Name is species concept thinking
I was looking through the latest DarwinCore and comments related to the TaxonConceptID.
Since the name serves as both a unique identifier and a phylogenetic hypothesis, you are effectively saying that observations labeled
*Aedes triseriatus*
and
*Ochlerotatus triseriatus*
Are separate species concepts, and should therefore be treated as separate things.
i.e. The name is the concept.
Also since there are several name variants for each "species", how do you distinguish which of these nameID's are the same species and which are different?
- Pete
---------------------------------------------------------------- Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base ------------------------------------------------------------
I don't think that's right. Names are pointers to species concepts, but are not concepts themselves. A label for a species concept would look like "Aus bus sensu Author/Citation" or "Aus bus sec. Author/Citation". Also, of course, the name is not, in itself, a unique identifier (in something like 10% of cases -- which in my mind is non-trivial).
Also, I disagree with the idea that Aedes triseriatus and Ochlerotatus triseriatus necessarily refer to different taxon concepts. To me, the "concept" is the circumscribed set of organisms. If I curcumscribe a set of organisms that I label with "Aedes triseriatus", and then later decide that this set of organisms is best classified in the genus Ocherotatus, then naming it as such does not change the circumscribed set of organisms. It certainly may have implications on the concepts for the genera "Aedes" and "Ochlerotatus", but in my mind, it has no effect on the implied circumscription (=Concept, sensu me) of what is indicated by the species epithet "triseriatus".
I do not think it makes sense to include hierarchical clasification as part of the terminal taxon "concept". Taxa at each hierarchical rank are, in my mind, defined by their contents; not their higher classification.
The way I visualize it, there is a many:many relationship between names and concepts (I *think* this applies no matter what you mean by "name", and no matter what you mean by "concept"). The same circumscription of organisms can be labelled by many different names, and the same name may apply to many different circumscriptions of organisms (not just homonyms/homographs, but also lumper/splitter issues).
Therefore, I don't think it is appropriate to try to equate names:concepts as 1:1, or even many:1.
Aloha, Rich
P.S. I certainly think that "Aedes triseriatus" and "Ochlerotatus triseriatus" are different "things", just not (necessarily) different taxon concepts. Actually, from an informatics perspective, I think that treating these different combinations as unique/identified objects doesn't gain us much. I think it's *MUCH* more robust to parse out the different individual usages of each combination as the identified objects, then derive the unique combinations/spellings/etc. from those usages. If the notion of indexing usages seems too intimidating, then start with the easy ones -- like the original useages of each of the name elements ("Aedes", "Ochlerotatus", and "triseriatus"), and the key treatments (e.g., whoever first combined "triseriatus" with the genus "Ochlerotatus", and/or whoever robustly defined alternate concepts for each).
_____
From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Peter DeVries Sent: Thursday, June 10, 2010 9:38 AM To: tdwg-content@lists.tdwg.org Subject: [tdwg-content] Name is species concept thinking
I was looking through the latest DarwinCore and comments related to the TaxonConceptID.
Since the name serves as both a unique identifier and a phylogenetic hypothesis, you are effectively saying that observations labeled
Aedes triseriatus
and
Ochlerotatus triseriatus
Are separate species concepts, and should therefore be treated as separate things.
i.e. The name is the concept.
Also since there are several name variants for each "species", how do you distinguish which of these nameID's are the same species and which are different?
- Pete
---------------------------------------------------------------- Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base ------------------------------------------------------------
Keep in mind that I recognize that my comments are only my own perspective/options. I guess a key question that I'd like to get feedback from others on is whether "Parent Taxon" and "circumscribed organisms" are *both* intrinsic properties of "taxon concept", or if "taxon concept" is effectively equivalent to "circumscription", relegating "parent taxon" as a property of Classification, separate from "Taxon Concept".
Aloha, Rich
________________________________
From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Richard Pyle Sent: Thursday, June 10, 2010 9:55 AM To: 'Peter DeVries'; tdwg-content@lists.tdwg.org Subject: Re: [tdwg-content] Name is species concept thinking I don't think that's right. Names are pointers to species concepts, but are not concepts themselves. A label for a species concept would look like "Aus bus sensu Author/Citation" or "Aus bus sec. Author/Citation". Also, of course, the name is not, in itself, a unique identifier (in something like 10% of cases -- which in my mind is non-trivial). Also, I disagree with the idea that Aedes triseriatus and Ochlerotatus triseriatus necessarily refer to different taxon concepts. To me, the "concept" is the circumscribed set of organisms. If I curcumscribe a set of organisms that I label with "Aedes triseriatus", and then later decide that this set of organisms is best classified in the genus Ocherotatus, then naming it as such does not change the circumscribed set of organisms. It certainly may have implications on the concepts for the genera "Aedes" and "Ochlerotatus", but in my mind, it has no effect on the implied circumscription (=Concept, sensu me) of what is indicated by the species epithet "triseriatus". I do not think it makes sense to include hierarchical clasification as part of the terminal taxon "concept". Taxa at each hierarchical rank are, in my mind, defined by their contents; not their higher classification. The way I visualize it, there is a many:many relationship between names and concepts (I *think* this applies no matter what you mean by "name", and no matter what you mean by "concept"). The same circumscription of organisms can be labelled by many different names, and the same name may apply to many different circumscriptions of organisms (not just homonyms/homographs, but also lumper/splitter issues). Therefore, I don't think it is appropriate to try to equate names:concepts as 1:1, or even many:1. Aloha, Rich P.S. I certainly think that "Aedes triseriatus" and "Ochlerotatus triseriatus" are different "things", just not (necessarily) different taxon concepts. Actually, from an informatics perspective, I think that treating these different combinations as unique/identified objects doesn't gain us much. I think it's *MUCH* more robust to parse out the different individual usages of each combination as the identified objects, then derive the unique combinations/spellings/etc. from those usages. If the notion of indexing usages seems too intimidating, then start with the easy ones -- like the original useages of each of the name elements ("Aedes", "Ochlerotatus", and "triseriatus"), and the key treatments (e.g., whoever first combined "triseriatus" with the genus "Ochlerotatus", and/or whoever robustly defined alternate concepts for each).
