is Euclidean):
Time, on the one hand, and spatial dimensions on the other, are
separate. Now, again, this thing does *not* happen in SR. All serious
reasons for drawing a bona-fide distinction between one of the four
dimensions and the other three has dissapeared: in order to
differentiate between these four dimensions one has to make recourse
to an obsolete theory (Newtonian mechanics), founded on obsolete
notions (space and time).
> > No principled distinction between time and the other dimensions
> > can be made *within* SR.
>
> Second point: time is represented by the variable t, and (normally) the
> three spatial dimensions are represented by the three variables x, y,
> and z. When space and time are 'added', the time variable is multiplied
> by i (the sqaure of minus 1) in order to keep it distinct from the
> spatial dimensions.
Again, see above. This is just for our peace of mind (for "legacy"
reasons, as they say), i.e. to make a connection between SR and
something very familiar--though utterly false (i.e. Newtonian
mechanics with its obsolete notions (space and time))). There is no
need to mention neither time, nor space, in SR. Reality (as described
by SR) has no room for space and time.
> I think that the natural sciences so far have been able to create
> theories with an increasing degree of truthlikeness, and hope that they
> will continue to try to do so. This means that I do *not* believe that
> today's sciences have got hold of the ultimate and specific bona fide
> boundaries in the world, but since all theories we know of postulate
> some such boundaries, it seems safe to say that the fiat/bona fide
> distinction will never be cancelled.
It probably won't be canceled, though it will always remain a
theory-relative distinction: note that there exist different but
equivalent theories of one and the same scientific field, each theory
postulating its own fiat/bona-fide boundaries, which boundaries do not
necessarily match (meaning that what is bona-fide in one appears as
fiat in others and vice versa). The fiat/bona-fide distinction, hence,
carries a lot *less* metaphysical weight than its proponents might be
inclined to bestow upon it.
> > an aura of absolute, where all physical
> > theories (formulated at the *same* granularity level), and all sane
> > physicists, should eventually converge. For them, I take it, there is such a
> > thing as THE ultimate reality carving-up.
>
> Don't you believe that your last sentence is true?
No, however I do believe in the existence of reality. No ultimate
(hence self-legitimizing) reality *carving-up* though. There is no way
to prefer a decoupage (carving-up) over another.
> Don't you believe,
> for instance, in THE ultimate reality of your two kids?
Of course I do believe in their reality, just as I equally and
equivalently believe in the ultimate reality of Entity#1 made up of
two parts (Gaby's upper part and Adam lower part), and Entity#2 made
up of two parts (Adam's upper part and Gaby's lower part). (Note that
I do *not* mean by this cutting the kids in two and sew the respective
parts together as the description might suggest). Both of these
decoupages are of equal importance to me.
Cheers,
C
--
"I don't want to achieve immortality through my work. I want to
achieve it through not dying." -- Woody Allen
Smith, Barry
Suppose we can see cells only through microscopes, each with its own
settings, lens specifications, etc. Would this mean that we should
stop describing cells and start describing microscopes?
BS
>Matthew
>
>
>
>
>
Ingvar Johansson
>> spatial dimensions, or that there is only a fiat distinction between the
>> temporal dimension and the spatial dimensions. SR postulates three
>> spatial dimensions and one temporal. If the theory is true, the world
>> is like that - bona fide.
>
> Minkowsky universe has four dimensions, i.e. it is "composed" of
> so-called quadrivectors. An orthonormal basis (reference frame) is,
> hence, made up of four unit orthogonal vectors, say x1, x2, x3, x4.
> Its metric is given by dR = dx1^2 + dx2^2 + dx3^2 + dx4^2. So far time
> (the temporal dimension) has nowhere been mentioned. In fact we could
> stop here, without ever mentioning "time" or "spatial dimensions."
No, we cannot -- if we are interested in physics and not only in
mathematics. Whereas x1, x2, and x3 without any qualifications represent
three orthogonal spatial dimensions, dx4 has to be written out as 'ict'
('i' = the root of minus 1; c = the light of velocity; t = the time
parameter). If, as is usually done, c is by convention given the
magnitude 1, then we get: dx4^2 = -t^2. And this marks dx4 off as being
very special in relation to the other three dimensions. (I guess 'dR' in
your metric formula should be 'dR^2' .)
> It
> should suffice. Time, or "spatial dimensions" are obsolete terms from
> the pov of SR.
No, they are not: (1) within each inertial frame of reference the
distincton between space and time is just as absolute as in Newtonian
mechanics; (2) the distinction between spatial and temporal dimenions is
essential to SR's distinction between pair of events that are
*space-like*, *time-like*, and *light-like*, respectively.
> All these books that present SR do nothing but
> establish the relation between a faint approximation of reality
> (described by Newtonian mechanics, with its nice clear "bona-fide"
> distinctions between time on the one hand, and spatial dimensions on
> the other), and reality itself (described by SR (let us, for the
> moment, assume that SR is true)), where no such separation occurs. Now
> *why* do they go to the trouble of presenting this relation between
> the two theories, when they could simply describe the Minkowsky
> universe? It is because this approximation happens to have a huge
> relevance for us, "obsolete" humans, whose life has been (so far!)
> confined to a world where relativistic notions have very little day to
> day impact. *That* is the only reason why you still see the word
> "time" mentioned in SR books. In fact, you could very well imagine
> that our civilization discovered SR directly, without going first
> through the long and arduous road of discovering Newtonian mechanics
> first. Moreover, in this scenario, Newtonian mechanics has never been
> discovered. There is no word "time" or "spatial dimension," just x1,
> x2, x3, x4.
>
> "Time" and "space" are the reflexes of a familiar, but unfortunately
> obsolete, world. It is an ordinary psychological fact that we are
> inclined to "find" familiar structures into new territory, however
> arcane this might turn out to be.
My arguments above are much more than a psychological refusal to
understand a new theory.
> There is something essential about time and spatial dimensions that
> happens in Newtonian mechanics, but does *not* happen in SR, which
> could, indeed, count as a bona-fide demarcation in Newtonian
> mechanics. And that is the fact that in Newtonian mechanics the metric
> of the space does not depend on time (the universe is Euclidean):
> Time, on the one hand, and spatial dimensions on the other, are
> separate. Now, again, this thing does *not* happen in SR. All serious
> reasons for drawing a bona-fide distinction between one of the four
> dimensions and the other three has dissapeared: in order to
> differentiate between these four dimensions one has to make recourse
> to an obsolete theory (Newtonian mechanics), founded on obsolete
> notions (space and time).
>
>> > No principled distinction between time and the other dimensions
>> > can be made *within* SR.
>>
>> Second point: time is represented by the variable t, and (normally) the
>> three spatial dimensions are represented by the three variables x, y,
>> and z. When space and time are 'added', the time variable is multiplied
>> by i (the sqaure of minus 1) in order to keep it distinct from the
>> spatial dimensions.
