POnt - Personal Ontology 1.0 alpha
Mázsa Péter
let me introduce you POnt - Personal Ontology 1.0 alpha.
Home: http://theunitedpersons.org/pont
Owl file: https://github.com/mazsa/Personal-Ontology/raw/master/pont.owl
It is a modification of BFO, based on this Constitution:
http://theunitedpersons.org/constitution
Motivation:
I'm a big fan of Parts - A Study in Ontology by Simons
http://www.amazon.com/Parts-Study-Ontology-Peter-Simons/dp/0199241465
and BFO, and, however short it is (< 1 page), I spent a lot of time
revealing/creating the Constitution above. I wanted to merge them all.
My central concern was that agents and persons as defined by the
Axioms 1-4. http://theunitedpersons.org/constitution/axiom of the
Constitution didn't fit well in BFO 1.1. I think we need a category
for Agents: Independent Continuants who are not necessarily Material
Entities (or Boundaries or Sites). E.g., I don't think we should
regard Constituted Persons (e.g. corporation / universitas
http://penelope.uchicago.edu/Thayer/E/Roman/Texts/secondary/SMIGRA*/Universitas.html
) as necessarily Objects.
Main deviation:
This is why I must replace the root BFO-category Entity for root
POnt-category Information. In POnt, (material) Objects are special
cases of Information ( = are constrained Information), ensuring a
place for not necessarily material independent continuants.
(As a matter of fact, I have another motivation too: I think it may be
useful to approach from this point of view the trade-off between
"private ownership" of information and the "public domain"
"Locke (1690 http://www.ilt.columbia.edu/academic/digitexts/locke/second/locke2nd.txt
) was one of the earliest writers to argue that ideas should be
appropriated by those who originally produced them and thereafter
protected for a period of time under the principle of natural law for
the benefit of the public. He identified the trade-off between private
ownership and the public domain"
https://www.stanford.edu/group/song/papers/ScienceandPropertyARLSS.pdf
)
Implication:
Notice that my suggestion (the replacement) implies an answer to this question:
"[...] it is important to clarify what the problem is not. We are not
asking whether the metaphorical interpretation of the universe as a
computer is more useful than mislead- ing. We are not even asking
whether an informational description of the universe, as we know it,
is possible, at least partly and piecemeal. [...] We are asking
whether the universe in itself could essentially be made of
information [...]"
http://num.math.uni-goettingen.de/schaback/info/mat/floridi_open_problems.pdf
p. 574
The replacement (of Entity for Information) implies that universe in
itself could possibly essentially be made of information with natural
processes, including causation, as special cases of it (Cf.
http://en.wikipedia.org/wiki/Digital_physics ).
Notice further that
- the replacement does not imply predetermination
- the definition of Agent is "Freedom of some Information"
- the basic concept "freedom" is not necessarily (but possibly)
ontologic, it is "at least" epistemologic, i.e., the replacement does
not exclude predetermination either - we are agnostic on this topic
and cf. http://wiki.lesswrong.com/wiki/Free_will_(solution)
Structure:
Some Agents [CommonKnowledgePartitionPart
http://theunitedpersons.org/constitution/axiom/ca2 , including
FinestPart and NonFinestPart] are equivalent to some Persons
[SocialPerson http://theunitedpersons.org/constitution/axiom/ca3 ,
including Individual and State], based on a Theorem (which includes
http://planetmath.org/encyclopedia/PartitionIsEquivalentToAnEquivalenceRelation.html
).
And these Persons are equivalent to some Objects [SocialPersonObject,
including Body and Territory], based on a Presumption.
Of course, some SocialPersons (States, Individuals) may have
constitutions, and some ConstitutedPersons may have "incarnations"
among Objects (e.g. the United Persons may affiliate some States:)
Question:
The replacement of Entity for Information is the main price I think we
should pay for Persons (or for not necessarily material independent
continuants) in our ontology. What do you think of it as the
BFO-community?
Thank you:
Peter Mazsa http://mazsa.com
Janna Hastings
I am curious. Can you give a definition for Information as you've used it?
Thanks
Janna
2010/11/22 Mázsa Péter <pe...@mazsa.com>:
> --
> You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
> To post to this group, send email to bfo-d...@googlegroups.com.
> To unsubscribe from this group, send email to bfo-discuss...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/bfo-discuss?hl=en.
>
>
Mázsa Péter
> I am curious. Can you give a definition for Information as you've used it?
Hi Janna,
the "natural" answer is that I use it as a basic concept, i.e. without
a definition (just like BFO uses Entity). But this does not tell the
whole story. I use Information (vs matter) as some(thing) not
necessarily local and/or temporal, but (like matter) likely causal
(cf. eg. http://lesswrong.com/lw/qr/timeless_causality ).
P.
Barry Smith
BS
Mázsa Péter
> Nice try. You might want to check this:
> http://www.gap5.de/proceedings/pdf/479-491_jansen.pdf
Thank you Ludger, I like it very much! We are pursuing the same...
subject:) (e.g. "[...] there are non-natural persons that have no
intelligence nor emotions of happiness or misery of there own, but
still are agents [= persons @Jansen] to which actions and their merit
are appropriated" Jansen,2003 p. 486.)
Main differences:
- I think agency is not what constitutes personship, but a necessary
condition of it. In POnt, there may be agents who are not persons. Or,
this is at least not excluded. (Cf. eg.
http://www.amazon.com/Personal-Agency-Metaphysics-Mind-Action/dp/0199217149
)
- I like your category "status persons" ("things that exist only
because we believe them to exist" Jansen cites Searle
"belief-dependent non-beliefs" Jansen,2003 p. 479). I think the
concept of person is itself a "status". In POnt, Individual (or:
[FinesPart, Body, Individual]) is not a "natural person" but a "status
person" as well, due to common knowledge (common knowledge as I use:
http://arxiv.org/PS_cache/cs/pdf/0006/0006009v1.pdf pp. 14-15.).
