This is a proposed update that changes inheres in to the characteristic of model

[leechuck]

No description provided.

[zhengj2007]

This change intends to apply to quality only or apply to 'specifically dependent continuant' including 'realizable entity' like disposition and role.
Current "realized in" relation is used to associate 'realizable entity' with process. If we just relabel the RO: "inheres in" (RO:0000052) and changed the range to "independent continuant" or "process", does it mean RO: "realized in" (BFO_0000054) is a subProperty of RO: "characteristic of"? If not, what's difference between a 'realizable entity' 'realized in' a process and a 'realizable entity' 'characteristic of' a process?

[cmungall]

This PR needs to be done against ro-edit rather than ro... I promised to do this but have not yet

[alanruttenberg]

It's a bit hard to tell, but it looks like the changes have more than just this change. Could someone check?

[cmungall]

There is an additional complication beyond just migrating the edits to the edit file.

TL;DR - characteristic-of can't be a super-property of a property representing what the current BFO2 reference calls inheres-in

For the phenotype work, the assumption was always that the relation linking the quality to the IC/process was functional. We set this out here:
http://webont.org/owled/2007/PapersPDF/paper_40.pdf

We didn't just make this up. It was entirely our understanding at that time (pre BFO formalization in a non-PDF form) that inheres-in was intended to be functional. See equation A7 in Fabian, Pierre, and Barry's paper, which we took to be canonical:
http://ontology.buffalo.edu/bfo/SQU.pdf
(the mapping to OWL maybe isn't quite so clear due to the use of n-ary relations for relational qualities)

(note that paper explicitly deferred on QoPs)

So if the URI we are using becomes the super-property of what is now what the BFO2 reference calls inheres-in, we cannot declare this to be functional with inducing the same characteristic on BFO2:inheres_in, which is not intentional according to the reference, due to the following definition that was introduced for relational qualities:
Definition: b is a relational quality = Def. for some independent continuants c, d and for some time t: b quality_of c at t & b quality_of d at t. [057-001]

If we want to declare RO_0000052 as functional, then this has to become a sibling of BFO2:inheres_in. This goes further than what we all agreed upon.

An alternate option is that BFO2 abandons its approach to RQs and adopts the one we use in PATO where inheres-in is used as a functional relation, and the additional relatum for a RQ gets a different object property (currently awkwardly called 'towards'). This is more in the spirit of the Neuhaus paper I think. (FWIW, I think the PATO community may be the only one using RQs, and they cause widespread confusion).

[bpeters42]

Discussed this on the call 6/4 which Chris M could unfortunately not join. Consensus that some solution needs to appear fast, and that Chris M needs to weigh in if what Alan is proposing above works for him.

[cmungall]

Discussed on RO call today, all in agreement, see https://docs.google.com/document/d/19dhzj5QoliQZQgDJRczI878rLtexzMDo1g0K9_LRUWk/edit#

We will implement this structure (names not yet decided):

• [new] characteristic of or inheres in [domain: SDC, range: Thing]
  • RO_0000052 functional characteristic of [properties: functional]
  • inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: IC]

We will decide on labels shortly.

[phismith]

can you tell ,me where process qualities fit in here? is the idea that they would be SDCs that would depend on processes?

…

On Tue, Jul 30, 2019 at 12:38 PM Chris Mungall ***@***.***> wrote: Discussed on RO call today. We will implement this structure (names not yet decided): - [new] characteristic of or inheres in [domain: SDC, range: Thing] - RO_0000052 functional characteristic of [properties: functional] - inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: IC] We will decide on labels shortly. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#284?email_source=notifications&email_token=AB7KUNY3BOHIGUDA5BK6NPLQCBVB3A5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3ESFFY#issuecomment-516498071>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AB7KUN3XV2MV4TGR7WSLJQ3QCBVB3ANCNFSM4F7LSHUQ> .