First an excuse: I happened to write 'the square of minus 1' instead of
'the root of minus 1'.
If we introduce my variables (x,y z t) in the paragraph above instead of
Cristian's (x1 x2, x3, x4), then we get (compare Cristian's first formula):
dR^2 = dx^2 + dy^2 + dz^2 - t^2.
Obvioulsy, the time variable is represented as being special, since it
(its square) is subtracted from, not added, to the other dimensions..
> This is just for our peace of mind (for "legacy"
> reasons, as they say), i.e. to make a connection between SR and
> something very familiar--though utterly false (i.e. Newtonian
> mechanics with its obsolete notions (space and time))). There is no
> need to mention neither time, nor space, in SR. Reality (as described
> by SR) has no room for space and time.
>
>> I think that the natural sciences so far have been able to create
>> theories with an increasing degree of truthlikeness, and hope that they
>> will continue to try to do so. This means that I do *not* believe that
>> today's sciences have got hold of the ultimate and specific bona fide
>> boundaries in the world, but since all theories we know of postulate
>> some such boundaries, it seems safe to say that the fiat/bona fide
>> distinction will never be cancelled.
>
> It probably won't be canceled, though it will always remain a
> theory-relative distinction:
A distinction (such as that between 'distinctions in nature' and 'purely
man-made distinctions') that occurs in all theories, but does it in
different ways, is in one sense theory-relative but in another it is not.
> note that there exist different but
> equivalent theories of one and the same scientific field, each theory
> postulating its own fiat/bona-fide boundaries, which boundaries do not
> necessarily match (meaning that what is bona-fide in one appears as
> fiat in others and vice versa). The fiat/bona-fide distinction, hence,
> carries a lot *less* metaphysical weight than its proponents might be
> inclined to bestow upon it.
>
>> > an aura of absolute, where all physical
>> > theories (formulated at the *same* granularity level), and all sane
>> > physicists, should eventually converge. For them, I take it, there
>> is such a
>> > thing as THE ultimate reality carving-up.
>>
>> Don't you believe that your last sentence is true?
>
> No, however I do believe in the existence of reality. No ultimate
> (hence self-legitimizing) reality *carving-up* though. There is no way
> to prefer a decoupage (carving-up) over another.
As far as I can understand, your view must imply that reality contains
no structure of its own. In fact, I have recently on internet stumbled
upon a philosopher who explicitly argues in favor of such a
'mereological nihilism'. There are obvious flaws in his argumentation.
>
>> Don't you believe,
>> for instance, in THE ultimate reality of your two kids?
>
> Of course I do believe in their reality, just as I equally and
> equivalently believe in the ultimate reality of Entity#1 made up of
> two parts (Gaby's upper part and Adam lower part), and Entity#2 made
> up of two parts (Adam's upper part and Gaby's lower part). (Note that
> I do *not* mean by this cutting the kids in two and sew the respective
> parts together as the description might suggest). Both of these
> decoupages are of equal importance to me.
Have I got you right? Are you really saying that there is no
ontologically important distinction between regarding, on the one hand,
Gaby and Adam as two unities, and, on the other hand, regarding Entity#1
and Entity#2 as two unities?
best,
Ingvar
>
> Cheers,
Matthew Pocock
> First an excuse: I happened to write 'the square of minus 1' instead of
> 'the root of minus 1'.
> If we introduce my variables (x,y z t) in the paragraph above instead of
> Cristian's (x1 x2, x3, x4), then we get (compare Cristian's first formula):
> dR^2 = dx^2 + dy^2 + dz^2 - t^2.
> Obvioulsy, the time variable is represented as being special, since it
> (its square) is subtracted from, not added, to the other dimensions..
Hum, I've not looked propperly at SR/GR since I was at school. I've had 2
cousins doing physics at degree, so a certain amount of xmas beer and
high-energy physics prattle has gone on since. My recolection was that there
is nothing baring any of x1, x2, x3 or x4 being complex values. Isn't this
part of the basis for botching QM and GR together when building models of
very small things with high energies, and very big things with low energies?
At this point, what we identify in every-day life with time realy does look
utterly indistinguishable from the other 3 'dimensions', except in that the
newtonian-congruent portion of space is in the real projections of x1..3 and
the complex projection of x4, while time is in the complex projection of
x1..x3 and the real projection of x4. Viewed like this, the (i) attached to
your t^2 is a mathematical hack.
However, I'm more than happy to punt this over to a physicist who knows
better.
> best,
> Ingvar
Matthew
Matthias Samwald
> Of course it would be essential to have some questions which depended
> on the answer to that classification. For example, if we know that
> something is a fiat boundary as opposed to a bona fide boundary, what
> do we now that helps in biological reasoning?
>
> Certainly I think it helps groups designing ontologies reach
> consensus quicker.
Really? My impression so far is that the distinction between fiat and
bona fide boundaries only creates a lot of controversy, without a hint
of any practical value (e.g. for reasoning or better understanding of
the things we are talking about).
William Bug
My impression so far is that the distinction between fiat and
bona fide boundaries only creates a lot of controversy, without a hint
of any practical value (e.g. for reasoning or better understanding of
the things we are talking about).
Matthew Pocock
I feel a little charicatured :)
Your question pertains to the conflation of observations with the underlying
reality, not that between theoretical entities and the bits of an underlying
reality they seek to explain.
Anyhow, running with it for now, what I would say is that we can only claim
with any certainty what we see, and how we went about seeing it. What we
believe gave rise to the things we see is conjecture - in this case, the
presence of cells are the reason we see prety shapes under high
magnifications, in accordance with what we know about optics. This is what
experimental science is realy like.
To make a claim that cells are real, and to do this while ignoring that the
only way to gain information about them is via a range of microscopes used in
a process of observation is to invite disaster. For example, if each
microscope consistently miss-represented something about cells, then if we
don't make the distinction between 'cells as viewed by microscopes' and 'real
cells' and at the same time also describe the process of observation, then
bad things will inevitably happen.
As a concrete example, if there where systematic defects in the lenzes so that
certain wavelengths are not transmitted correctly, then some cell types could
be invisible, or look similar to solid lumps of calcium. Things that
are 'cells in reality' would be absent or be something else in the
observation. If we are stating an identity between what we see and what is
there, then we are now reaching wrong conclusions about the real world.
Oh, and just in case of confusion about what I mean by 'what we see', I would
use it in the same way that when I watch telly I see things moving, but what
is realy happening is that many still pictures are flashed up on the same
screen many times a seccond.
This seems to be in a similar vein with others on this list who seem to
believe that their personal experience of, for example, the sunyness of the
weather in their city is a statment of truth about reality because they have
experienced it. I'd just hedge and say it is a truth about their observation
of reality, since I am inclidend to believe that they observed sunny weather.