Natural [person] (be it a result of evolution or creation) is not
preferred over non-natural [person] (be it a result of anyone's
creation or belief).
- In POnt, there are 2 possible origins of your status objects: (1)
States and Individuals, who are status objects because they are
elements of a common knowledge partition of agents, based on Axiom 2
("2. All parts of a common knowledge partition of agents are agents")
and (2) ConstitutedPersons who are status objects based on Axiom 4
("4. All ‘persons’ declared by persons and able to speak with one
voice are persons.")
- Both of us think that status persons are not, or not exclusively,
(material) Objects. (In POnt, persons are originally not material
Objects, however some persons, SocialPersons, are equivalent to some
Objects, Social PersonObjects, and some ConstitutedPersons may have
"incarnations" among Objects as well). But you are not explicit about
it, I mean how would you implement your system in BFO 1.1, where you
can not find a place for not necessarily material
IndependentContinuants?
> Personally, I would try to seperate the formal ontology of persons from the
> person ethos thing.
2 answers:
1) You are right. For me, this 2 above are inseparable: I'm generating
persons from 'persons' (said to be persons) in the following way:
"1. All ‘persons’ are agents.
2. All parts of a common knowledge partition of agents are agents.
3. All members of a possibly fair society of agents are persons.
4. All ‘persons’ declared by persons and able to speak with one voice
are persons."
http://theunitedpersons.org/constitution/axiom
The Axiom-system is determined by the claim that the concept of person
which is generated by it should be invariant to "any origin, natural
or authored" and to "any consent, by common right or by constitution".
So the person ethos thing (the claim of invariance) determines the
structure of Axioms.
This unseparability is undeniably a constraint. But maybe a good kind
of constraint: I think it makes possible to determine the meaning of
the concept of person, without any contingency, overdetermination or
underdetermination.
(By the way, ethical foundation of personhood is not new, cf. e.g.
"[...] this book will begin with an ethical assumption about the
nature of persons, which it will then take as a critical and defining
starting point for further metaphysical investigation into the kind
'person'." http://www.amazon.com/Bounds-Agency-Carol-Rovane/dp/0691017166
p. 5.)
2) Formally, you can work with the bare axioms without any reference
to their origin. Starting from here, the formal ontology of person is
completely separated from the person ethics thing.
> By the way: What is a "common knowledge partition part"?
When we speak about "common knowledge partition part", we use Axiom 2
above: if the set B is a part (here: element) of a partition [<==> a
collection of disjoint nonempty subsets of the set A whose union is
all of the set A] of a set A of agents, and the partition is common
knowledge among the elements of the set A of agents, then the set B is
an agent. I call B CommonKnowledgePartitionPart.
Examles are the equivalent [FinesPart, Body, Individual] and the
equivalent [NonFinestPart, Territory, State]
I'm curious... what do you think?
Peter
> ----- Original Message ----- From: "Mázsa Péter" <pe...@mazsa.com>
> To: <bfo-d...@googlegroups.com>
> Sent: Monday, November 22, 2010 9:51 AM
> Subject: [bfo-discuss] POnt - Personal Ontology 1.0 alpha
Mázsa Péter
> Your proposal, unfortunately, has the unacceptable consequence that every
> instance of material entity is an instance of information.
> I believe that if you really believe this, then DOLCE would be a more
> suitable environment for your work; or potentially also HL7 RIM.
> BS
Barry, I think BFO you initiated is superior to any other ontologies I
checked - BFO is my revealed preference:)
Regarding that both the Entity of BFO and the Information of POnt are
basic concepts, i.e. both without definition (thanks Janna!), I do not
insist on Information as a root category. POnt can call the root
category Entity, so I need not give up BFO-tradition + will not lose
generality (thanks Barry!). I may like information to be the
"infrastucture" of matter but you are right: for an ontology this
would be a superfluous presumption and a logical fallacy of this kind:
http://www.philosophyexperiments.com/mary/Default.aspx .
What is absolutely necessary for POnt is that there should be a place
for Independent Continuants who are not necessarily Material Entities
(or Boundaries or Sites).
Of course I am not informed enough on the history of BFO, so I am not
able to decide whether it is a bug of BFO or a feature - hopefully a
feature, and in this case POnt will be just a fork of BFO, not a
suggestion for improvement of it.
***
I got 2 private questions independently:
"I still do not understand the square bracket terms in POnt."
"What are the brackets for?"
Square brackets are generated by SWOOP 2.3beta4 if you declare 2
categories to be equivalent. If '=' is 'is equivalent to', I declared
that State = Territory = NonFinestPart, and that Individual = Body =
FinestPart, and that SocialPerson = SocialPersonObject =
CommonKnowledgePartitionPart. E.g. the meaning of [State, Territory,
NonFinestPart] (cf. http://theunitedpersons.org/pont ) is that the
categories of State, Territory and NonFinestPart are equivalent.
Alan Ruttenberg
>
> What is absolutely necessary for POnt is that there should be a place
> for Independent Continuants who are not necessarily Material Entities
> (or Boundaries or Sites).
Why do you need them to be independent? In what way would dependent
continuant not be correct? Are there some things that do not have a
material basis?
For example, we would consider "Legal person" as a role, and could
then define a class of material entities that bear this role. Such a
class does not have to be a subclass of homo sapiens.