[alanruttenberg]

On Tue, Jun 4, 2019 at 12:14 AM Chris Mungall ***@***.***> wrote: For the phenotype work, the assumption was always that the relation linking the quality to the IC/process was functional. We set this out here: http://webont.org/owled/2007/PapersPDF/paper_40.pdf We didn't just make this up. It was entirely our understanding at that time (pre BFO formalization in a non-PDF form) that inheres-in was intended to be functional. See equation A7 in Fabian, Pierre, and Barry's paper, which we took to be canonical: http://ontology.buffalo.edu/bfo/SQU.pdf

I'm not reading that as functional in the sense you mean. From SQU: "If v is a d-place quality, then v inheres in d entities or there is an entity which inheres in d entities ..." i.e. you can inhere in more than one entity. They talk about d-place qualities. I'm still trying to grok the formal notation. Alan

[alanruttenberg]

On Tue, Jul 30, 2019 at 10:34 PM Alan Ruttenberg <alanruttenberg@gmail.com> wrote:

On Tue, Jun 4, 2019 at 12:14 AM Chris Mungall ***@***.***> wrote: > For the phenotype work, the assumption was always that the relation > linking the quality to the IC/process was functional. We set this out here: > > http://webont.org/owled/2007/PapersPDF/paper_40.pdf > > We didn't just make this up. It was entirely our understanding at that > time (pre BFO formalization in a non-PDF form) that inheres-in was intended > to be functional. See equation A7 in Fabian, Pierre, and Barry's paper, > which we took to be canonical: > http://ontology.buffalo.edu/bfo/SQU.pdf > I'm not reading that as functional in the sense you mean. From SQU: "If v is a d-place quality, then v inheres in d entities or there is an entity which inheres in d entities ..."

My understanding of what you mean by inheres in is function: A quality can inhere in only one entity. My understanding of A7: A quality can inhere in only one tuple of entities.

…

i.e. you can inhere in more than one entity. They talk about d-place qualities. I'm still trying to grok the formal notation. Alan

[alanruttenberg]

On Tue, Jul 30, 2019 at 12:38 PM Chris Mungall ***@***.***> wrote: Discussed on RO call today. We will implement this structure (names not yet decided): - [new] characteristic of or inheres in [domain: SDC, range: Thing] - RO_0000052 functional characteristic of [properties: functional] - inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: IC] We will decide on labels shortly.

This makes the assumption that a process characteristic is a SDC. The problem here is that SDCs only and necessarily depend on independent continuants that aren't spatial regions, or specifically dependent continuants , but not processes. One might argue that functional-characteristic-of is a distinct relation from s-depends-on, but that would still leave a SDC with a s-depends-on relation to who-knows-what. I think the change needs to be coupled with a revision of PATO:Quality. a class process-characteristic is added. PATO:Quality becomes quality or process-characteristic, Then the question is where process-characteristic lies in the BFO hierarchy. We haven't worked all the way through this, but currently I think some might be continuants and some might be occurrents based on whether they can be temporally sliced or not (duration:yes, rate:no). A more careful review of the existing PATO process qualities might shed light. Maybe some of us can get together to do that. A safe bet, though I'm not sure how happy Barry would be about it, would be to do the following: Add subclass of occurrent : occurrent-process-characteristic Add a subclass of continuant: continuant-process-characteristic Add a defined class process-characteristic = occurrent-process-characteristic OR continuant-process-characteristic Add a defined class characteristic = process-characteristic or SDC Make domain of 'characteristic of or inheres in' characteristic Make domain of 'functional characteristic of' characteristic As usual, the labels are placeholders that I'm not vested in. Alan

…

— You are receiving this because you commented. Reply to this email directly, view it on GitHub <#284?email_source=notifications&email_token=AAB3CDULUMSJMLCM4IUNLSTQCBVB3A5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3ESFFY#issuecomment-516498071>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAB3CDQI2QMOW2FZJIUJBZTQCBVB3ANCNFSM4F7LSHUQ> .

[bpeters42]

My simpler proposal, in order to just get something logically consistent out now, is to drop the SDC constraint on the domain of 'characteristic of'. A refactoring of PATO has been promised for at least five years, so we should not make that the 'first step'. On Wed, Jul 31, 2019 at 8:45 AM Alan Ruttenberg <notifications@github.com> wrote:

…

On Tue, Jul 30, 2019 at 12:38 PM Chris Mungall ***@***.***> wrote: > Discussed on RO call today. > > We will implement this structure (names not yet decided): > > - [new] characteristic of or inheres in [domain: SDC, range: Thing] > - RO_0000052 functional characteristic of [properties: functional] > - inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: > IC] > > We will decide on labels shortly. > This makes the assumption that a process characteristic is a SDC. The problem here is that SDCs only and necessarily depend on independent continuants that aren't spatial regions, or specifically dependent continuants , but not processes. One might argue that functional-characteristic-of is a distinct relation from s-depends-on, but that would still leave a SDC with a s-depends-on relation to who-knows-what. I think the change needs to be coupled with a revision of PATO:Quality. a class process-characteristic is added. PATO:Quality becomes quality or process-characteristic, Then the question is where process-characteristic lies in the BFO hierarchy. We haven't worked all the way through this, but currently I think some might be continuants and some might be occurrents based on whether they can be temporally sliced or not (duration:yes, rate:no). A more careful review of the existing PATO process qualities might shed light. Maybe some of us can get together to do that. A safe bet, though I'm not sure how happy Barry would be about it, would be to do the following: Add subclass of occurrent : occurrent-process-characteristic Add a subclass of continuant: continuant-process-characteristic Add a defined class process-characteristic = occurrent-process-characteristic OR continuant-process-characteristic Add a defined class characteristic = process-characteristic or SDC Make domain of 'characteristic of or inheres in' characteristic Make domain of 'functional characteristic of' characteristic As usual, the labels are placeholders that I'm not vested in. Alan > — > You are receiving this because you commented. > Reply to this email directly, view it on GitHub > < #284?email_source=notifications&email_token=AAB3CDULUMSJMLCM4IUNLSTQCBVB3A5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3ESFFY#issuecomment-516498071 >, > or mute the thread > < https://github.com/notifications/unsubscribe-auth/AAB3CDQI2QMOW2FZJIUJBZTQCBVB3ANCNFSM4F7LSHUQ > > . > — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#284?email_source=notifications&email_token=ADJX2ISPDHZGSLJ6536227TQCGCOZA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HD4TI#issuecomment-516832845>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADJX2IVSRRUZICEC6IK2CKDQCGCOZANCNFSM4F7LSHUQ> .

-- Bjoern Peters Professor La Jolla Institute for Allergy and Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters

[alanruttenberg]

On Wed, Jul 31, 2019 at 9:49 AM bpeters42 ***@***.***> wrote: My simpler proposal, in order to just get something logically consistent out now, is to drop the SDC constraint on the domain of 'characteristic of'.

Doing that will get rid of 1/2 the problems. There's still the question of *what is* the domain, and leaving it unconstrained will mean that there will be terms without a path to BFO root as well as opportunity for nonsense assertions, such as asserting a material entity as characteristic of something.

A refactoring of PATO has been promised for at least five years, so we should not make that the 'first step'.

Didn't understand what you mean here. Alan

…

On Wed, Jul 31, 2019 at 8:45 AM Alan Ruttenberg ***@***.***> wrote: > On Tue, Jul 30, 2019 at 12:38 PM Chris Mungall ***@***.*** > > wrote: > > > Discussed on RO call today. > > > > We will implement this structure (names not yet decided): > > > > - [new] characteristic of or inheres in [domain: SDC, range: Thing] > > - RO_0000052 functional characteristic of [properties: functional] > > - inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: > > IC] > > > > We will decide on labels shortly. > > > This makes the assumption that a process characteristic is a SDC. The > problem here is that SDCs only and necessarily depend on independent > continuants that aren't spatial regions, or specifically dependent > continuants , but not processes. One might argue that > functional-characteristic-of is a distinct relation from s-depends-on, but > that would still leave a SDC with a s-depends-on relation to > who-knows-what. > > I think the change needs to be coupled with a revision of PATO:Quality. > > a class process-characteristic is added. > PATO:Quality becomes quality or process-characteristic, > > Then the question is where process-characteristic lies in the BFO > hierarchy. We haven't worked all the way through this, but currently I > think some might be continuants and some might be occurrents based on > whether they can be temporally sliced or not (duration:yes, rate:no). A > more careful review of the existing PATO process qualities might shed > light. Maybe some of us can get together to do that. A safe bet, though > I'm not sure how happy Barry would be about it, would be to do the > following: > > Add subclass of occurrent : occurrent-process-characteristic > > Add a subclass of continuant: continuant-process-characteristic > > Add a defined class process-characteristic = > occurrent-process-characteristic OR continuant-process-characteristic > > Add a defined class characteristic = process-characteristic or SDC > > Make domain of 'characteristic of or inheres in' characteristic > > Make domain of 'functional characteristic of' characteristic > > As usual, the labels are placeholders that I'm not vested in. > > Alan > > > > > — > > You are receiving this because you commented. > > Reply to this email directly, view it on GitHub > > < > #284?email_source=notifications&email_token=AAB3CDULUMSJMLCM4IUNLSTQCBVB3A5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3ESFFY#issuecomment-516498071 > >, > > or mute the thread > > < > https://github.com/notifications/unsubscribe-auth/AAB3CDQI2QMOW2FZJIUJBZTQCBVB3ANCNFSM4F7LSHUQ > > > > . > > > > — > You are receiving this because you commented. > Reply to this email directly, view it on GitHub > < #284?email_source=notifications&email_token=ADJX2ISPDHZGSLJ6536227TQCGCOZA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HD4TI#issuecomment-516832845 >, > or mute the thread > < https://github.com/notifications/unsubscribe-auth/ADJX2IVSRRUZICEC6IK2CKDQCGCOZANCNFSM4F7LSHUQ > > . > -- Bjoern Peters Professor La Jolla Institute for Allergy and Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#284?email_source=notifications&email_token=AAB3CDXZPSGG7L6UHFKNLN3QCGJ6FA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HJ2LY#issuecomment-516857135>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAB3CDRMCFPRUWGKVW2CRD3QCGJ6FANCNFSM4F7LSHUQ> .