However, my time in and around biology has taught me that we mostly all wear
rose-tinted spectacles all the time, and that you nearly never see what is
actually there, as it is realy is hence my insistance on the hedge.
Matthew
Smith, Barry
>On Thursday 19 April 2007, Smith, Barry wrote:
> > At 12:21 PM 4/19/2007, you wrote:
> > >If this is true, then we should not be claiming to be describing reality.
> > > We should be claiming to be describing theories. We chose to restrict
> > > ourselves to describing those theories relevant to some aspect of
> > > reality.
> >
> > Suppose we can see cells only through microscopes, each with its own
> > settings, lens specifications, etc. Would this mean that we should
> > stop describing cells and start describing microscopes?
> > BS
>
>I feel a little charicatured :)
>
>Your question pertains to the conflation of observations with the underlying
>reality, not that between theoretical entities and the bits of an underlying
>reality they seek to explain.
>
>Anyhow, running with it for now, what I would say is that we can only claim
>with any certainty what we see, and how we went about seeing it. What we
>believe gave rise to the things we see is conjecture - in this case, the
>presence of cells are the reason we see prety shapes under high
>magnifications, in accordance with what we know about optics. This is what
>experimental science is realy like.
We leave the scientists to formulate the conjectures (they do pretty
well, after all, and they have lots of evidence for most of what they
conjecture); it is not up to the ontologist to add an extra layer of
(fussing about) conjectures; our job is to do our best to represent
the reality that scientists are describing for us.
BS
Matthew Pocock
> well, after all, and they have lots of evidence for most of what they
> conjecture); it is not up to the ontologist to add an extra layer of
> (fussing about) conjectures; our job is to do our best to represent
> the reality that scientists are describing for us.
> BS
But this is the root of the cause of the missunderstanding. Scientists don't
describe reality. They describe theories about reality. If this where not the
case, then scientists would not use hedges like 'relativistic mass' but just
go direct to saying 'mass'. I am at a loss to understand how an ontologist
can not see this distinction.
Matthew
Alan Ruttenberg
I didn't say that it makes arrival at consensus quick. I said
quicker. For instance, at this moment we do not have consensus on
what neurondb means by a compartment, and in particular, what makes
the axon hillock the same sort of entity as the cell body. I'd
venture to say that having the distinction between fiat and bona
fide, by introducing an appropriate way to talk about the
controversy *is* moving us in the right direction.
However, I suppose that I might have to retract my assertion if it
doesn't land up enticing you to respond to the issues I started
enumerating in the message you are quoting from. Will you?
-Alan
Ingvar Johansson
> On Thursday 19 April 2007, Ingvar Johansson wrote:
>
>> If we introduce my variables (x,y z t) in the paragraph above instead of
>> Cristian's (x1 x2, x3, x4), then we get (compare Cristian's first formula):
>> dR^2 = dx^2 + dy^2 + dz^2 - t^2.
>> Obvioulsy, the time variable is represented as being special, since it
>> (its square) is subtracted from, not added, to the other dimensions..
>>
>
> Hum, I've not looked propperly at SR/GR since I was at school. I've had 2
> cousins doing physics at degree, so a certain amount of xmas beer and
> high-energy physics prattle has gone on since.
Then, next time you meet your cousins you can ask them the following
questions the theory of special relativity (SR) and compare their
answers with mine:
(i) Isn't c (the velocity of light) the basic constant par preferance in SR?
Answer: Yes, it is.
(ii) Doesn't c have the physical dimension m/s, i.e., 'length/time'?
Answer: Yes, it has.
(iii) Doesn't a formula in *physics* such as 'dR^2 = dx^2 + dy^2 + dz^2
- t^2' need to have the same physical dimension on both sides of the
identity sign?
Answer: Yes, this is necessary for the formula to be physically meaningful.
(iv) But how can the formula above fulfill this requirement if R, x, y,
and z have the dimension 'length' and t 'time'?
Answer: Because c is given the value 1, and is therefore not visible in
the formula. If everything is made explicit, the formula looks as
follows: 'dR^2 = dx^2 + dy^2 + dz^2 - (ct)^2'. Since here the time
variable t (with dimension 'time') is multiplied by a velocity (with
dimension 'length/time') the whole expression (ct) has the dimension
'length', too.
(v) Can the distinction between spatial dimensions and the temporal
dimension without loss be taken away from SR?
Answer: No, definitely not.
Best spatiotemporal regards,
Ingvar
Ingvar Johansson
> This seems to be in a similar vein with others on this list who seem to
> believe that their personal experience of, for example, the sunyness of the
> weather in their city is a statment of truth about reality because they have
> experienced it. I'd just hedge and say it is a truth about their observation
> of reality, since I am inclidend to believe that they observed sunny weather.
> However, my time in and around biology has taught me that we mostly all wear
> rose-tinted spectacles all the time, and that you nearly never see what is
> actually there, as it is realy is hence my insistance on the hedge.
>
Let's make a new try. If I am saying 'Today the sun is shining in
Saarbrücken', then I am not trying to describe any theory/concepts. I am
using everyday language and common sense theories in order to describe a
part and an aspect of reality. In order to check whether I am right or
wrong (you never know with *absolute certainty* anything about empirical
matters) you should look at the weather, not at some theory or some
concepts. Perception is our primary link to mind-independent reality.
Another point. You are putting forward an *inconsistent position*. On
the one hand you say that we can never know anything, but on the other
hand you also say that you *know* that we all "wear rose-tinted
spectacles all the time".
Ingvar
Matthew Pocock
> Matthew Pocock schrieb:
> > This seems to be in a similar vein with others on this list who seem to
> > believe that their personal experience of, for example, the sunyness of
> > the weather in their city is a statment of truth about reality because
> > they have experienced it. I'd just hedge and say it is a truth about
> > their observation of reality, since I am inclidend to believe that they
> > observed sunny weather. However, my time in and around biology has taught
> > me that we mostly all wear rose-tinted spectacles all the time, and that
> > you nearly never see what is actually there, as it is realy is hence my
> > insistance on the hedge.
>
> Let's make a new try. If I am saying 'Today the sun is shining in
> Saarbrücken', then I am not trying to describe any theory/concepts. I am
> using everyday language and common sense theories in order to describe a
> part and an aspect of reality. In order to check whether I am right or
> wrong (you never know with *absolute certainty* anything about empirical
> matters) you should look at the weather, not at some theory or some
> concepts.
Hum - you say you are not describing any theory and then go on to say that
a "common sense" theory is involved with this description. Hence, you are
saying that your statement about the weather depends in part on the weather,
in part on your way of observing it and in part on your common sense theory
of weather. If any one of these three where different, then you may tell me
something else. This is starting to sound alarmingly close to what I have
been claiming throughout this thread.