-Alan
Barry Smith
Mázsa Péter
> 2010/11/24 Mázsa Péter <pe...@mazsa.com>:
>> What is absolutely necessary for POnt is that there should be a place
>> for Independent Continuants who are not necessarily Material Entities
>> (or Boundaries or Sites).
>
> continuant not be correct? Are there some things that do not have a
> material basis?
1. You are a person. Are you sure you think that a material basis is
absolutely necessary for you? If you had the opportunity do you think
you would "upload" yourself (= convert yourself into information
http://en.wikipedia.org/wiki/Mind_uploading ) before you die? [I think
I would. And I would use e.g. Git
http://en.wikipedia.org/wiki/Git_(software) for version control, i.e.
branching and merging of my bodiless minds & minds "downloaded" into
other bodies criss-crossing the Earth & the Universe and being happy
not to be dead:]
2. Suppose a (likewise hypothetical) Artificial Intelligence [AI]
running on the internet is a person. Do you think that she has a
material basis / is an independently continuant material object
(bearing a role of person)? [I think it would not be useful to regard
her as something that has a material basis. I think you think this AI
is an independent continuant, but I can't imagine how you think she is
a material object. I think filing them into the existing categories of
BFO would be like working with celestial spheres: still usable but
ugly.]
I think that what we think of as persons are "things" that do not
necessarily and intuitively have a material basis. Are you convinced
by the thought experiments above?
On 24 November 2010 18:48, Barry Smith <phis...@buffalo.edu> wrote:
> did you look at IAO, which extends Buffalo, and places information artifacts
> under BFO:generically dependent continuants:
> http://code.google.com/p/information-artifact-ontology/
The uploaded and version controlled (information artifact) me might be
somehow depend on the decesed me - perhaps otherwise than "Borges" on
"me" in http://www.amherstlecture.org/perry2007/Borges%20and%20I.pdf ,
but both of us may live with the consciousness that you filed (the
uploaded) me into the category BFO:generically dependent continuants.
But what would be the independent continuant material basis of the AI
on the internet? A set of electrons / photons?
On 24 November 2010 17:42, Alan Ruttenberg <alanrut...@gmail.com> wrote:
> For example, we would consider "Legal person" as a role, and could
> then define a class of material entities that bear this role. Such a
> class does not have to be a subclass of homo sapiens.
Persons are continuants without doubt.
As I understand you think that
- you as a homo sapiens are an instance of independently continuant
material object, and
- a legal person is a class of independently continuant material
objects [or a class of other legal persons and independently
continuant material objects, etc.] bearing a (dependently continuant)
role of "legal person".
The question whether personship is a role has a very long history, my
favorite contemporary texts on this history are
http://www.amazon.com/Category-Person-Anthropology-Philosophy-History/dp/0521277574
and
http://www.amazon.com/Pluralism-Personality-State-Ideas-Context/dp/0521022630
(cf. of course http://oregonstate.edu/instruct/phl302/texts/hobbes/leviathan-contents.html
esp ch. 16. and 2nd part where personship is a role and
http://www.amazon.com/Reasons-Persons-Oxford-Paperbacks-Parfit/dp/019824908X
on the logical construction of personship).
If you think that you are an independently continuant material object
bearing a role of "natural person", then you think personship is a
role. (Or you think that natural persons are unique among persons in
being independently continuant material objects without bearing a
role?)
I think personship is not a role: roles are a kind of connection
between persons. As I wrote above I think that persons are not
independently continuant material objects bearing (dependently
continuant) roles but independent continuants who do not necessarily
and intuitively have a material basis ("Independent Continuants who
are not necessarily Material Entities (or Boundaries or Sites)"). E.g.
I think I am an independent continuant who is necessarily material as
a body (of a homo sapiens) and as an individual person but not as a
person.
The question remains whether the Person-Object divide has ontological
(vs. "mere" ethical) significance (like the Continuant-Occurent
divide). If you accept (as I do) or suppose for a moment that "An
objective fact is one that is invariant under all admissible
transformations" (Nozick: Invariances
http://www.amazon.com/Invariances-Structure-Objective-Robert-Nozick/dp/0674012453
p. 82., in general in part "2. Invariance and Objectivity" pp.
75-119., and in particular to ethics in chapter "Ethical Truth and
Ethical Objectivity", pp. 284-294.) you can see
- why I insist on invariance in my ethics,
- why I think that implementing the invariance-requirement is the way
how we can avoid a contingent definition of person.
My criticism of BFO "embodied" in POnt stems from the fact that I think that
- the (hypothetical) transformation of the origin of a person (natural
vs authored, e.g. material vs uploaded, or individual vs AI) is an
ontological transformation, and
- personship is invariant to this transformation as well.
This is why I think that POnt is a pure ontology, i.e., not
"contaminated" with ethics, and perhaps with relevance to BFO.
Peter
Alan Ruttenberg
> On 24 November 2010 17:42, Alan Ruttenberg <alanrut...@gmail.com> wrote:
>> 2010/11/24 Mázsa Péter <pe...@mazsa.com>:
>>> What is absolutely necessary for POnt is that there should be a place
>>> for Independent Continuants who are not necessarily Material Entities
>>> (or Boundaries or Sites).
>>
>> Why do you need them to be independent? In what way would dependent
>> continuant not be correct? Are there some things that do not have a
>> material basis?
>
> 1. You are a person. Are you sure you think that a material basis is
> absolutely necessary for you?
Yes.
> If you had the opportunity do you think
> you would "upload" yourself (= convert yourself into information
> http://en.wikipedia.org/wiki/Mind_uploading ) before you die? [I think
> I would. And I would use e.g. Git
> http://en.wikipedia.org/wiki/Git_(software) for version control, i.e.