[bpeters42]

Well, wouldn't it be nice to get rid of 1/2 the problems? On Wed, Jul 31, 2019, 9:55 AM Alan Ruttenberg <notifications@github.com> wrote:

…

On Wed, Jul 31, 2019 at 9:49 AM bpeters42 ***@***.***> wrote: > My simpler proposal, in order to just get something logically consistent > out now, is to drop the SDC constraint on the domain of 'characteristic > of'. Doing that will get rid of 1/2 the problems. There's still the question of *what is* the domain, and leaving it unconstrained will mean that there will be terms without a path to BFO root as well as opportunity for nonsense assertions, such as asserting a material entity as characteristic of something. > A refactoring of PATO has been promised for at least five years, so we > should not make that the 'first step'. > Didn't understand what you mean here. Alan > > > On Wed, Jul 31, 2019 at 8:45 AM Alan Ruttenberg < ***@***.***> > wrote: > > > On Tue, Jul 30, 2019 at 12:38 PM Chris Mungall < ***@***.*** > > > > wrote: > > > > > Discussed on RO call today. > > > > > > We will implement this structure (names not yet decided): > > > > > > - [new] characteristic of or inheres in [domain: SDC, range: Thing] > > > - RO_0000052 functional characteristic of [properties: functional] > > > - inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: > > > IC] > > > > > > We will decide on labels shortly. > > > > > This makes the assumption that a process characteristic is a SDC. The > > problem here is that SDCs only and necessarily depend on independent > > continuants that aren't spatial regions, or specifically dependent > > continuants , but not processes. One might argue that > > functional-characteristic-of is a distinct relation from s-depends-on, > but > > that would still leave a SDC with a s-depends-on relation to > > who-knows-what. > > > > I think the change needs to be coupled with a revision of PATO:Quality. > > > > a class process-characteristic is added. > > PATO:Quality becomes quality or process-characteristic, > > > > Then the question is where process-characteristic lies in the BFO > > hierarchy. We haven't worked all the way through this, but currently I > > think some might be continuants and some might be occurrents based on > > whether they can be temporally sliced or not (duration:yes, rate:no). A > > more careful review of the existing PATO process qualities might shed > > light. Maybe some of us can get together to do that. A safe bet, though > > I'm not sure how happy Barry would be about it, would be to do the > > following: > > > > Add subclass of occurrent : occurrent-process-characteristic > > > > Add a subclass of continuant: continuant-process-characteristic > > > > Add a defined class process-characteristic = > > occurrent-process-characteristic OR continuant-process-characteristic > > > > Add a defined class characteristic = process-characteristic or SDC > > > > Make domain of 'characteristic of or inheres in' characteristic > > > > Make domain of 'functional characteristic of' characteristic > > > > As usual, the labels are placeholders that I'm not vested in. > > > > Alan > > > > > > > > > — > > > You are receiving this because you commented. > > > Reply to this email directly, view it on GitHub > > > < > > > #284?email_source=notifications&email_token=AAB3CDULUMSJMLCM4IUNLSTQCBVB3A5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3ESFFY#issuecomment-516498071 > > >, > > > or mute the thread > > > < > > > https://github.com/notifications/unsubscribe-auth/AAB3CDQI2QMOW2FZJIUJBZTQCBVB3ANCNFSM4F7LSHUQ > > > > > > . > > > > > > > — > > You are receiving this because you commented. > > Reply to this email directly, view it on GitHub > > < > #284?email_source=notifications&email_token=ADJX2ISPDHZGSLJ6536227TQCGCOZA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HD4TI#issuecomment-516832845 > >, > > or mute the thread > > < > https://github.com/notifications/unsubscribe-auth/ADJX2IVSRRUZICEC6IK2CKDQCGCOZANCNFSM4F7LSHUQ > > > > . > > > > > -- > Bjoern Peters > Professor > La Jolla Institute for Allergy and Immunology > 9420 Athena Circle > La Jolla, CA 92037, USA > Tel: 858/752-6914 > Fax: 858/752-6987 > http://www.liai.org/pages/faculty-peters > > — > You are receiving this because you commented. > Reply to this email directly, view it on GitHub > < #284?email_source=notifications&email_token=AAB3CDXZPSGG7L6UHFKNLN3QCGJ6FA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HJ2LY#issuecomment-516857135 >, > or mute the thread > < https://github.com/notifications/unsubscribe-auth/AAB3CDRMCFPRUWGKVW2CRD3QCGJ6FANCNFSM4F7LSHUQ > > . > — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#284?email_source=notifications&email_token=ADJX2IRAKE5KC6RIILK77UTQCGKVFA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HKNIY#issuecomment-516859555>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADJX2IUJH7LGMOEACQOUTWLQCGKVFANCNFSM4F7LSHUQ> .