> Perception is our primary link to mind-independent reality.
On what basis do you take this position? 1) that there is mind independent
reality 2) that perception links to this, rather than something else. Is this
a statement of belief, convenience, or something that you can empracally or
logically demonstrate? I am not saying I think you are wrong, but would like
to hear why you personally think it is a justifiable position.
> Another point. You are putting forward an *inconsistent position*. On
> the one hand you say that we can never know anything, but on the other
> hand you also say that you *know* that we all "wear rose-tinted
> spectacles all the time".
I don't think I said that I *know* :) I said that "my time in and around
biology has taught me ..." which is not the same thing. Regardless, to
axiomatically equate what we know with what is true must be the first
cardinal sin of any student of philosophy, surely :D
What I thought I'd explained is that I am taking the position that every
observation depends upon both the means of observation and the thing
observed. I go further, and claim that to deny this is likely to lead to
systematic misconceptions about the thing observed which arise due to
artefacts of the means of observation. You can't dissagree with this, surely?
Matthew
>
> Ingvar
>
>
Matthias Samwald
realist world view is really accurate or not. It might be useful to
start a separate thread on this topic, and keep the other threads
clear of these kinds of discussion. BFO is based on this world view,
and when we want to work with BFO for any reason we should try to play
along as best as we can.
Personally, I consider myself as a radical constructivist [1] -- like
many people that have spent some time studying how our brain and
cognition work. However, I still find that the BFO ontology can be
very practical in many use cases (e.g. implementing reasoning,
implementing biological models that can be used for mathematical
simulation). Even if some of us do not agree with the underlying world
view, we can still use BFO as a practical tool, and accept it its
underlying premises when we are using it
-- Matthias Samwald
Matthias Samwald
Observing" interesting in this context. It is available online at
http://www.inteco.cl/biology/ontology/
-- Matthias
(I won't go further off-topic, I promise)
Matthew Pocock
> Matthew Pocock schrieb:
> > On Thursday 19 April 2007, Ingvar Johansson wrote:
> >> If we introduce my variables (x,y z t) in the paragraph above instead of
> >> Cristian's (x1 x2, x3, x4), then we get (compare Cristian's first
> >> formula): dR^2 = dx^2 + dy^2 + dz^2 - t^2.
> >> Obvioulsy, the time variable is represented as being special, since it
> >> (its square) is subtracted from, not added, to the other dimensions..
> >
> > Hum, I've not looked propperly at SR/GR since I was at school. I've had 2
> > cousins doing physics at degree, so a certain amount of xmas beer and
> > high-energy physics prattle has gone on since.
>
> Then, next time you meet your cousins you can ask them the following
> questions the theory of special relativity (SR) and compare their
> answers with mine:
> (iv) But how can the formula above fulfill this requirement if R, x, y,
> and z have the dimension 'length' and t 'time'?
> Answer: Because c is given the value 1, and is therefore not visible in
> the formula. If everything is made explicit, the formula looks as
> follows: 'dR^2 = dx^2 + dy^2 + dz^2 - (ct)^2'. Since here the time
> variable t (with dimension 'time') is multiplied by a velocity (with
> dimension 'length/time') the whole expression (ct) has the dimension
> 'length', too.
This is getting rather off topic, so I'll restrict myself to facts and
conjecture, not points of philosophy. Again, if there is a physicist
listening, please put us out of our misery and tell us what is realy going
on. Please.
As I said earlier, SR (and GR, which superceeded it) doesn't restrict any of
these dimensions to real numbers. They can be, and to explain certain
phenomena, must be complex. If we start from a Newtonian model with a
spato-temporal location we can do the mapng thus:
(x,y,z,t) --fromNewtonianReference--> ((x,c0i), (y,c0i), (z,c0i), (0, cti))
If we are modelling some fancy bit of physics we may end up with a coordinate
where all values can be filled in with potentially non-zero values, rather
more like:
((x, cwi), (y, cvi), (z, cui), (a, cti))
The formula for dR^2 will work just dandy with this, and with a little
re-labeling of the four terms above to (x1, x2, x3, x4) we are home and dry,
back to the formula you quote from Christian, with no distinction made
between the four SR/GR dimensions. No rabits up my sleeve! I promiss. The
entire equation is even dimension-sound.
In SR/GR, subjective and possibly Newtonian-friendly time arises from how you
decompose your locating 4-vector. You only need to do this at the point you
want to tell a person that their twin has aged remarkably well, or draw
new-scientist friendly diagrams of how during an eclipse you can weigh the
Sun by how much it bends the light of stars. As far as I know, you don't need
to drop into explicit 3-space+1-time within SR/GR to describe and compare
paths in spacetime, including temporal ordering or to demonstrate causality.
A quick scout through wikipedia would seem to indicate that I am not wholely
on the wrong track. At least in my head, substituting in complex numbers
throughout and flipping signs as needed seemed to work for all the equations
I can find that don't hide all the fun bits inside tensors.
http://en.wikipedia.org/wiki/Proper_time
http://en.wikipedia.org/wiki/Minkowski_metric
I freely admit that my understanding of SR/GR is somewhat blured by the
environment within which the knowledge was imparted to me. My brain starts
to melt when thinking of 4-d manifolds and metric tensors. Give me the
haskell class system any day.
Matthew
Ingvar Johansson
> I think that this quite a fundamental discussion about whether a
> realist world view is really accurate or not. It might be useful to
> start a separate thread on this topic, and keep the other threads
> clear of these kinds of discussion. BFO is based on this world view,
> and when we want to work with BFO for any reason we should try to play
> along as best as we can.
>
> Personally, I consider myself as a radical constructivist [1] -- like
> many people that have spent some time studying how our brain and
> cognition work.
Now, think for a moment. Have these people studied (a) real
mind-independently existing brains, or have they studied merely (b) a
social construction that goes under the name 'brain'? If you choose
alternative (a) then you are not a radical constructivist, and if you
choose (b) I can see no reason why this fact should turn you into a
radical social constructivisst.
Ingvar
> However, I still find that the BFO ontology can be
> very practical in many use cases (e.g. implementing reasoning,
> implementing biological models that can be used for mathematical
> simulation). Even if some of us do not agree with the underlying world
> view, we can still use BFO as a practical tool, and accept it its
> underlying premises when we are using it
>
> -- Matthias Samwald
>
> [1] http://www.univie.ac.at/constructivism/
>
>
> >
Ingvar Johansson
> On Friday 20 April 2007, Ingvar Johansson wrote:
>
>> Matthew Pocock schrieb:
>>
>>> On Thursday 19 April 2007, Ingvar Johansson wrote:
>>>
>>>> If we introduce my variables (x,y z t) in the paragraph above instead of
>>>> Cristian's (x1 x2, x3, x4), then we get (compare Cristian's first
>>>> formula): dR^2 = dx^2 + dy^2 + dz^2 - t^2.