> branching and merging of my bodiless minds & minds "downloaded" into
> other bodies criss-crossing the Earth & the Universe and being happy
> not to be dead:]
If possible, I would. Whether possible is an open question.
After I uploaded I would have a different material basis (some digital
form of memory)
In BFO we call such entities generically dependent continuants.
> 2. Suppose a (likewise hypothetical) Artificial Intelligence [AI]
> running on the internet is a person. Do you think that she has a
> material basis / is an independently continuant material object
> (bearing a role of person)?
The BFO approach would be to deal with that eventuality when and if it obtains.
In any case, I think, in your formulation, yes, it would have at least
one material basis
>[I think it would not be useful to regard
> her as something that has a material basis.
It isn't a matter of useful. It's a matter of true.
> I think you think this AI is an independent continuant, but I can't imagine how you think she is
> a material object.
I didn't say the AI was a material entity. I said it was a generically
dependent continuant.
> I think filing them into the existing categories of
> BFO would be like working with celestial spheres: still usable but
> ugly.]
Sorry to have offended.
> I think that what we think of as persons are "things" that do not
> necessarily and intuitively have a material basis. Are you convinced
> by the thought experiments above?
No.
-Alan
Barry Smith
On 24 November 2010 17:42, Alan Ruttenberg <alanrut...@gmail.com> wrote:
> 2010/11/24 Mázsa Péter <pe...@mazsa.com>:
>> What is absolutely necessary for POnt is that there should be a place
>> for Independent Continuants who are not necessarily Material Entities
>> (or Boundaries or Sites).
>
> Why do you need them to be independent? In what way would dependent1. You are a person. Are you sure you think that a material basis is
> continuant not be correct? Are there some things that do not have a
> material basis?
absolutely necessary for you? If you had the opportunity do you think
you would "upload" yourself (= convert yourself into information
http://en.wikipedia.org/wiki/Mind_uploading ) before you die? [I think
I would. And I would use e.g. Git
http://en.wikipedia.org/wiki/Git_(software) for version control, i.e.
branching and merging of my bodiless minds & minds "downloaded" into
other bodies criss-crossing the Earth & the Universe and being happy
not to be dead:]
2. Suppose a (likewise hypothetical) Artificial Intelligence [AI]
running on the internet is a person. Do you think that she has a
material basis / is an independently continuant material object
(bearing a role of person)? [I think it would not be useful to regard
her as something that has a material basis. I think you think this AI
is an independent continuant, but I can't imagine how you think she is
a material object. I think filing them into the existing categories of
BFO would be like working with celestial spheres: still usable but
ugly.]
I think that what we think of as persons are "things" that do not
necessarily and intuitively have a material basis. Are you convinced
by the thought experiments above?
BS
Mázsa Péter
based on the conversation I have set Entity back as a root category in
POnt https://github.com/mazsa/Personal-Ontology/raw/master/pont.owl :
thank you for your help.
On 29 November 2010 19:56, Alan Ruttenberg <alanrut...@gmail.com> wrote:
> 2010/11/29 Mázsa Péter <pe...@mazsa.com>:
>> Suppose a (likewise hypothetical) Artificial Intelligence [AI]
>> running on the internet is a person. Do you think that she has a
>> material basis / is an independently continuant material object
>> (bearing a role of person)?
>
> The BFO approach would be to deal with that eventuality when and if it obtains.
On 29 November 2010 20:01, Barry Smith <phis...@buffalo.edu> wrote:
>> 1. You are a person. Are you sure you think that a material basis is
>> absolutely necessary for you? If you had the opportunity do you think
>> you would "upload" yourself (= convert yourself into information
>> http://en.wikipedia.org/wiki/Mind_uploading ) before you die? [...]
>
> when this is shown to be possible, we will re-address the issue;
1. Thank you, this is a pragmatic answer, I can live with it.
2. Alan, in a way it have obtained: I raised this issue as a routine
http://plato.stanford.edu/entries/thought-experiment :)
On 29 November 2010 19:56, Alan Ruttenberg <alanrut...@gmail.com> wrote:
> 2010/11/29 Mázsa Péter <pe...@mazsa.com>:
>>>> What is absolutely necessary for POnt is that there should be a place
>>>> for Independent Continuants who are not necessarily Material Entities
>>>> (or Boundaries or Sites).
>>>
>>> Why do you need them to be independent? In what way would dependent
>>> continuant not be correct? Are there some things that do not have a
>>> material basis?
>>
>> I think you think this AI is an independent continuant, but I can't imagine how you think she is
>> a material object.
>
> I didn't say the AI was a material entity. I said it was a generically
> dependent continuant.
>
>> If you had the opportunity do you think
>> you would "upload" yourself (= convert yourself into information
>> http://en.wikipedia.org/wiki/Mind_uploading ) before you die? [...]
>
> If possible, I would. Whether possible is an open question.
> After I uploaded I would have a different material basis (some digital
> form of memory)
> In BFO we call such entities generically dependent continuants.
You say that AIs running on the internet would be generically
dependent continuants (not MaterialEntities) and all generically
dependent continuants have material bases, which are presumably
MaterialEntities, specifically, spatially dispersed sets of Objects.
You say that I'm an independent continuant while I'm flesh and blood,
"uploading" would change me (still me) into a dependent continuant
with a material basis (presumably a set of Objects), and "downloading"
again would presumably change me back into an independent continuant.
Similarly, if dependent continuant AIs running on the internet would
"download" themselves exclusively into robots partitioned in space,
they would presumably become independent continuants.
OK, I understand. Whether possible is an open question, but eventually
and occasionally I / AI might switch my / her ontological status in
BFO, without ceasing to be myself / herself, so this me / her is
jumping across ontological categories.