[alanruttenberg]

On Wed, Jul 31, 2019 at 10:56 AM bpeters42 ***@***.***> wrote: Well, wouldn't it be nice to get rid of 1/2 the problems?

It would. I'm just not sure if the new problems offset the benefit.

…

On Wed, Jul 31, 2019, 9:55 AM Alan Ruttenberg ***@***.***> wrote: > On Wed, Jul 31, 2019 at 9:49 AM bpeters42 ***@***.***> > wrote: > > > My simpler proposal, in order to just get something logically consistent > > out now, is to drop the SDC constraint on the domain of 'characteristic > > of'. > > > Doing that will get rid of 1/2 the problems. There's still the question of > *what is* the domain, and leaving it unconstrained will mean that there > will be terms without a path to BFO root as well as opportunity for > nonsense assertions, such as asserting a material entity as characteristic > of something. > > > > A refactoring of PATO has been promised for at least five years, so we > > should not make that the 'first step'. > > > > Didn't understand what you mean here. > > Alan > > > > > > > > On Wed, Jul 31, 2019 at 8:45 AM Alan Ruttenberg < > ***@***.***> > > wrote: > > > > > On Tue, Jul 30, 2019 at 12:38 PM Chris Mungall < > ***@***.*** > > > > > > wrote: > > > > > > > Discussed on RO call today. > > > > > > > > We will implement this structure (names not yet decided): > > > > > > > > - [new] characteristic of or inheres in [domain: SDC, range: Thing] > > > > - RO_0000052 functional characteristic of [properties: functional] > > > > - inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: > > > > IC] > > > > > > > > We will decide on labels shortly. > > > > > > > This makes the assumption that a process characteristic is a SDC. The > > > problem here is that SDCs only and necessarily depend on independent > > > continuants that aren't spatial regions, or specifically dependent > > > continuants , but not processes. One might argue that > > > functional-characteristic-of is a distinct relation from s-depends-on, > > but > > > that would still leave a SDC with a s-depends-on relation to > > > who-knows-what. > > > > > > I think the change needs to be coupled with a revision of PATO:Quality. > > > > > > a class process-characteristic is added. > > > PATO:Quality becomes quality or process-characteristic, > > > > > > Then the question is where process-characteristic lies in the BFO > > > hierarchy. We haven't worked all the way through this, but currently I > > > think some might be continuants and some might be occurrents based on > > > whether they can be temporally sliced or not (duration:yes, rate:no). A > > > more careful review of the existing PATO process qualities might shed > > > light. Maybe some of us can get together to do that. A safe bet, though > > > I'm not sure how happy Barry would be about it, would be to do the > > > following: > > > > > > Add subclass of occurrent : occurrent-process-characteristic > > > > > > Add a subclass of continuant: continuant-process-characteristic > > > > > > Add a defined class process-characteristic = > > > occurrent-process-characteristic OR continuant-process-characteristic > > > > > > Add a defined class characteristic = process-characteristic or SDC > > > > > > Make domain of 'characteristic of or inheres in' characteristic > > > > > > Make domain of 'functional characteristic of' characteristic > > > > > > As usual, the labels are placeholders that I'm not vested in. > > > > > > Alan > > > > > > > > > > > > > — > > > > You are receiving this because you commented. > > > > Reply to this email directly, view it on GitHub > > > > < > > > > > > #284?email_source=notifications&email_token=AAB3CDULUMSJMLCM4IUNLSTQCBVB3A5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3ESFFY#issuecomment-516498071 > > > >, > > > > or mute the thread > > > > < > > > > > > https://github.com/notifications/unsubscribe-auth/AAB3CDQI2QMOW2FZJIUJBZTQCBVB3ANCNFSM4F7LSHUQ > > > > > > > > . > > > > > > > > > > — > > > You are receiving this because you commented. > > > Reply to this email directly, view it on GitHub > > > < > > > #284?email_source=notifications&email_token=ADJX2ISPDHZGSLJ6536227TQCGCOZA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HD4TI#issuecomment-516832845 > > >, > > > or mute the thread > > > < > > > https://github.com/notifications/unsubscribe-auth/ADJX2IVSRRUZICEC6IK2CKDQCGCOZANCNFSM4F7LSHUQ > > > > > > . > > > > > > > > > -- > > Bjoern Peters > > Professor > > La Jolla Institute for Allergy and Immunology > > 9420 Athena Circle > > La Jolla, CA 92037, USA > > Tel: 858/752-6914 > > Fax: 858/752-6987 > > http://www.liai.org/pages/faculty-peters > > > > — > > You are receiving this because you commented. > > Reply to this email directly, view it on GitHub > > < > #284?email_source=notifications&email_token=AAB3CDXZPSGG7L6UHFKNLN3QCGJ6FA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HJ2LY#issuecomment-516857135 > >, > > or mute the thread > > < > https://github.com/notifications/unsubscribe-auth/AAB3CDRMCFPRUWGKVW2CRD3QCGJ6FANCNFSM4F7LSHUQ > > > > . > > > > — > You are receiving this because you commented. > Reply to this email directly, view it on GitHub > < #284?email_source=notifications&email_token=ADJX2IRAKE5KC6RIILK77UTQCGKVFA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HKNIY#issuecomment-516859555 >, > or mute the thread > < https://github.com/notifications/unsubscribe-auth/ADJX2IUJH7LGMOEACQOUTWLQCGKVFANCNFSM4F7LSHUQ > > . > — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#284?email_source=notifications&email_token=AAB3CDRNHKXVSQMGUEPCLIDQCGR3PA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HQ33A#issuecomment-516885996>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAB3CDRVBECRHBXS5DCMI63QCGR3PANCNFSM4F7LSHUQ> .

[cmungall]

@dosumis and @matentzn advocate for something like:

• [new] ugly disjunctive name [domain: SDC, range: Thing]
  • RO_0000052 "characteristic of" [properties: functional]
  • inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: IC]

[cmungall]

Should not be merged - this PR is now serving as more of a ticket for a new PR. See other comments

[phismith]

I vote for this BS On Wed, Jul 31, 2019 at 8:45 AM Alan Ruttenberg <notifications@github.com> wrote:

…

On Tue, Jul 30, 2019 at 12:38 PM Chris Mungall ***@***.***> wrote: > Discussed on RO call today. > > We will implement this structure (names not yet decided): > > - [new] characteristic of or inheres in [domain: SDC, range: Thing] > - RO_0000052 functional characteristic of [properties: functional] > - inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: > IC] > > We will decide on labels shortly. > This makes the assumption that a process characteristic is a SDC. The problem here is that SDCs only and necessarily depend on independent continuants that aren't spatial regions, or specifically dependent continuants , but not processes. One might argue that functional-characteristic-of is a distinct relation from s-depends-on, but that would still leave a SDC with a s-depends-on relation to who-knows-what. I think the change needs to be coupled with a revision of PATO:Quality. a class process-characteristic is added. PATO:Quality becomes quality or process-characteristic, Then the question is where process-characteristic lies in the BFO hierarchy. We haven't worked all the way through this, but currently I think some might be continuants and some might be occurrents based on whether they can be temporally sliced or not (duration:yes, rate:no). A more careful review of the existing PATO process qualities might shed light. Maybe some of us can get together to do that. A safe bet, though I'm not sure how happy Barry would be about it, would be to do the following: Add subclass of occurrent : occurrent-process-characteristic Add a subclass of continuant: continuant-process-characteristic Add a defined class process-characteristic = occurrent-process-characteristic OR continuant-process-characteristic Add a defined class characteristic = process-characteristic or SDC Make domain of 'characteristic of or inheres in' characteristic Make domain of 'functional characteristic of' characteristic As usual, the labels are placeholders that I'm not vested in. Alan > — > You are receiving this because you commented. > Reply to this email directly, view it on GitHub > < #284?email_source=notifications&email_token=AAB3CDULUMSJMLCM4IUNLSTQCBVB3A5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3ESFFY#issuecomment-516498071 >, > or mute the thread > < https://github.com/notifications/unsubscribe-auth/AAB3CDQI2QMOW2FZJIUJBZTQCBVB3ANCNFSM4F7LSHUQ > > . > — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#284?email_source=notifications&email_token=AB7KUNZBRFBLI66WC6L5C5TQCGCOXA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HD4TI#issuecomment-516832845>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AB7KUN2RXQ5VVQFQPATJOIDQCGCOXANCNFSM4F7LSHUQ> .

[phismith]

More specifically, I vote for the part of what Alan wrote highlighted in yellow below BS

…

On Thu, Aug 1, 2019 at 1:02 PM Barry Smith ***@***.***> wrote: I vote for this BS On Wed, Jul 31, 2019 at 8:45 AM Alan Ruttenberg ***@***.***> wrote: > On Tue, Jul 30, 2019 at 12:38 PM Chris Mungall ***@***.***> > wrote: > > > Discussed on RO call today. > > > > We will implement this structure (names not yet decided): > > > > - [new] characteristic of or inheres in [domain: SDC, range: Thing] > > - RO_0000052 functional characteristic of [properties: functional] > > - inheres_in [equivalent to inheres in in the bfo2 ref doc] [range: > > IC] > > > > We will decide on labels shortly. > > > This makes the assumption that a process characteristic is a SDC. The > problem here is that SDCs only and necessarily depend on independent > continuants that aren't spatial regions, or specifically dependent > continuants , but not processes. One might argue that > functional-characteristic-of is a distinct relation from s-depends-on, but > that would still leave a SDC with a s-depends-on relation to > who-knows-what. > > I think the change needs to be coupled with a revision of PATO:Quality. > > a class process-characteristic is added. > PATO:Quality becomes quality or process-characteristic, > > Then the question is where process-characteristic lies in the BFO > hierarchy. We haven't worked all the way through this, but currently I > think some might be continuants and some might be occurrents based on > whether they can be temporally sliced or not (duration:yes, rate:no). A > more careful review of the existing PATO process qualities might shed > light. Maybe some of us can get together to do that. A safe bet, though > I'm not sure how happy Barry would be about it, would be to do the > following: > > Add subclass of occurrent : occurrent-process-characteristic > > Add a subclass of continuant: continuant-process-characteristic > > Add a defined class process-characteristic = > occurrent-process-characteristic OR continuant-process-characteristic > > Add a defined class characteristic = process-characteristic or SDC > > Make domain of 'characteristic of or inheres in' characteristic > > Make domain of 'functional characteristic of' characteristic > > As usual, the labels are placeholders that I'm not vested in. > > Alan > > > > > — > > You are receiving this because you commented. > > Reply to this email directly, view it on GitHub > > < > #284?email_source=notifications&email_token=AAB3CDULUMSJMLCM4IUNLSTQCBVB3A5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3ESFFY#issuecomment-516498071 > >, > > or mute the thread > > < > https://github.com/notifications/unsubscribe-auth/AAB3CDQI2QMOW2FZJIUJBZTQCBVB3ANCNFSM4F7LSHUQ > > > > . > > > > — > You are receiving this because you commented. > Reply to this email directly, view it on GitHub > <#284?email_source=notifications&email_token=AB7KUNZBRFBLI66WC6L5C5TQCGCOXA5CNFSM4F7LSHU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3HD4TI#issuecomment-516832845>, > or mute the thread > <https://github.com/notifications/unsubscribe-auth/AB7KUN2RXQ5VVQFQPATJOIDQCGCOXANCNFSM4F7LSHUQ> > . >

[cmungall]

new PR here: #442