>>>> Obvioulsy, the time variable is represented as being special, since it
>>>> (its square) is subtracted from, not added, to the other dimensions..
>>>>
>>> Hum, I've not looked propperly at SR/GR since I was at school. I've had 2
>>> cousins doing physics at degree, so a certain amount of xmas beer and
>>> high-energy physics prattle has gone on since.
>>>
>> Then, next time you meet your cousins you can ask them the following
>> questions the theory of special relativity (SR) and compare their
>> answers with mine:
>>
>
>
>> (iv) But how can the formula above fulfill this requirement if R, x, y,
>> and z have the dimension 'length' and t 'time'?
>> Answer: Because c is given the value 1, and is therefore not visible in
>> the formula. If everything is made explicit, the formula looks as
>> follows: 'dR^2 = dx^2 + dy^2 + dz^2 - (ct)^2'. Since here the time
>> variable t (with dimension 'time') is multiplied by a velocity (with
>> dimension 'length/time') the whole expression (ct) has the dimension
>> 'length', too.
>>
>
> This is getting rather off topic, so I'll restrict myself to facts and
> conjecture, not points of philosophy.
I have not brought in this topic; and what is being said above is not
especially philosophical. It is physics. Contact your cousins.
> Again, if there is a physicist
> listening, please put us out of our misery and tell us what is realy going
> on. Please.
>
> As I said earlier, SR (and GR, which superceeded it) doesn't restrict any of
> these dimensions to real numbers.
And I said in one of my mails (to Cristian Cocos) that his expression
'dx4' abstracts away from the fact that it can be written 'ict'. The
letter i represents the imaginary number 'the root of minus one', and
everything that is multiplied by i becomes thereby an imaginary number.
A complex number is a real number plus an imaginary number; imaginary
numbers figure in SR but only in connection with the time variable. This
is one of several ways to make clear that the time dimension in SR
cannot be put on a par with the spatial dimensions.
End of discussion of SR as far as I am concerned.
Ingvar
Matthew Pocock
> And I said in one of my mails (to Cristian Cocos) that his expression
> 'dx4' abstracts away from the fact that it can be written 'ict'. The
> letter i represents the imaginary number 'the root of minus one', and
> everything that is multiplied by i becomes thereby an imaginary number.
> A complex number is a real number plus an imaginary number; imaginary
> numbers figure in SR but only in connection with the time variable. This
> is one of several ways to make clear that the time dimension in SR
> cannot be put on a par with the spatial dimensions.
>
> End of discussion of SR as far as I am concerned.
Gha!!! In SR/GR, there is no single 'time variable', and no time dimension.
This only exists when you project into a subjective frame of reference, by
which point you are talking about a non-SR/GR system.
Just to re-iterate, each of the 4 SR/GR coordinates are complex numbers. All
four of them can have both a real and a complex part. If you'd scanned
through those wikipedia pages I linked to, you'd see that SR/GR do not
distinguish between the real and complex parts for any of the tensor/manifold
stuff within which all the physics is done. To get out a subjective time, you
manipulate all four of the components, finding the length of an arc through
spacetime. None of the four components is labelled 'time', or contributes to
it any more or less than the others. All four of the complex coordinates
contribute to subjective space and subjective time equally if you chose to
project out of spacetime into a subjective reference frame. All the physics
in SR/GR is done within spacetime, not subjective frames of reference.
Anyway, this has been done to death. The equations speak for themselves. I
found them relatively clear.
> Ingvar
Matthew
Matthew Pocock
> > On Friday 20 April 2007, Ingvar Johansson wrote:
> >> Matthew Pocock schrieb:
> >>> This seems to be in a similar vein with others on this list who seem to
> >>> believe that their personal experience of, for example, the sunyness of
> >>> the weather in their city is a statment of truth about reality because
> >>> they have experienced it. I'd just hedge and say it is a truth about
> >>> their observation of reality, since I am inclidend to believe that they
> >>> observed sunny weather. However, my time in and around biology has
> >>> taught me that we mostly all wear rose-tinted spectacles all the time,
> >>> and that you nearly never see what is actually there, as it is realy is
> >>> hence my insistance on the hedge.
> >> Another point. You are putting forward an *inconsistent position*. On
> >> the one hand you say that we can never know anything, but on the other
> >> hand you also say that you *know* that we all "wear rose-tinted
> >> spectacles all the time".
> >
> > I don't think I said that I *know* :) I said that "my time in and around
> > biology has taught me ..."
>
> No, you didn't; you said 'know'.
Just a point of fact, but if you look up to the paragraph at the top which is
a direct quote of what I originally wrote, you will see that I said:
"However, my time in and around biology has taught me that we mostly all
wear rose-tinted spectacles all the time, and that you nearly never see what
is actually there, as it is realy is hence my insistance on the hedge."
If you are going to use my own words against them, please have the courtesy to
ensure that you are in fact using my own words, and not what you would have
liked me to write.
Unless you where refering to a different place that I mentioned rose tinted
spectacles? If so, send me the google groups permalink and I'll duly
apologise.
> > What I thought I'd explained is that I am taking the position that every
> > observation depends upon both the means of observation and the thing
> > observed. I go further, and claim that to deny this is likely to lead to
> > systematic misconceptions about the thing observed which arise due to
> > artefacts of the means of observation. You can't dissagree with this,
> > surely?
>
> No, I agree.
Good. It has been tortuous getting this far. May we build upon this moment of
shared understanding fruitfully.
> But if this is what you have wanted to say all along, then
> you are really bad at giving explanations.
Possibly. We clearly have had a problem with mutual understanding. Perhaps the
difference in useage of words in the sciences and in philosophy and how
examples are used to support arguments has contributed to a certain degree.
That would be somewhat ironic, given the situation, wouldn't it.
Matthew
Phillip Lord
>> Again, if there is a physicist listening, please put us out of
>> our misery and tell us what is realy going on. Please.
>>
>> As I said earlier, SR (and GR, which superceeded it) doesn't
>> restrict any of these dimensions to real numbers.
IJ> And I said in one of my mails (to Cristian Cocos) that his
IJ> expression 'dx4' abstracts away from the fact that it can be
IJ> written 'ict'. The letter i represents the imaginary number 'the
IJ> root of minus one', and everything that is multiplied by i
IJ> becomes thereby an imaginary number. A complex number is a real
IJ> number plus an imaginary number; imaginary numbers figure in SR
IJ> but only in connection with the time variable. This is one of
IJ> several ways to make clear that the time dimension in SR cannot
IJ> be put on a par with the spatial dimensions.
Ingvar
Which is sort of the point. We are all aware that some circumstances
of SR can be represented as a Newtonian mechanics, where time has a
clear bona fide distinction from space. (Incidentally, I suspect we
are all also aware of what a complex number is).