________________________________
From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Peter DeVries Sent: Thursday, June 10, 2010 9:38 AM To: tdwg-content@lists.tdwg.org Subject: [tdwg-content] Name is species concept thinking I was looking through the latest DarwinCore and comments related to the TaxonConceptID.
Since the name serves as both a unique identifier and a phylogenetic hypothesis, you are effectively saying that observations labeled
Aedes triseriatus
and
Ochlerotatus triseriatus
Are separate species concepts, and should therefore be treated as separate things.
i.e. The name is the concept.
Also since there are several name variants for each "species", how do you distinguish which of these nameID's are the same species and which are different?
- Pete
---------------------------------------------------------------- Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base ------------------------------------------------------------
On Thu, Jun 10, 2010 at 1:00 PM, Richard Pyle deepreef@bishopmuseum.org wrote:
I guess a key question that I'd like to get feedback from others on is whether "Parent Taxon" and "circumscribed organisms" are *both* intrinsic properties of "taxon concept", or if "taxon concept" is effectively equivalent to "circumscription", relegating "parent taxon" as a property of Classification, separate from "Taxon Concept".
But "Parent Taxon" is a circumscription itself. Its children are circumscriptions of circumscriptions. A specimen is the ultimate circumscription.
///ark
Hi Richard
You are agreeing with me.
My point is that the DarwinCore should include an identifier that resolves to a site that provides some information about how that concept is defined.
1) First by mapping the concept to the various names and related identifiers 2) By providing additional information that helps one determine if a specimen is a close match to that species concept.
The field nameAccordingTo does not provide this kind of information.
- Pete
On Thu, Jun 10, 2010 at 2:55 PM, Richard Pyle deepreef@bishopmuseum.orgwrote:
I don't think that's right. Names are pointers to species concepts, but are not concepts themselves. A label for a species concept would look like "Aus bus sensu Author/Citation" or "Aus bus sec. Author/Citation". Also, of course, the name is not, in itself, a unique identifier (in something like 10% of cases -- which in my mind is non-trivial).
Also, I disagree with the idea that Aedes triseriatus and Ochlerotatus triseriatus necessarily refer to different taxon concepts. To me, the "concept" is the circumscribed set of organisms. If I curcumscribe a set of organisms that I label with "Aedes triseriatus", and then later decide that this set of organisms is best classified in the genus Ocherotatus, then naming it as such does not change the circumscribed set of organisms. It certainly may have implications on the concepts for the genera "Aedes" and "Ochlerotatus", but in my mind, it has no effect on the implied circumscription (=Concept, sensu me) of what is indicated by the species epithet "triseriatus".
I do not think it makes sense to include hierarchical clasification as part of the terminal taxon "concept". Taxa at each hierarchical rank are, in my mind, defined by their contents; not their higher classification.
The way I visualize it, there is a many:many relationship between names and concepts (I *think* this applies no matter what you mean by "name", and no matter what you mean by "concept"). The same circumscription of organisms can be labelled by many different names, and the same name may apply to many different circumscriptions of organisms (not just homonyms/homographs, but also lumper/splitter issues).
Therefore, I don't think it is appropriate to try to equate names:concepts as 1:1, or even many:1.
Aloha, Rich
P.S. I certainly think that "Aedes triseriatus" and "Ochlerotatus triseriatus" are different "things", just not (necessarily) different taxon concepts. Actually, from an informatics perspective, I think that treating these different combinations as unique/identified objects doesn't gain us much. I think it's *MUCH* more robust to parse out the different individual usages of each combination as the identified objects, then derive the unique combinations/spellings/etc. from those usages. If the notion of indexing usages seems too intimidating, then start with the easy ones -- like the original useages of each of the name elements ("Aedes", "Ochlerotatus", and "triseriatus"), and the key treatments (e.g., whoever first combined "triseriatus" with the genus "Ochlerotatus", and/or whoever robustly defined alternate concepts for each).
*From:* tdwg-content-bounces@lists.tdwg.org [mailto: tdwg-content-bounces@lists.tdwg.org] *On Behalf Of *Peter DeVries *Sent:* Thursday, June 10, 2010 9:38 AM *To:* tdwg-content@lists.tdwg.org *Subject:* [tdwg-content] Name is species concept thinking
I was looking through the latest DarwinCore and comments related to the TaxonConceptID.
Since the name serves as both a unique identifier and a phylogenetic hypothesis, you are effectively saying that observations labeled
*Aedes triseriatus*
and
*Ochlerotatus triseriatus*
Are separate species concepts, and should therefore be treated as separate things.
i.e. The name is the concept.
Also since there are several name variants for each "species", how do you distinguish which of these nameID's are the same species and which are different?
- Pete
Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base
Hmmm....OK, I'm not sure I follow, then.
I certainly wouldn't use nameAccordingTo/ID to reference attributes of a taxon concept -- that simply points to a documentation source in which a usage instance occurred. However, I *do* think that acceptedNameUsage/ID could resolve to an object that would lead you directly to all the name mappings, at least. As for the additional information to resolve concepts (e.g., diagnoses, keys, etc.), I have always imagined that sort of thing as a service layer that assembles sets of usage instances and establishes typed relationships among them. This, then, would bridge to the correct set of concept-definition information (diagnoses, etc.).
Also I thought that's what taxonConcept/ID was intended for? (I was never very clear on what it's exact intended purpose was.)
So....perhaps I should ask you to re-state your original question? The premise of your question was, "The name is the concept"; which is what I disagreed with. Also, I'm not sure what you mean by "these nameID's". As far as I can tell, all of the XXXXNameID terms in DwC have been superseded. The only exception seems to be "scientificNameID" of 2009-08-24, which is listed as "recommended", and which is said to replace "taxonNameID" of 2009-04-24. But the 2009-07-06 version of "scientificNameID" is said to be replaced by "taxonID" of 2009-08-04. I'm not quite sure how to interpret that.