>> I think filing them into the existing categories of
>> BFO would be like working with celestial spheres: still usable but
>> ugly.]
>
> Sorry to have offended.
:)))
perhaps I was not PC: celestial spheres are perfect companies for us
to work with them, e.g. to predict planetary movements and
configurations; they are just ... under-attractive:) to my taste
On 29 November 2010 20:01, Barry Smith <phis...@buffalo.edu> wrote:
>> I think filing them into the existing categories of
>> BFO would be like working with celestial spheres: still usable but
>> ugly.]
>>
> there are so many problems facing use of BFO in support of established
> science, we cannot waste time on merely hypothetical issues; for this you
> may wish to use DOLCE
>
>> I think that what we think of as persons are "things" that do not
>> necessarily and intuitively have a material basis. Are you convinced
>> by the thought experiments above?
>>
> This is not the issue; the issue is how to ensure BFO is successful in
> performing the jobs it needs to perform today
> BS
OK, this is the (acceptable) pragmatic line. Of course I cannot insist
on the modification of the preferences of the BFO-community based on
my unproven thoughts.
But do I have the right to fork
http://en.wikipedia.org/wiki/Fork_(software_development) BFO at all? I
feel so that the default copyright for ontologies is
http://creativecommons.org/licenses/by/3.0/ but I'd appreciate if you
would publish the copyright (e.g.
http://creativecommons.org/about/licenses/ ) of BFO.
On 29 November 2010 19:56, Alan Ruttenberg <alanrut...@gmail.com> wrote:
>>[I think it would not be useful to regard
>> her [the AI] as something that has a material basis.
>
> It isn't a matter of useful. It's a matter of true.
I'm sorry I was negligent.
1. Clarification: I think it would not be useful to regard an eventual
AI or the uploaded me as a dependent continuant with an independent
continuant material basis which is a scattered set of material
Objects. I think it would be useful to regard an AI and myself to be
independent continuants, invariant to eventual changes in our material
bases.
2. Truth: I think what I say (modified by the Clarification above) is
(ontologically) true = invariant to (ontological) transformations (cf.
the end of my previous letter).
3. Usefulness: I think usefulness (and elegance etc.) is not
superfluous in ontological research either. Pragmatic approach to
ontology presupposes usefulness (and at least non-falseness). I think
there are useful ontologies just like there are useful theorems:
"[...] - elegant theorem: theorem whose statement is short and somewhat unique
- interesting theorem [...]: theorem that cannot readily be deduced
from earlier ones, but is well connected
- boring theorem: theorem for which there are many others very much like it
- useful theorem: theorem that leads to many new ones
- powerful theorem: theorem that substantially reduces the lengths of
proofs needed for many others
- surprising theorem: theorem that appears in an otherwise sparse part
of the network
- deep theorem: theorem that connects components of the network that
otherwise far away
- important theorem: theorem that allows a broad new area of the
network to be reached"
https://www.wolframscience.com/nksonline/page-1176
P.
Alan Ruttenberg
> Dear list-members, Alan, Barry and Janna,
>
>
> based on the conversation I have set Entity back as a root category in
> POnt https://github.com/mazsa/Personal-Ontology/raw/master/pont.owl :
> thank you for your help.
>
> On 29 November 2010 19:56, Alan Ruttenberg <alanrut...@gmail.com> wrote:
>> 2010/11/29 Mázsa Péter <pe...@mazsa.com>:
>>> Suppose a (likewise hypothetical) Artificial Intelligence [AI]
>>> running on the internet is a person. Do you think that she has a
>>> material basis / is an independently continuant material object
>>> (bearing a role of person)?
>>
>> The BFO approach would be to deal with that eventuality when and if it obtains.
>
> On 29 November 2010 20:01, Barry Smith <phis...@buffalo.edu> wrote:
>>> 1. You are a person. Are you sure you think that a material basis is
>>> absolutely necessary for you? If you had the opportunity do you think
>>> you would "upload" yourself (= convert yourself into information
>>> http://en.wikipedia.org/wiki/Mind_uploading ) before you die? [...]
>>
>> when this is shown to be possible, we will re-address the issue;
>
> 1. Thank you, this is a pragmatic answer, I can live with it.
>
> 2. Alan, in a way it have obtained: I raised this issue as a routine
> http://plato.stanford.edu/entries/thought-experiment :)
I don't consider thought experiments to be evidence. They are good
argument. In this case there are a number of substantive issues that
we are far from understanding, so I don't think the thought experiment
is enough.
;-)
>
> On 29 November 2010 19:56, Alan Ruttenberg <alanrut...@gmail.com> wrote:
>> 2010/11/29 Mázsa Péter <pe...@mazsa.com>:
>>>>> What is absolutely necessary for POnt is that there should be a place
>>>>> for Independent Continuants who are not necessarily Material Entities
>>>>> (or Boundaries or Sites).
>>>>
>>>> Why do you need them to be independent? In what way would dependent
>>>> continuant not be correct? Are there some things that do not have a
>>>> material basis?
>>>
>>> I think you think this AI is an independent continuant, but I can't imagine how you think she is
>>> a material object.
>>
>> I didn't say the AI was a material entity. I said it was a generically
>> dependent continuant.
>>
>>> If you had the opportunity do you think
>>> you would "upload" yourself (= convert yourself into information
>>> http://en.wikipedia.org/wiki/Mind_uploading ) before you die? [...]
>>
>> If possible, I would. Whether possible is an open question.
>> After I uploaded I would have a different material basis (some digital
>> form of memory)
>> In BFO we call such entities generically dependent continuants.