As you say, Cristians equations use a representation which "abstract
away" from time. Of course, it may be possible to project a
representation where time is strongly distinguishable from space --
after all a projection into a Newtonian frame of reference (where the
situation allows it) is one such.
From which I draw the conclusion, that the distinction between space
and time is one that you either use, or do not use depending on which
representation you want to use.
Which is fine. The bona fide/fiat nature of a distinction is one that
you make, based on the circumstances and functional requirements of
your model. Or, in short, the distinction between bona fide and fiat
distinctions is a fiat distinction.
Have I got this right?
Phil
Ingvar Johansson
>>>>>> "IJ" == Ingvar Johansson <ingvar.j...@ifomis.uni-saarland.de> writes:
>>>>>>
>
> >> Again, if there is a physicist listening, please put us out of
> >> our misery and tell us what is realy going on. Please.
> >>
> >> As I said earlier, SR (and GR, which superceeded it) doesn't
> >> restrict any of these dimensions to real numbers.
>
> IJ> And I said in one of my mails (to Cristian Cocos) that his
> IJ> expression 'dx4' abstracts away from the fact that it can be
> IJ> written 'ict'. The letter i represents the imaginary number 'the
> IJ> root of minus one', and everything that is multiplied by i
> IJ> becomes thereby an imaginary number. A complex number is a real
> IJ> number plus an imaginary number; imaginary numbers figure in SR
> IJ> but only in connection with the time variable. This is one of
> IJ> several ways to make clear that the time dimension in SR cannot
> IJ> be put on a par with the spatial dimensions.
>
>
>
> Ingvar
>
>
> Which is sort of the point. We are all aware that some circumstances
> of SR can be represented as a Newtonian mechanics,
I don't know what you mean by the last sentence. What kind of
"circumstances" are you talking about? But the following holds true:
(a) taken in isolation, *each* inertial frame of reference in SR follows
the laws of Newtonian Mechanics.
(b) *no* transformation of values from one inertial frame to another is
the same in SR and NM; in SR there are Lorentz transformations and in NM
there are Galilei transformations.
> where time has a
> clear bona fide distinction from space. (Incidentally, I suspect we
> are all also aware of what a complex number is).
>
Well, Matthew Pocock seemed not be aware of the fact that I had from the
start brought in imaginary numbers.
> As you say, Cristians equations use a representation which "abstract
> away" from time.
I would say that the equation he used has abstracted both space and time
away, and is a purely mathematical formula. The fact that Minkowskian
space can be investigated also from a purely mathematical point of view,
does not imply that the *physical theory of SR* does not contain a
distinction between three spatial and one temporal dimension.
> Of course, it may be possible to project a
> representation where time is strongly distinguishable from space --
> after all a projection into a Newtonian frame of reference (where the
> situation allows it) is one such.
>
As long as one sticks to SR as a theory with more than mathematical
content, then SR contains - in and of itself - an essential distinction
between three spatial and one temporal dimension.
> >From which I draw the conclusion, that the distinction between space
> and time is one that you either use, or do not use depending on which
> representation you want to use.
>
Re-read what I have said above.
> Which is fine. The bona fide/fiat nature of a distinction is one that
> you make, based on the circumstances and functional requirements of
> your model.
But the model of SR, which we have been discussing, contains a
distinction between spatial and temporal dimensions, even though these
dimensions follow other laws in SR than in NM.
> Or, in short, the distinction between bona fide and fiat
> distinctions is a fiat distinction.
>
No, the distinction between 'bona fide' and 'fiat' is a bona fide
distinction.
> Have I got this right?
>
No.
Ingvar
> Phil
Matthew Pocock
and utterly my final post on this topic. Since Ingvar clearly does not belive
me, and also does not belive the maths, I shall quote some people who are (or
where) in a position to know better than any of us what is going on.
"From henceforth, space by itself, and time by itself, have vanished into the
merest shadows and only a kind of blend of the two exists in its own right."
Herman Minkowski, 1908, in an address to the 80 th Assembly of German Natural
Scientists and Physicians.
In the general theory of relativity the doctrine of space and time, or
kinematics, no longer figures as a fundamental independent of the rest of
physics. (Albert Einstein, 1919)
The common-sense distinctions we may make between space, time, objects,
gravity, motion, mass, are all frame-of-reference dependent in GR. They even
differ for the /same/ observer at /different subjective times/ if that
observer is in an accelerating frame of reference - which we all are, as we
are rotating about the Earth, in a gravity well. GR does have invariant
properties of systems, but they aren't the ones Newton would have recognised
and they for sure aren't the ones we meet in day to day life.
It follows that the distinctions made between space and time from my personal
point of view 5 min ago, are ever so slightly different to those I make now.
Hence, assuming GR, there are no inertial frames of reference that provide
the kind of space/time split invisaged by BFO that can possibly be inhabited
by any of us on Earth.
If what we are interested in is modeling reality, then we must reject space
and time as distinct entities. If we are interested in modeling something
that is good enough for every day circumstances, then the numerically small
the differences between a reality described by assuming falsely that we are
in a common inertial frame are well within acceptable limits. Either way, the
things considered 'foundational' have been shown to be theory-specific (which
is where we got on this train), and the most accurate theories to date don't
fit either BFO or our common-sense intuitions that space and time are
seperate and distinguishable.
Anyway, as I promissed, I will never ever (promiss on my mother's grave)
contribute to this thread/topic again. Even if baited. Or offered sweeties.
Honest. Scouts honor. Stick a needle...
Matthew
Phillip Lord
>>
>> Which is sort of the point. We are all aware that some
>> circumstances of SR can be represented as a Newtonian mechanics,
IJ> I don't know what you mean by the last sentence. What kind of
IJ> "circumstances" are you talking about? But the following holds
IJ> true:
IJ> (a) taken in isolation, *each* inertial frame of reference in SR
IJ> follows
IJ> the laws of Newtonian Mechanics.
Thats the circumstances.
IJ> Well, Matthew Pocock seemed not be aware of the fact that I had
IJ> from the start brought in imaginary numbers.
>> As you say, Cristians equations use a representation which
>> "abstract away" from time.
IJ> I would say that the equation he used has abstracted both space
IJ> and time away, and is a purely mathematical formula. The fact
IJ> that Minkowskian space can be investigated also from a purely
IJ> mathematical point of view, does not imply that the *physical
IJ> theory of SR* does not contain a distinction between three
IJ> spatial and one temporal dimension.
Yes, as I have said, and you agree, we can make a distinction between
spatial and temporal dimensions, if we choose to use a mathematical
model which uses this.
>> Of course, it may be possible to project a representation where
>> time is strongly distinguishable from space -- after all a
>> projection into a Newtonian frame of reference (where the
>> situation allows it) is one such.
>>
IJ> As long as one sticks to SR as a theory with more than
IJ> mathematical content,
This I don't understand.