Rich
________________________________
From: Peter DeVries [mailto:pete.devries@gmail.com] Sent: Thursday, June 10, 2010 10:09 AM To: Richard Pyle Cc: tdwg-content@lists.tdwg.org Subject: Re: [tdwg-content] Name is species concept thinking Hi Richard
You are agreeing with me.
My point is that the DarwinCore should include an identifier that resolves to a site that provides some information about how that concept is defined. 1) First by mapping the concept to the various names and related identifiers 2) By providing additional information that helps one determine if a specimen is a close match to that species concept.
The field nameAccordingTo does not provide this kind of information.
- Pete On Thu, Jun 10, 2010 at 2:55 PM, Richard Pyle deepreef@bishopmuseum.org wrote:
I don't think that's right. Names are pointers to species concepts, but are not concepts themselves. A label for a species concept would look like "Aus bus sensu Author/Citation" or "Aus bus sec. Author/Citation". Also, of course, the name is not, in itself, a unique identifier (in something like 10% of cases -- which in my mind is non-trivial). Also, I disagree with the idea that Aedes triseriatus and Ochlerotatus triseriatus necessarily refer to different taxon concepts. To me, the "concept" is the circumscribed set of organisms. If I curcumscribe a set of organisms that I label with "Aedes triseriatus", and then later decide that this set of organisms is best classified in the genus Ocherotatus, then naming it as such does not change the circumscribed set of organisms. It certainly may have implications on the concepts for the genera "Aedes" and "Ochlerotatus", but in my mind, it has no effect on the implied circumscription (=Concept, sensu me) of what is indicated by the species epithet "triseriatus". I do not think it makes sense to include hierarchical clasification as part of the terminal taxon "concept". Taxa at each hierarchical rank are, in my mind, defined by their contents; not their higher classification. The way I visualize it, there is a many:many relationship between names and concepts (I *think* this applies no matter what you mean by "name", and no matter what you mean by "concept"). The same circumscription of organisms can be labelled by many different names, and the same name may apply to many different circumscriptions of organisms (not just homonyms/homographs, but also lumper/splitter issues). Therefore, I don't think it is appropriate to try to equate names:concepts as 1:1, or even many:1. Aloha, Rich P.S. I certainly think that "Aedes triseriatus" and "Ochlerotatus triseriatus" are different "things", just not (necessarily) different taxon concepts. Actually, from an informatics perspective, I think that treating these different combinations as unique/identified objects doesn't gain us much. I think it's *MUCH* more robust to parse out the different individual usages of each combination as the identified objects, then derive the unique combinations/spellings/etc. from those usages. If the notion of indexing usages seems too intimidating, then start with the easy ones -- like the original useages of each of the name elements ("Aedes", "Ochlerotatus", and "triseriatus"), and the key treatments (e.g., whoever first combined "triseriatus" with the genus "Ochlerotatus", and/or whoever robustly defined alternate concepts for each).
________________________________
From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Peter DeVries Sent: Thursday, June 10, 2010 9:38 AM To: tdwg-content@lists.tdwg.org Subject: [tdwg-content] Name is species concept thinking I was looking through the latest DarwinCore and comments related to the TaxonConceptID.
Since the name serves as both a unique identifier and a phylogenetic hypothesis, you are effectively saying that observations labeled
Aedes triseriatus
and
Ochlerotatus triseriatus
Are separate species concepts, and should therefore be treated as separate things.
i.e. The name is the concept.
Also since there are several name variants for each "species", how do you distinguish which of these nameID's are the same species and which are different?
- Pete
---------------------------------------------------------------- Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base ------------------------------------------------------------
-- ---------------------------------------------------------------- Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base ------------------------------------------------------------
On Jun 10, 2010, at 10:09 PM, Peter DeVries wrote:
Hi Richard
You are agreeing with me.
My point is that the DarwinCore should include an identifier that resolves to a site that provides some information about how that concept is defined.
Wouldn't the taxonConceptID provide that?
- First by mapping the concept to the various names and related
identifiers
- By providing additional information that helps one determine if a
specimen is a close match to that species concept.
What sort of additional information? Literature, specimens, and descriptions can be included even using the text guidelines as we have done. There must be even more detailed ways to do it. These are tied to the taxonID which could be linked to a resolvable taxon concept ID.
The field nameAccordingTo does not provide this kind of information.
- Pete
On Thu, Jun 10, 2010 at 2:55 PM, Richard Pyle <deepreef@bishopmuseum.org
wrote:
I don't think that's right. Names are pointers to species concepts, but are not concepts themselves. A label for a species concept would look like "Aus bus sensu Author/Citation" or "Aus bus sec. Author/Citation". Also, of course, the name is not, in itself, a unique identifier (in something like 10% of cases -- which in my mind is non-trivial).
Also, I disagree with the idea that Aedes triseriatus and Ochlerotatus triseriatus necessarily refer to different taxon concepts.
I agree with Rich here. The two names themselves can only be inferred to refer to the same type. In a given instance they could either be used to refer to the same or a different concept. I would think that this must be true for any binomial combination referring to the same type. On the other hand, names referring to different types could not validly refer to the same concept if my current late night thinking is correct. A circumscription that includes multiple types must follow a set of rules to determine the name.
To me, the "concept" is the circumscribed set of organisms. If I curcumscribe a set of organisms that I label with "Aedes triseriatus", and then later decide that this set of organisms is best classified in the genus Ocherotatus, then naming it as such does not change the circumscribed set of organisms. It certainly may have implications on the concepts for the genera "Aedes" and "Ochlerotatus", but in my mind, it has no effect on the implied circumscription (=Concept, sensu me) of what is indicated by the species epithet "triseriatus".
I do not think it makes sense to include hierarchical clasification as part of the terminal taxon "concept". Taxa at each hierarchical rank are, in my mind, defined by their contents; not their higher classification.
I also agree with Rich here. A concept shouldn't change based on how it is categorized.