>
> You say that AIs running on the internet would be generically
> dependent continuants (not MaterialEntities) and all generically
> dependent continuants have material bases, which are presumably
> MaterialEntities, specifically, spatially dispersed sets of Objects.
Yes.
> You say that I'm an independent continuant while I'm flesh and blood,
> "uploading" would change me (still me) into a dependent continuant
> with a material basis (presumably a set of Objects), and "downloading"
> again would presumably change me back into an independent continuant.
I don't believe I said that.
Rather: I don't really know what constitutes the identity of a person.
If it came to be known that this identity could be preserved when
uploading, then we would know that it was a kind of generically
dependent continuant. When in your body, the body would be the
material entity that bears the concretization of that GDC.
BFO doesn't have such an entity (your sense of person/AI) so nothing
would need to be changed in BFO should we learn this is possible.
However if other ontologies had made a commitment that contradicted
this they would need to be changed.
>
> Similarly, if dependent continuant AIs running on the internet would
> "download" themselves exclusively into robots partitioned in space,
> they would presumably become independent continuants.
See above.
> OK, I understand. Whether possible is an open question, but eventually
> and occasionally I / AI might switch my / her ontological status in
> BFO, without ceasing to be myself / herself, so this me / her is
> jumping across ontological categories.
Except you have misunderstood me, as pointed out above.
>>> I think filing them into the existing categories of
>>> BFO would be like working with celestial spheres: still usable but
>>> ugly.]
>>
>> Sorry to have offended.
>
> :)))
> perhaps I was not PC: celestial spheres are perfect companies for us
> to work with them, e.g. to predict planetary movements and
> configurations; they are just ... under-attractive:) to my taste
Understood.
I will add comments to your responses to Barry's note, but Barry may
have other things to say.
>
> On 29 November 2010 20:01, Barry Smith <phis...@buffalo.edu> wrote:
>>> I think filing them into the existing categories of
>>> BFO would be like working with celestial spheres: still usable but
>>> ugly.]
>>>
>> there are so many problems facing use of BFO in support of established
>> science, we cannot waste time on merely hypothetical issues; for this you
>> may wish to use DOLCE
>>
>>> I think that what we think of as persons are "things" that do not
>>> necessarily and intuitively have a material basis. Are you convinced
>>> by the thought experiments above?
>>>
>> This is not the issue; the issue is how to ensure BFO is successful in
>> performing the jobs it needs to perform today
>> BS
>
> OK, this is the (acceptable) pragmatic line. Of course I cannot insist
> on the modification of the preferences of the BFO-community based on
> my unproven thoughts.
>
> But do I have the right to fork
> http://en.wikipedia.org/wiki/Fork_(software_development) BFO at all?
You do, but it would be considered bad practice to change any of the
intended meanings of BFO terms, or to distribute it in any way that it
might be confused by someone to be the real BFO.
Better to create another ontology and import and use BFO terms where
appropriate.
> I feel so that the default copyright for ontologies is
> http://creativecommons.org/licenses/by/3.0/ but I'd appreciate if you
> would publish the copyright (e.g.
> http://creativecommons.org/about/licenses/ ) of BFO.
It is so said in the OWL file
<dc:rights>http://creativecommons.org/licenses/by/3.0</dc:rights>
>
> On 29 November 2010 19:56, Alan Ruttenberg <alanrut...@gmail.com> wrote:
>>>[I think it would not be useful to regard
>>> her [the AI] as something that has a material basis.
>>
>> It isn't a matter of useful. It's a matter of true.
>
> I'm sorry I was negligent.
>
> 1. Clarification: I think it would not be useful to regard an eventual
> AI or the uploaded me as a dependent continuant with an independent
> continuant material basis which is a scattered set of material
> Objects. I think it would be useful to regard an AI and myself to be
> independent continuants, invariant to eventual changes in our material
> bases.
That's fine. Buy you are not using the terms as BFO intends them.
Better to define new additional terms that define things the way think
they should be. FWIW, I think it is possible that you don't understand
what it means to be a material basis in BFO.
> 2. Truth: I think what I say (modified by the Clarification above) is
> (ontologically) true = invariant to (ontological) transformations (cf.
> the end of my previous letter).
Yes, but the truth I was referring to was deeper. There isn't
anything, according to BFO, that doesn't depend on something material.
> 3. Usefulness: I think usefulness (and elegance etc.) is not
> superfluous in ontological research either. Pragmatic approach to
> ontology presupposes usefulness (and at least non-falseness). I think
> there are useful ontologies just like there are useful theorems:
> "[...] - elegant theorem: theorem whose statement is short and somewhat unique
> - interesting theorem [...]: theorem that cannot readily be deduced
> from earlier ones, but is well connected
> - boring theorem: theorem for which there are many others very much like it
> - useful theorem: theorem that leads to many new ones
> - powerful theorem: theorem that substantially reduces the lengths of
> proofs needed for many others
> - surprising theorem: theorem that appears in an otherwise sparse part
> of the network
> - deep theorem: theorem that connects components of the network that
> otherwise far away
> - important theorem: theorem that allows a broad new area of the
> network to be reached"
> https://www.wolframscience.com/nksonline/page-1176
It depends on what your definition of ontology is. BFO is trying to be
an ontology in the sense of being a catalog of types of things that
exist. It is proposed that building ontologies in this sense is
already damned useful for it's targeted purposes.
-Alan
Mázsa Péter
>> 2. Alan, in a way it have obtained: I raised this issue as a routine
>> http://plato.stanford.edu/entries/thought-experiment :)
>
> I don't consider thought experiments to be evidence. They are good
> argument. In this case there are a number of substantive issues that
> we are far from understanding, so I don't think the thought experiment
> is enough.