>> Or, in short, the distinction between bona fide and fiat
>> distinctions is a fiat distinction.
>>
IJ> No, the distinction between 'bona fide' and 'fiat' is a bona
IJ> fide distinction.
>> Have I got this right?
>>
IJ> No.
Thats a pity. I thought I was started to understand what you meant.
Phil
Ingvar Johansson
> I promissed not to contribute to this thread any more. This will be absolutely
> and utterly my final post on this topic. Since Ingvar clearly does not belive
> me, and also does not belive the maths, I shall quote some people who are (or
> where) in a position to know better than any of us what is going on.
>
I have once said that I will not spend more time on discussing SR with
you, but confronted with these quotations I am happy to write even this
mail.
> "From henceforth, space by itself, and time by itself, have vanished into the
> merest shadows and only a kind of blend of the two exists in its own right."
> Herman Minkowski, 1908, in an address to the 80 th Assembly of German Natural
> Scientists and Physicians.
>
First, A "blend" of two things does not make the blended things
identical. Second, the "shadows" are said to be there. Even though it
can truly be said that "space by itself" and "time by itself" have no
meaning in SR, it CANNOT truly be said that SR can as a physical (and
not merely mathematical) theory do without a distinction between the
spatial dimensions and the temporal one. Minkowski talks of space and
time as two kinds of "shadows", not as if the distinction between them
can be completely collapsed.
> In the general theory of relativity the doctrine of space and time, or
> kinematics, no longer figures as a fundamental independent of the rest of
> physics. (Albert Einstein, 1919)
I have never claimed that in SR or GR the spatial and the temporal
dimensions are *independent* of each other; only that they cannot be
regarded as being of the same kind. Let me give an example from SR
(where we only have to bother about one spatial dimension). Assume that
in a particular inertial frame A, we have two events (E1 and E2) that
are x meter and y seconds apart. In every other inertial frame the
events E1 and E2 keep their identity, but the values for x and y cannot
be the same as in frame A. However, in every inertial frame whatsoever
E1 and E2 will be assigned one spatial distance and one temporal interval.
Ingvar
Ingvar Johansson
>
> Yes, as I have said, and you agree, we can make a distinction between
> spatial and temporal dimensions, if we choose to use a mathematical
> model which uses this.
>
No pure mathematical model brings in space and time, but the mathematics
of the physical theory of SR brings with it an important distinction
between three spatial and one temporal dimension as long as at it has
any interesting physical content.
>
>
> >> Of course, it may be possible to project a representation where
> >> time is strongly distinguishable from space -- after all a
> >> projection into a Newtonian frame of reference (where the
> >> situation allows it) is one such.
> >>
>
> IJ> As long as one sticks to SR as a theory with more than
> IJ> mathematical content,
>
> This I don't understand.
>
Perhaps because you have never thought of the distinction between pure
mathematics and mathematical physics. The fact that physical truths can
be represented mathematically do not turn them into mathematical truths.
best,
Ingvar
>
> >> Or, in short, the distinction between bona fide and fiat
> >> distinctions is a fiat distinction.
> >>
>
> IJ> No, the distinction between 'bona fide' and 'fiat' is a bona
> IJ> fide distinction.
>
> >> Have I got this right?
> >>
>
> IJ> No.
>
>
> Thats a pity. I thought I was started to understand what you meant.
>
> Phil
>
> >
Phillip Lord
IJ> As long as one sticks to SR as a theory with more than
IJ> mathematical content,
>>
>> This I don't understand.
>>
IJ> Perhaps because you have never thought of the distinction
IJ> between pure mathematics and mathematical physics. The fact that
IJ> physical truths can be represented mathematically do not turn
IJ> them into mathematical truths.
Maybe. For me, I think it's more that I don't understand how you
determine which mathematical representation defines physical truth and
which does not.
What are the criteria for this determining this?
Phil
Ingvar Johansson
I am not determining anything of this kind; I am trying to make explict
what the physicists are doing. In my opinion, *no* mathematical
representation can *define a physical truth*. To me, you seem think
"top-down", from mathematics to physical reality, whereas I am firmly
convinced one should think "bottom-up". First there is an aspect of
physical reality, and then it can be asked whether this aspect can be
given a mathematical representation. Such representations are *never*
pure mathematics. They rely on linking (mostly) the real numbers to a
physical dimension such as length, time, mass, and electric current.
This linking is possible only if the determinates of the physical
dimension in question allow themselves to be ordered the linear way the
real numbers are ordered. When it is possible, it tells us something
about the world. When several physical dimensions have been quantified,
one can try to find laws that relate them to each other. In order to
come to know what laws are more or less truthlike, observations and
experiments have to be made. Why? Answer: because our primary link to
physical reality is perception.
Ingvar
> What are the criteria for this determining this?
>
> Phil
>
>
> >
Phillip Lord
>> Maybe. For me, I think it's more that I don't understand how you
>> determine which mathematical representation defines physical
>> truth and which does not.
>>
IJ> I am not determining anything of this kind; I am trying to make
IJ> explict what the physicists are doing. In my opinion, *no*
IJ> mathematical representation can *define a physical truth*. To
IJ> me, you seem think "top-down", from mathematics to physical
IJ> reality, whereas I am firmly convinced one should think
IJ> "bottom-up".
No, I don't think so. I just stop at the mathematical
representation. We have some observations both with and without
perturbation of the system. Then we have some mathematics (or other
model, informal, or formal), which describes these.
I stopped worrying about whether the mathematics represents reality
sometime ago, as I wasn't sure what "reality" means. Especially with
physics, I find that this makes life an awful lot easier. I am
guessing that you have had the opposite experience.
IJ> This linking is possible only if the determinates of the
IJ> physical dimension in question allow themselves to be ordered
IJ> the linear way the real numbers are ordered. When it is
IJ> possible, it tells us something about the world.
Yes. We can fit our observations into some mathematical framework.
IJ> When several physical dimensions have been quantified, one can
IJ> try to find laws that relate them to each other. In order to
IJ> come to know what laws are more or less truthlike, observations
IJ> and experiments have to be made.
Yes. But then you can have more one mathematical model. Does the earth
go around the sun, the sun go around the earth, do they both rotate
around a point, or is it a meaningless question? We have models for
all of these. And they can't all be experimental differentiated.
Phil
Ingvar Johansson
>>>>>> "IJ" == Ingvar Johansson <ingvar.j...@ifomis.uni-saarland.de> writes:
>>>>>>
>
> >> Maybe. For me, I think it's more that I don't understand how you
> >> determine which mathematical representation defines physical
> >> truth and which does not.
> >>
>
> IJ> I am not determining anything of this kind; I am trying to make
> IJ> explict what the physicists are doing. In my opinion, *no*
> IJ> mathematical representation can *define a physical truth*. To
> IJ> me, you seem think "top-down", from mathematics to physical
> IJ> reality, whereas I am firmly convinced one should think
> IJ> "bottom-up".