The way I visualize it, there is a many:many relationship between names and concepts (I *think* this applies no matter what you mean by "name", and no matter what you mean by "concept"). The same circumscription of organisms can be labelled by many different names, and the same name may apply to many different circumscriptions of organisms (not just homonyms/homographs, but also lumper/splitter issues).
Therefore, I don't think it is appropriate to try to equate names:concepts as 1:1, or even many:1.
Aloha, Rich
P.S. I certainly think that "Aedes triseriatus" and "Ochlerotatus triseriatus" are different "things", just not (necessarily) different taxon concepts. Actually, from an informatics perspective, I think that treating these different combinations as unique/identified objects doesn't gain us much. I think it's *MUCH* more robust to parse out the different individual usages of each combination as the identified objects, then derive the unique combinations/spellings/etc. from those usages. If the notion of indexing usages seems too intimidating, then start with the easy ones -- like the original useages of each of the name elements ("Aedes", "Ochlerotatus", and "triseriatus"), and the key treatments (e.g., whoever first combined "triseriatus" with the genus "Ochlerotatus", and/or whoever robustly defined alternate concepts for each).
From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org ] On Behalf Of Peter DeVries Sent: Thursday, June 10, 2010 9:38 AM To: tdwg-content@lists.tdwg.org Subject: [tdwg-content] Name is species concept thinking
I was looking through the latest DarwinCore and comments related to the TaxonConceptID.
Since the name serves as both a unique identifier and a phylogenetic hypothesis, you are effectively saying that observations labeled
Aedes triseriatus
and
Ochlerotatus triseriatus
Are separate species concepts, and should therefore be treated as separate things.
i.e. The name is the concept.
Also since there are several name variants for each "species", how do you distinguish which of these nameID's are the same species and which are different?
- Pete
Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base
--
Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base
tdwg-content mailing list tdwg-content@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-content
Yes, good point!
This is something that has been slightly confused over the years, ie there seems to be 2 ways of defining a "taxon concept":
1. A Taxon Name (nomenclatural data) + Literature Reference - ie Name X as defined in article Y
2. As you have said a grouping of data that define a taxon concept (Name + Reference + Synonyms + Type Specimen + Protologue, ...)
1 has been covered quite well with the various schemas we have come up with over the years, but I think these schemas have failed to capture 2 very well (the data fields are there, but the encompassing ID is not), ie
TaxonName ID = N1, FullName = "Aus bus" Reference ID = R1, Citation = "Richards, how to define a taxon concept" TaxonConcept ID = C1, NameID = N1, ReferenceID = R1
BUT, the taxon concept C1 does not encompass all related data that defines that concept (synonyms etc)
To do that we need more Concept Ids and relationships between these concepts, eg
TaxonName ID = N2, FullName = "Aus cus" Reference ID = R2, Citation = "Richards, how to define a taxon concept" TaxonConcept ID = C2, NameID = N2, ReferenceID = R2 ConceptRelationship ID=CR1 ConceptFromID=C2, ConceptToID =C1, RelationshipType='has preferred name'
BUT this means there is no "Concept" ID that covers all of this data??
Kevin
From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Peter DeVries Sent: Friday, 11 June 2010 8:09 a.m. To: Richard Pyle Cc: tdwg-content@lists.tdwg.org Subject: Re: [tdwg-content] Name is species concept thinking
Hi Richard
You are agreeing with me.
My point is that the DarwinCore should include an identifier that resolves to a site that provides some information about how that concept is defined.
1) First by mapping the concept to the various names and related identifiers 2) By providing additional information that helps one determine if a specimen is a close match to that species concept.
The field nameAccordingTo does not provide this kind of information.
- Pete On Thu, Jun 10, 2010 at 2:55 PM, Richard Pyle <deepreef@bishopmuseum.orgmailto:deepreef@bishopmuseum.org> wrote: I don't think that's right. Names are pointers to species concepts, but are not concepts themselves. A label for a species concept would look like "Aus bus sensu Author/Citation" or "Aus bus sec. Author/Citation". Also, of course, the name is not, in itself, a unique identifier (in something like 10% of cases -- which in my mind is non-trivial).
Also, I disagree with the idea that Aedes triseriatus and Ochlerotatus triseriatus necessarily refer to different taxon concepts. To me, the "concept" is the circumscribed set of organisms. If I curcumscribe a set of organisms that I label with "Aedes triseriatus", and then later decide that this set of organisms is best classified in the genus Ocherotatus, then naming it as such does not change the circumscribed set of organisms. It certainly may have implications on the concepts for the genera "Aedes" and "Ochlerotatus", but in my mind, it has no effect on the implied circumscription (=Concept, sensu me) of what is indicated by the species epithet "triseriatus".
I do not think it makes sense to include hierarchical clasification as part of the terminal taxon "concept". Taxa at each hierarchical rank are, in my mind, defined by their contents; not their higher classification.
The way I visualize it, there is a many:many relationship between names and concepts (I *think* this applies no matter what you mean by "name", and no matter what you mean by "concept"). The same circumscription of organisms can be labelled by many different names, and the same name may apply to many different circumscriptions of organisms (not just homonyms/homographs, but also lumper/splitter issues).
Therefore, I don't think it is appropriate to try to equate names:concepts as 1:1, or even many:1.
Aloha, Rich
P.S. I certainly think that "Aedes triseriatus" and "Ochlerotatus triseriatus" are different "things", just not (necessarily) different taxon concepts. Actually, from an informatics perspective, I think that treating these different combinations as unique/identified objects doesn't gain us much. I think it's *MUCH* more robust to parse out the different individual usages of each combination as the identified objects, then derive the unique combinations/spellings/etc. from those usages. If the notion of indexing usages seems too intimidating, then start with the easy ones -- like the original useages of each of the name elements ("Aedes", "Ochlerotatus", and "triseriatus"), and the key treatments (e.g., whoever first combined "triseriatus" with the genus "Ochlerotatus", and/or whoever robustly defined alternate concepts for each).