> ;-)
ok
>> You say that AIs running on the internet would be generically
>> dependent continuants (not MaterialEntities) and all generically
>> dependent continuants have material bases, which are presumably
>> MaterialEntities, specifically, spatially dispersed sets of Objects.
>
> Yes.
ok
>> You say that I'm an independent continuant while I'm flesh and blood,
>> "uploading" would change me (still me) into a dependent continuant
>> with a material basis (presumably a set of Objects), and "downloading"
>> again would presumably change me back into an independent continuant.
>
> I don't believe I said that.
> Rather: I don't really know what constitutes the identity of a person.
> If it came to be known that this identity could be preserved when
> uploading, then we would know that it was a kind of generically
> dependent continuant. When in your body, the body would be the
> material entity that bears the concretization of that GDC.
> BFO doesn't have such an entity (your sense of person/AI) so nothing
> would need to be changed in BFO should we learn this is possible.
> However if other ontologies had made a commitment that contradicted
> this they would need to be changed.
makes sense and consistent in itself: you say that BFO is independent
(and you are agnostic, at least as long as possible) on a lot of
issues. The concept of agency/personship is one of them.
>> But do I have the right to fork
>> http://en.wikipedia.org/wiki/Fork_(software_development) BFO at all?
>
> You do, but it would be considered bad practice to change any of the
> intended meanings of BFO terms, or to distribute it in any way that it
> might be confused by someone to be the real BFO.
>
> Better to create another ontology and import and use BFO terms where
> appropriate.
>
>> I feel so that the default copyright for ontologies is
>> http://creativecommons.org/licenses/by/3.0/ but I'd appreciate if you
>> would publish the copyright (e.g.
>> http://creativecommons.org/about/licenses/ ) of BFO.
>
> It is so said in the OWL file
> <dc:rights>http://creativecommons.org/licenses/by/3.0</dc:rights>
Copyright: thank you, I didn't notice it.
Disambiguation: I understand your concern. I'll do it bona fide.
>> On 29 November 2010 19:56, Alan Ruttenberg <alanrut...@gmail.com> wrote:
>>>>[I think it would not be useful to regard
>>>> her [the AI] as something that has a material basis.
>>>
>>> It isn't a matter of useful. It's a matter of true.
>>
>> I'm sorry I was negligent.
>>
>> 1. Clarification: I think it would not be useful to regard an eventual
>> AI or the uploaded me as a dependent continuant with an independent
>> continuant material basis which is a scattered set of material
>> Objects. I think it would be useful to regard an AI and myself to be
>> independent continuants, invariant to eventual changes in our material
>> bases.
>
> That's fine. Buy you are not using the terms as BFO intends them.
> Better to define new additional terms that define things the way think
> they should be. FWIW, I think it is possible that you don't understand
> what it means to be a material basis in BFO.
ad1. see below
>> 2. Truth: I think what I say (modified by the Clarification above) is
>> (ontologically) true = invariant to (ontological) transformations (cf.
>> the end of my previous letter).
>
> Yes, but the truth I was referring to was deeper. There isn't
> anything, according to BFO, that doesn't depend on something material.
ad2. see below
>> 3. Usefulness: I think usefulness (and elegance etc.) is not
>> superfluous in ontological research either. Pragmatic approach to
>> ontology presupposes usefulness (and at least non-falseness). I think
>> there are useful ontologies just like there are useful theorems:
>> "[...]
>> - useful theorem: theorem that leads to many new ones [...]"
>
> It depends on what your definition of ontology is. BFO is trying to be
> an ontology in the sense of being a catalog of types of things that
> exist. It is proposed that building ontologies in this sense is
> already damned useful for it's targeted purposes.
I think I understand what material basis means in BFO (but that
thought of mine may be a part of the problem:) Let's check it and see.
You say that "exists = (matter itself and/or) have material basis".
'Exists = (matter itself and/or) have material basis' := 'X=Mb'
Case I.
If you think X=Mb is an axiom of BFO 1.1, it's ok. Then you can say that
- either you can slide "information" into the pigeon holes of the BFO
1.1 (e.g. an AI on the internet or the uploaded me or an imaginary
magenta unicorn pegasus are generally dependent continuants with
material basis),
- or "information" (which does not have material basis) is nonexistent
(it would contradict to the axiom).
This is the scenario
- where you say that "BFO is trying to be an ontology in the sense of
being a catalog of types of things that exist", and you reject that
meaning of "information" which may or may not have material basis
(obviously along with other similar "things", e.g.
http://en.wikipedia.org/wiki/Luminiferous_aether ) presumably as
"speculative metaphysics" (à la Kant, Heidegger), and
- where every continuant with a speculative mind (like me:), be it
independent or dependent, using the fact that in BFO 1.1 owl:Thing =/=
bfo:Entity, have the legitimate opportunity to reveal other categories
under owl:Thing but outside the category bfo:Entity, e.g. a category
Nonentity, or a category dolce-lite:Quale.
- This approach is too "postmodern" to be an ontology in one (the
"continental") meaning ("[Philosophy] the branch of metaphysics that
deals with the nature of being" [= owl:Thing, including a rigorous and
exhaustive organization of it that is hierarchical and contains all
the relevant entities and their relations]),
- but it's ok in the ("analytic") other ("[Logic] the set of entities
presupposed by a theory" / "[computer science] a rigorous and
exhaustive organization of some knowledge domain that is usually
hierarchical and contains all the relevant entities and their
relations") http://www.thefreedictionary.com/ontology
Case II.
But if
- you think that X=Mb is a deep (basic) truth (of ontology in
philosophical meaning) mapped into an axiom of BFO, i.e. BFO is a
basic ontology in the philosophical meaning above, and
- tertium non datur,
then I think X=Mb is simply false.