>
> No, I don't think so. I just stop at the mathematical
> representation. We have some observations both with and without
> perturbation of the system. Then we have some mathematics (or other
> model, informal, or formal), which describes these.
>
No *pure mathematics* can lay claim to describe physical reality! The
statement "in all inertial frames of reference dR2 = dx2 + dy2 + dz2 +
du2 " is physically meaningless before you link the variables to
physical dimensions.
But if you first say "let x, y, and z be three spatial dimensions, t the
temporal dimension, and c the velocity of light", then the statement "in
all inertial frames of reference dR2 = dx2 + dy2 + dz2 + (ict)2 " lays
claim to describe physical reality. If SR is true, it does; otherwise not.
> I stopped worrying about whether the mathematics represents reality
> sometime ago, as I wasn't sure what "reality" means. Especially with
> physics, I find that this makes life an awful lot easier. I am
> guessing that you have had the opposite experience.
I have no problems whatsoever in thinking reality-claims away from
physical theories and regard them as merely being about fictive objects
and phenomena (as many obsolete theories are; phlogiston is a fiction);
or, for that matter, abot hypothetically assumed objects and phenomena
(e.g. entanglement in QM). But if one stops talking about "reality"
altogether, one cannot make sense of the importance that observations
and experiments have in physics. Perception is our primary link to reality.
> IJ> This linking is possible only if the determinates of the
> IJ> physical dimension in question allow themselves to be ordered
> IJ> the linear way the real numbers are ordered. When it is
> IJ> possible, it tells us something about the world.
>
>
> Yes. We can fit our observations into some mathematical framework.
>
Not always. Here is a task for you: try to map all our observations of
different kinds of two-dimensional closed geometrical shapes on the real
number line (the way this is done with length, time, mass, etc.). Why
hasn't this mapping been done already?
>
> IJ> When several physical dimensions have been quantified, one can
> IJ> try to find laws that relate them to each other. In order to
> IJ> come to know what laws are more or less truthlike, observations
> IJ> and experiments have to be made.
>
> Yes. But then you can have more one mathematical model.
But, to repeat, there are *no* purely mathematical models of physical
reality!
Ingvar
> Does the earth
> go around the sun, the sun go around the earth, do they both rotate
> around a point, or is it a meaningless question? We have models for
> all of these. And they can't all be experimental differentiated.
>
>
> Phil
>
> >
Phillip Lord
IJ> No *pure mathematics* can lay claim to describe physical
IJ> reality!
That's fine. I don't make this claim.
>> I stopped worrying about whether the mathematics represents
>> reality sometime ago, as I wasn't sure what "reality"
>> means. Especially with physics, I find that this makes life an
>> awful lot easier. I am guessing that you have had the opposite
>> experience.
IJ> But if one stops talking about "reality" altogether, one cannot
IJ> make sense of the importance that observations and experiments
IJ> have in physics.
I think you mean that you cannot make sense of this importance. It all
seems fine to me. Being able to describe observation and predict the
response to perturbation seems both important and rather fun.
IJ> Perception is our primary link to reality.
So, you keep saying. Reminds me of the "reality is an illusion caused
by lack of alcohol" t-shirt that you (or I at least) see periodically.
IJ> This linking is possible only if the determinates of the
IJ> physical dimension in question allow themselves to be ordered
IJ> the linear way the real numbers are ordered. When it is
IJ> possible, it tells us something about the world.
>>
>>
>> Yes. We can fit our observations into some mathematical
>> framework.
>>
IJ> Not always.
Apologies, you misread my text, which wasn't very clear. I meant that
in this case, fitting our observations against the reals tells us that
we can fit these observations to some (specifically, the reals)
framework.
Clearly, we can't do this in general, or at least, we haven't show that
we can do this in general. I don't think that we have shown that we
can't either, which is good news.
IJ> When several physical dimensions have been quantified, one can
IJ> try to find laws that relate them to each other. In order to
IJ> come to know what laws are more or less truthlike, observations
IJ> and experiments have to be made.
>>
>> Yes. But then you can have more one mathematical model.
IJ> But, to repeat, there are *no* purely mathematical models of
IJ> physical reality!
And to repeat, I don't know what you mean by physical reality.
Why do I care about this? Well, I don't and I stopped worrying about
it a long time ago.
Which is why I am rather confused by criteria like "does it exist",
"is it a real-world entity" while trying to build models. I have no
idea what the questions mean, yet alone how to answer them. I need
something more straightforward, understandable and testable to work
with.
This probably stems from my biological upbringing. Physicists have all
these grand theories. Biologists worry about what happens when you
kick a dog or grind up a fly.
Phil
Matthew Pocock
> This probably stems from my biological upbringing. Physicists have all
> these grand theories. Biologists worry about what happens when you
> kick a dog or grind up a fly.
That one I can answer - you get dogs and flys that don't work very well any
more.
Matthew
>
>
> Phil
>
>
>
>
Ingvar Johansson
> Why do I care about this? Well, I don't and I stopped worrying about
> it a long time ago.
>
> Which is why I am rather confused by criteria like "does it exist",
> "is it a real-world entity" while trying to build models. I have no
> idea what the questions mean, yet alone how to answer them.
According to modern physics and chemistry, oxygen exists but phlogiston
does not. Notwithstanding the fact that once two competing models for
combustion were built; one used the concept of "oxygen" and the other
used "phlogiston".
> I need
> something more straightforward, understandable and testable to work
> with.
>
>
> This probably stems from my biological upbringing. Physicists have all
> these grand theories.
which have made it possible, among other things, to blow atomic bombs,
put a man on the moon, and create laser beams.
> Biologists worry about what happens when you
> kick a dog or grind up a fly.
Contemporary biologists worry a lot about molecular biology; especially
about the grand theory of DNA and its functioning.
best wishes - and end of this discussion,
Ingvar
Phillip Lord
IJ> According to modern physics and chemistry, oxygen exists but
IJ> phlogiston does not. Notwithstanding the fact that once two
IJ> competing models for combustion were built; one used the concept
IJ> of "oxygen" and the other used "phlogiston".
God I am confused. Now, we're back to modelling the concept of
oxygen.
>> I need something more straightforward, understandable and
>> testable to work with.
No tests then. Pity.
>> Biologists worry about what happens when you kick a dog or grind
>> up a fly.
IJ> Contemporary biologists worry a lot about molecular biology;
IJ> especially about the grand theory of DNA and its functioning.
I know; I am a contemporary biologist. There is no grand theory of
DNA; there are lots of little pieces which we try to stick together.
IJ> best wishes - and end of this discussion, Ingvar
Well, thank you for your time. I have learnt a lot.
Phil