________________________________ From: tdwg-content-bounces@lists.tdwg.orgmailto:tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.orgmailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Peter DeVries Sent: Thursday, June 10, 2010 9:38 AM To: tdwg-content@lists.tdwg.orgmailto:tdwg-content@lists.tdwg.org Subject: [tdwg-content] Name is species concept thinking I was looking through the latest DarwinCore and comments related to the TaxonConceptID.
Since the name serves as both a unique identifier and a phylogenetic hypothesis, you are effectively saying that observations labeled
Aedes triseriatus
and
Ochlerotatus triseriatus
Are separate species concepts, and should therefore be treated as separate things.
i.e. The name is the concept.
Also since there are several name variants for each "species", how do you distinguish which of these nameID's are the same species and which are different?
- Pete
---------------------------------------------------------------- Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base ------------------------------------------------------------
-- ---------------------------------------------------------------- Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base ------------------------------------------------------------
________________________________ Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
This is something that has been slightly confused over the years, ie there
seems to be 2 ways of defining a "taxon concept":
- A Taxon Name (nomenclatural data) + Literature Reference - ie Name X as
defined in article Y
- As you have said a grouping of data that define a taxon concept (Name +
Reference + Synonyms + Type Specimen + Protologue, .)
I don't think of these as two different ways of defining a concept. I see #1 as a way of *pointing to* a taxon concept definition, and #2 as the concept definition itself. Basically, #1 (usage instance) is effectively a container or an identifier for the taxon concept definition.
However, there is somewhat of a dichotmy in the way that taxon concepts are defined - one is by included members (i.e., specimens, presumably including at least one name-bearing type specimen, from which a name-label is derived), the other is by properties (i.e. characters -- morphologic, genetic, or otherwise). In practice, most concept definitions include both. But I think the "definition" of the concept (i.e., the circumscription boundaries) is the same for both -- it's just that those boundaries can be articulated in different ways (i.e., by examplar members, and by purported properties).
1 has been covered quite well with the various schemas we have come up with over the years, but I think these schemas have failed to capture 2 very well (the data fields are there, but the encompassing ID is not),
ie
Agreed -- sort of. I think the schemas are there, but have not been organized appropriately (yet). See below.
TaxonName ID = N1, FullName = "Aus bus" Reference ID = R1, Citation = "Richards, how to define a taxon concept" TaxonConcept ID = C1, NameID = N1, ReferenceID = R1 BUT, the taxon concept C1 does not encompass all related data that defines
that concept (synonyms etc)
No, but it could, through a network of linkages, as I tried to describe in one of my recent posts.
To do that we need more Concept Ids and relationships between these
concepts, eg
Exactly! And we need a schema-based process to capture the relevant information (diagnoses, etc.), anchored to the Concept Ids. At a basic level, Plazi/TaxonX does this. However, it usually only goes as far as the text-blob. To parse the text blob, we need to either look towards SDD (for character-based concept definition stuff) or DwC/Occurrence (for specimen-based concept definition stuff).
ConceptRelationship ID=CR1 ConceptFromID=C2, ConceptToID =C1,
RelationshipType='has preferred name'
Yes, I agree we need this as well! But again, I see this as a way of networking pointers to taxon concept defintions, not describing the definitions themselves.
Man, these conversations really hurt my brain.... :-)
Aloha, Rich
I think my main point here was the fact that in most schemas we (TDWG, et al) have created, we have not really provided an ID field for (2). As you said (2) is the "concept definition" but there is no ID field (that I have come across), for referring to it explicitly.
My thought would be to have something like:
http://zoobank.org/taxonconcept/12345-ABCDE
that returns data for the "whole" taxon concept (ie 2), not just the Name + Reference
I think it is really a data/technical issue - ie the way the schemas/models are defined, a Taxon Concept ID includes, and only includes, a Name ID and a Reference ID. This is based on my understanding of TCS - perhaps DwC is different??
Kevin
-----Original Message----- From: Richard Pyle [mailto:deepreef@bishopmuseum.org] Sent: Friday, 11 June 2010 9:03 a.m. To: Kevin Richards; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
This is something that has been slightly confused over the years, ie there
seems to be 2 ways of defining a "taxon concept":
- A Taxon Name (nomenclatural data) + Literature Reference - ie Name X as
defined in article Y
- As you have said a grouping of data that define a taxon concept (Name +
Reference + Synonyms + Type Specimen + Protologue, .)
I don't think of these as two different ways of defining a concept. I see #1 as a way of *pointing to* a taxon concept definition, and #2 as the concept definition itself. Basically, #1 (usage instance) is effectively a container or an identifier for the taxon concept definition.
However, there is somewhat of a dichotmy in the way that taxon concepts are defined - one is by included members (i.e., specimens, presumably including at least one name-bearing type specimen, from which a name-label is derived), the other is by properties (i.e. characters -- morphologic, genetic, or otherwise). In practice, most concept definitions include both. But I think the "definition" of the concept (i.e., the circumscription boundaries) is the same for both -- it's just that those boundaries can be articulated in different ways (i.e., by examplar members, and by purported properties).
1 has been covered quite well with the various schemas we have come up with over the years, but I think these schemas have failed to capture 2 very well (the data fields are there, but the encompassing ID is not),
ie
Agreed -- sort of. I think the schemas are there, but have not been organized appropriately (yet). See below.
TaxonName ID = N1, FullName = "Aus bus" Reference ID = R1, Citation = "Richards, how to define a taxon concept" TaxonConcept ID = C1, NameID = N1, ReferenceID = R1 BUT, the taxon concept C1 does not encompass all related data that defines
that concept (synonyms etc)
No, but it could, through a network of linkages, as I tried to describe in one of my recent posts.
To do that we need more Concept Ids and relationships between these
concepts, eg
Exactly! And we need a schema-based process to capture the relevant information (diagnoses, etc.), anchored to the Concept Ids. At a basic level, Plazi/TaxonX does this. However, it usually only goes as far as the text-blob. To parse the text blob, we need to either look towards SDD (for character-based concept definition stuff) or DwC/Occurrence (for specimen-based concept definition stuff).