(I don't know whether BFO has this - type II - ambition at all:
- on one hand, it is so close to it that it would be a sin:) to miss
the opportunity,
- on the other, in BFO 1.1, owl:Thing =/= bfo:Entity as if BFO would
not be confident enough to be type II and would ensure place for the
diversity of ontologies.)
Why false? Suppose for a while that I completely accept the solution
of BFO 1.1 for AI-s running on the internet (i.e, they are generically
dependent continuants, not MaterialEntities, and all generically
dependent continuants have material bases, which are MaterialEntities,
specifically, spatially dispersed sets of Objects).
However, this would not solve
- my original problem (that universe in itself could possibly
essentially be made of information with matter as special case of it,
cf. http://www.idsia.ch/~juergen/computeruniverse.html
http://blog.wolfram.com/2007/09/11/my-hobby-hunting-for-our-universe/
) and
- similar problems (e.g., do you think that the not-yet-observed
not-even-verified superstrings are axiomatically non-existent as if
they were not real? or rather they have material bases cf.
https://www.newgrounds.com/portal/view/525347 ?? or matter is defined
by Mb:=X???),
and so would not mean that X=Mb.
If you say that matter is the nesessary building block for owl:Thing,
you make the mistake in logic I made at first ("I may like information
to be the “infrastucture” of matter but you are right: for an ontology
this would be a superfluous presumption and a logical fallacy of this
kind: http://www.philosophyexperiments.com/mary/Default.aspx ").
So in Case II we do the right thing if we abandon our demand for the
preeminence of information and of matter and in general of any
specific owl:Thing.
What the universe (more exactly: owl:Thing) is made of at the
"fundamental level" is an open empirical (rather than a speculative /
a thought experiment kind of) question yet (cf. e.g.
http://arstechnica.com/science/news/2010/12/lhc-spots-no-black-holes-eliminates-some-versions-of-string-theory.ars
): I don't think so that BFO in Case II should decide on it in
advance. BFO in Case II should be explicitely independent on all
"speculative mathaphysical" answers.
Solution
In Case I: There is no problem to solve.
Your axioms/categories are yours, mine are mine: there is no further
need to argue (you don't argue fiercely e.g. on Dewey numbers:) and no
need to change anything on BFO 1.1. The only (absolutely legitimate!)
task left is to keep BFO away form revolution and let it evolve into
other disciplines.
In Case II: Back to the future!
Fortunately, even if I am right in what I am saying above, BFO
itself(!) has a solution for the problem of Case II: your internal
solution in Case II is BFO 1.0 ( http://www.ifomis.org/bfo/1.0 ) The
structure of the previous version of BFO is "agnostic": independent on
what owl:Thing is supposed to made of.
Of course, BFO 1.1 is an improvement on 1.0: Disposotion, Function and
Role are bundled into a category in 1.0, and bundling Object,
FiatObjectPart and ObjectAggregate into a new category in 1.1 is a
parallel idea. BFO may retain the new structure in the next version of
BFO as well.
The problem is, as the forensic investigations show:), that this
stuctural change was accompanied by the silent introduction of X=Mb by
calling the new category 'MaterialEntity'. This is why BFO 1.1 is not
just a harmless improvement of 1.0. From the point of view of Case II,
1.1 is 1.0 constrained by the false X=Mb.
I think (the Platonic idea of) BFO is deeper than 1.1, and 1.0 is closer to it.
Suggestions
I'm a just a user of BFO, not even a heavy one, one among many
thousands. You should put my suggestions in their right place.
For me it is very important to clarify: how would look like the BFO I
would be happy to work with, without the need to fork it?
1. It would be modular: a core and auxiliary axioms (or alternative
auxiliary axiom systems). BFO is independent on a lot of issues. The
concept of agency/personship is one of them. For me, agency/personship
is ontologically important, but you are right: this doesn't imply that
the core of BFO should contain personship. However, it implies that
BFO should not preclude the ontological concept of personship.
BFO should be modular: its core
- should remain indenpendent on theories in ontology (like the
ontological construction of the concept of agency and personship) and
- should become independent on open empirical questions (like the
fundamental constitution of owl:Things).
2. It would have a Type II core.
The core of BFO seems to me the right place for the
philosophical/continental meaning of ontology (Case II above). This is
why
2.1. I would consider removing the the root category bfo:Entity from
BFO. In philosophical meaning of ontology bfo:Entity is not different
to owl:Thing, it would be an inappropriate postmodern modesty:) not to
erease it.
2.2. I would change the label of the BFO 1.1 category 'MaterialEntity'
into another label without reference to matter and to any other
specific hypothetical constitution of owl:Thing.
(2.3. I would revisit the examples and the wording of the definitions
of BFO 1.1 based on the arguments above.)
3. It would have different Type I auxiliary axioms.
3.A. With the BFO core + X=Mb as an auxiliary axiom, we could
essentially reconstruct BFO 1.1.
3.B. With the BFO core + an auxiliary axiom alternative to X=Mb, I
would be able to describe what I think to be ontologically important.
Results
In this https://groups.google.com/group/bfo-discuss/browse_thread/thread/433bfa9718cae15
conversation BFO 1.0 has revealed itself as more profound than BFO
1.1. Minor changes in BFO 1.0 could result in a future BFO core
independent on material or any other specific hypothetical
constitution of owl:Thing. BFO core + auxiliary axiom X=Mb would
reconstruct BFO 1.1. BFO core + alternative auxiliary axioms would
ensure conceptual space for alternative ontologies without the need to
change or fork the BFO core, and without any interference with the
reconstructed BFO 1.1.