ConceptRelationship ID=CR1 ConceptFromID=C2, ConceptToID =C1,
RelationshipType='has preferred name'
Yes, I agree we need this as well! But again, I see this as a way of networking pointers to taxon concept defintions, not describing the definitions themselves.
Man, these conversations really hurt my brain.... :-)
Aloha, Rich
Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
Are these concepts fixed to a particular phylogenetic hypothesis i.e. a particular genus?
What I am getting at is that for a number of uses. Aedes triseriatus and Ochlerotatus triseriatus are the same "species concept"
Because the character "pattern" is the same and the set of individuals that are considered instances of that concept are the same.
The only difference is that someone has made the assertion that the instances of that concept have a different phylogeny.
Also in the case of this particular species, the original description is not very informative and *I believe* the original type specimen has been lost.
- Pete
On Thu, Jun 10, 2010 at 4:21 PM, Kevin Richards < RichardsK@landcareresearch.co.nz> wrote:
I think my main point here was the fact that in most schemas we (TDWG, et al) have created, we have not really provided an ID field for (2). As you said (2) is the "concept definition" but there is no ID field (that I have come across), for referring to it explicitly.
My thought would be to have something like:
http://zoobank.org/taxonconcept/12345-ABCDE
that returns data for the "whole" taxon concept (ie 2), not just the Name + Reference
I think it is really a data/technical issue - ie the way the schemas/models are defined, a Taxon Concept ID includes, and only includes, a Name ID and a Reference ID. This is based on my understanding of TCS - perhaps DwC is different??
Kevin
-----Original Message----- From: Richard Pyle [mailto:deepreef@bishopmuseum.org] Sent: Friday, 11 June 2010 9:03 a.m. To: Kevin Richards; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
This is something that has been slightly confused over the years, ie
there seems to be 2 ways of defining a "taxon concept":
- A Taxon Name (nomenclatural data) + Literature Reference - ie Name X
as defined in article Y
- As you have said a grouping of data that define a taxon concept (Name
Reference + Synonyms + Type Specimen + Protologue, .)
I don't think of these as two different ways of defining a concept. I see #1 as a way of *pointing to* a taxon concept definition, and #2 as the concept definition itself. Basically, #1 (usage instance) is effectively a container or an identifier for the taxon concept definition.
However, there is somewhat of a dichotmy in the way that taxon concepts are defined - one is by included members (i.e., specimens, presumably including at least one name-bearing type specimen, from which a name-label is derived), the other is by properties (i.e. characters -- morphologic, genetic, or otherwise). In practice, most concept definitions include both. But I think the "definition" of the concept (i.e., the circumscription boundaries) is the same for both -- it's just that those boundaries can be articulated in different ways (i.e., by examplar members, and by purported properties).
1 has been covered quite well with the various schemas we have come up with over the years, but I think these schemas have failed to capture 2 very well (the data fields are there, but the encompassing ID is not),
ie
Agreed -- sort of. I think the schemas are there, but have not been organized appropriately (yet). See below.
TaxonName ID = N1, FullName = "Aus bus" Reference ID = R1, Citation = "Richards, how to define a taxon concept" TaxonConcept ID = C1, NameID = N1, ReferenceID = R1 BUT, the taxon concept C1 does not encompass all related data that
defines that concept (synonyms etc)
No, but it could, through a network of linkages, as I tried to describe in one of my recent posts.
To do that we need more Concept Ids and relationships between these
concepts, eg
Exactly! And we need a schema-based process to capture the relevant information (diagnoses, etc.), anchored to the Concept Ids. At a basic level, Plazi/TaxonX does this. However, it usually only goes as far as the text-blob. To parse the text blob, we need to either look towards SDD (for character-based concept definition stuff) or DwC/Occurrence (for specimen-based concept definition stuff).
ConceptRelationship ID=CR1 ConceptFromID=C2, ConceptToID =C1,
RelationshipType='has preferred name'
Yes, I agree we need this as well! But again, I see this as a way of networking pointers to taxon concept defintions, not describing the definitions themselves.
Man, these conversations really hurt my brain.... :-)
Aloha, Rich
Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
You said: "What I am getting at is that for a number of uses. Aedes triseriatus and Ochlerotatus triseriatus are the same "species concept""
I would think that would be up to the consumer of the data - they could interpret it that way if they like. (but this is then probably another concept in it self?? Ie "Aedes triseriatus and Ochlerotatus triseriatus are the same "species concept" according to Consumer X" ??
From: Peter DeVries [mailto:pete.devries@gmail.com] Sent: Friday, 11 June 2010 9:29 a.m. To: Kevin Richards Cc: Richard Pyle; tdwg-content@lists.tdwg.org; Jerry Cooper Subject: Re: [tdwg-content] Name is species concept thinking
Are these concepts fixed to a particular phylogenetic hypothesis i.e. a particular genus?
What I am getting at is that for a number of uses. Aedes triseriatus and Ochlerotatus triseriatus are the same "species concept"
Because the character "pattern" is the same and the set of individuals that are considered instances of that concept are the same.
The only difference is that someone has made the assertion that the instances of that concept have a different phylogeny.
Also in the case of this particular species, the original description is not very informative and I believe the original type specimen has been lost.
- Pete On Thu, Jun 10, 2010 at 4:21 PM, Kevin Richards <RichardsK@landcareresearch.co.nzmailto:RichardsK@landcareresearch.co.nz> wrote: I think my main point here was the fact that in most schemas we (TDWG, et al) have created, we have not really provided an ID field for (2). As you said (2) is the "concept definition" but there is no ID field (that I have come across), for referring to it explicitly.
My thought would be to have something like:
http://zoobank.org/taxonconcept/12345-ABCDE
that returns data for the "whole" taxon concept (ie 2), not just the Name + Reference
I think it is really a data/technical issue - ie the way the schemas/models are defined, a Taxon Concept ID includes, and only includes, a Name ID and a Reference ID. This is based on my understanding of TCS - perhaps DwC is different??
Kevin
-----Original Message----- From: Richard Pyle [mailto:deepreef@bishopmuseum.orgmailto:deepreef@bishopmuseum.org] Sent: Friday, 11 June 2010 9:03 a.m. To: Kevin Richards; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.orgmailto:tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
This is something that has been slightly confused over the years, ie there
seems to be 2 ways of defining a "taxon concept":
- A Taxon Name (nomenclatural data) + Literature Reference - ie Name X as
defined in article Y
- As you have said a grouping of data that define a taxon concept (Name +
Reference + Synonyms + Type Specimen + Protologue, .)
I don't think of these as two different ways of defining a concept. I see #1 as a way of *pointing to* a taxon concept definition, and #2 as the concept definition itself. Basically, #1 (usage instance) is effectively a container or an identifier for the taxon concept definition.
However, there is somewhat of a dichotmy in the way that taxon concepts are defined - one is by included members (i.e., specimens, presumably including at least one name-bearing type specimen, from which a name-label is derived), the other is by properties (i.e. characters -- morphologic, genetic, or otherwise). In practice, most concept definitions include both. But I think the "definition" of the concept (i.e., the circumscription boundaries) is the same for both -- it's just that those boundaries can be articulated in different ways (i.e., by examplar members, and by purported properties).
1 has been covered quite well with the various schemas we have come up with over the years, but I think these schemas have failed to capture 2 very well (the data fields are there, but the encompassing ID is not),
ie
Agreed -- sort of. I think the schemas are there, but have not been organized appropriately (yet). See below.
TaxonName ID = N1, FullName = "Aus bus" Reference ID = R1, Citation = "Richards, how to define a taxon concept" TaxonConcept ID = C1, NameID = N1, ReferenceID = R1 BUT, the taxon concept C1 does not encompass all related data that defines
that concept (synonyms etc)
No, but it could, through a network of linkages, as I tried to describe in one of my recent posts.
To do that we need more Concept Ids and relationships between these
concepts, eg
Exactly! And we need a schema-based process to capture the relevant information (diagnoses, etc.), anchored to the Concept Ids. At a basic level, Plazi/TaxonX does this. However, it usually only goes as far as the text-blob. To parse the text blob, we need to either look towards SDD (for character-based concept definition stuff) or DwC/Occurrence (for specimen-based concept definition stuff).
ConceptRelationship ID=CR1 ConceptFromID=C2, ConceptToID =C1,
RelationshipType='has preferred name'
Yes, I agree we need this as well! But again, I see this as a way of networking pointers to taxon concept defintions, not describing the definitions themselves.
Man, these conversations really hurt my brain.... :-)
Aloha, Rich
Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
-- ---------------------------------------------------------------- Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base ------------------------------------------------------------
________________________________ Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
Yes, the consumer could interpret these anyway they want.
What I am wondering about is if we are missing the intent of the identifier.
If they choose a particular concept then you know that name can probably be changed to match the current phylogenetic thinking.
If they choose only a particular lexical variant you don't know if they intended that particular combination of genus epithet authority or if they simply copied down the name that was next to where key ended.
- Pete
On Thu, Jun 10, 2010 at 4:33 PM, Kevin Richards < RichardsK@landcareresearch.co.nz> wrote:
You said: “What I am getting at is that for a number of uses. Aedes triseriatus and Ochlerotatus triseriatus are the same "species concept"“
I would think that would be up to the consumer of the data – they could interpret it that way if they like. (but this is then probably another concept in it self?? Ie “Aedes triseriatus and Ochlerotatus triseriatus are the same "species concept" according to Consumer X” ??
*From:* Peter DeVries [mailto:pete.devries@gmail.com] *Sent:* Friday, 11 June 2010 9:29 a.m. *To:* Kevin Richards *Cc:* Richard Pyle; tdwg-content@lists.tdwg.org; Jerry Cooper *Subject:* Re: [tdwg-content] Name is species concept thinking
Are these concepts fixed to a particular phylogenetic hypothesis i.e. a particular genus?
What I am getting at is that for a number of uses. Aedes triseriatus and Ochlerotatus triseriatus are the same "species concept"
Because the character "pattern" is the same and the set of individuals that are considered instances of that concept are the same.
The only difference is that someone has made the assertion that the instances of that concept have a different phylogeny.
Also in the case of this particular species, the original description is not very informative and *I believe* the original type specimen has been lost.
- Pete
On Thu, Jun 10, 2010 at 4:21 PM, Kevin Richards < RichardsK@landcareresearch.co.nz> wrote:
I think my main point here was the fact that in most schemas we (TDWG, et al) have created, we have not really provided an ID field for (2). As you said (2) is the "concept definition" but there is no ID field (that I have come across), for referring to it explicitly.
My thought would be to have something like:
http://zoobank.org/taxonconcept/12345-ABCDE
that returns data for the "whole" taxon concept (ie 2), not just the Name + Reference
I think it is really a data/technical issue - ie the way the schemas/models are defined, a Taxon Concept ID includes, and only includes, a Name ID and a Reference ID. This is based on my understanding of TCS - perhaps DwC is different??
Kevin
-----Original Message----- From: Richard Pyle [mailto:deepreef@bishopmuseum.org] Sent: Friday, 11 June 2010 9:03 a.m. To: Kevin Richards; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
This is something that has been slightly confused over the years, ie
there seems to be 2 ways of defining a "taxon concept":
- A Taxon Name (nomenclatural data) + Literature Reference - ie Name X
as defined in article Y
- As you have said a grouping of data that define a taxon concept (Name
Reference + Synonyms + Type Specimen + Protologue, .)
I don't think of these as two different ways of defining a concept. I see #1 as a way of *pointing to* a taxon concept definition, and #2 as the concept definition itself. Basically, #1 (usage instance) is effectively a container or an identifier for the taxon concept definition.
However, there is somewhat of a dichotmy in the way that taxon concepts are defined - one is by included members (i.e., specimens, presumably including at least one name-bearing type specimen, from which a name-label is derived), the other is by properties (i.e. characters -- morphologic, genetic, or otherwise). In practice, most concept definitions include both. But I think the "definition" of the concept (i.e., the circumscription boundaries) is the same for both -- it's just that those boundaries can be articulated in different ways (i.e., by examplar members, and by purported properties).
1 has been covered quite well with the various schemas we have come up with over the years, but I think these schemas have failed to capture 2 very well (the data fields are there, but the encompassing ID is not),
ie
Agreed -- sort of. I think the schemas are there, but have not been organized appropriately (yet). See below.
TaxonName ID = N1, FullName = "Aus bus" Reference ID = R1, Citation = "Richards, how to define a taxon concept" TaxonConcept ID = C1, NameID = N1, ReferenceID = R1 BUT, the taxon concept C1 does not encompass all related data that
defines that concept (synonyms etc)
No, but it could, through a network of linkages, as I tried to describe in one of my recent posts.
To do that we need more Concept Ids and relationships between these
concepts, eg
Exactly! And we need a schema-based process to capture the relevant information (diagnoses, etc.), anchored to the Concept Ids. At a basic level, Plazi/TaxonX does this. However, it usually only goes as far as the text-blob. To parse the text blob, we need to either look towards SDD (for character-based concept definition stuff) or DwC/Occurrence (for specimen-based concept definition stuff).
ConceptRelationship ID=CR1 ConceptFromID=C2, ConceptToID =C1,
RelationshipType='has preferred name'
Yes, I agree we need this as well! But again, I see this as a way of networking pointers to taxon concept defintions, not describing the definitions themselves.
Man, these conversations really hurt my brain.... :-)
Aloha, Rich
Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
--
Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base
Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
Well, for starters, zoobank is not in the business of dealing with taxon concepts -- so I wouldn't use that domain name in the example.
Also, what *is* the correct pl;ace to get DwC? Peter was using http://code.google.com/p/darwincore/wiki/Taxon; but I've been using http://rs.tdwg.org/dwc/index.htm. Now I'm thoroughly confused....
Rich
-----Original Message----- From: Kevin Richards [mailto:RichardsK@landcareresearch.co.nz] Sent: Thursday, June 10, 2010 11:22 AM To: Richard Pyle; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
I think my main point here was the fact that in most schemas we (TDWG, et al) have created, we have not really provided an ID field for (2). As you said (2) is the "concept definition" but there is no ID field (that I have come across), for referring to it explicitly.
My thought would be to have something like:
http://zoobank.org/taxonconcept/12345-ABCDE
that returns data for the "whole" taxon concept (ie 2), not just the Name + Reference
I think it is really a data/technical issue - ie the way the schemas/models are defined, a Taxon Concept ID includes, and only includes, a Name ID and a Reference ID. This is based on my understanding of TCS - perhaps DwC is different??
Kevin
-----Original Message----- From: Richard Pyle [mailto:deepreef@bishopmuseum.org] Sent: Friday, 11 June 2010 9:03 a.m. To: Kevin Richards; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
This is something that has been slightly confused over the
years, ie
there
seems to be 2 ways of defining a "taxon concept":
- A Taxon Name (nomenclatural data) + Literature Reference
- ie Name
X as
defined in article Y
- As you have said a grouping of data that define a taxon concept
(Name +
Reference + Synonyms + Type Specimen + Protologue, .)
I don't think of these as two different ways of defining a concept. I see #1 as a way of *pointing to* a taxon concept definition, and #2 as the concept definition itself. Basically, #1 (usage instance) is effectively a container or an identifier for the taxon concept definition.
However, there is somewhat of a dichotmy in the way that taxon concepts are defined - one is by included members (i.e., specimens, presumably including at least one name-bearing type specimen, from which a name-label is derived), the other is by properties (i.e. characters -- morphologic, genetic, or otherwise). In practice, most concept definitions include both. But I think the "definition" of the concept (i.e., the circumscription boundaries) is the same for both -- it's just that those boundaries can be articulated in different ways (i.e., by examplar members, and by purported properties).
1 has been covered quite well with the various schemas we
have come up
with over the years, but I think these schemas have failed
to capture
2 very well (the data fields are there, but the encompassing ID is not),
ie
Agreed -- sort of. I think the schemas are there, but have not been organized appropriately (yet). See below.
TaxonName ID = N1, FullName = "Aus bus" Reference ID = R1, Citation = "Richards, how to define a
taxon concept"
TaxonConcept ID = C1, NameID = N1, ReferenceID = R1 BUT, the taxon concept C1 does not encompass all related data that defines
that concept (synonyms etc)
No, but it could, through a network of linkages, as I tried to describe in one of my recent posts.
To do that we need more Concept Ids and relationships between these
concepts, eg
Exactly! And we need a schema-based process to capture the relevant information (diagnoses, etc.), anchored to the Concept Ids. At a basic level, Plazi/TaxonX does this. However, it usually only goes as far as the text-blob. To parse the text blob, we need to either look towards SDD (for character-based concept definition stuff) or DwC/Occurrence (for specimen-based concept definition stuff).
ConceptRelationship ID=CR1 ConceptFromID=C2, ConceptToID =C1,
RelationshipType='has preferred name'
Yes, I agree we need this as well! But again, I see this as a way of networking pointers to taxon concept defintions, not describing the definitions themselves.
Man, these conversations really hurt my brain.... :-)
Aloha, Rich
Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
Yep, sorry about that - got carried away.
I always look at http://rs.tdwg.org/dwc/terms/ when looking at DwC stuff. The http://code.google.com/p/darwincore/wiki/Taxon is just for discussions - so does not include all the DwC terms/fields etc.
Kevin
-----Original Message----- From: Richard Pyle [mailto:deepreef@bishopmuseum.org] Sent: Friday, 11 June 2010 9:36 a.m. To: Kevin Richards; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
Well, for starters, zoobank is not in the business of dealing with taxon concepts -- so I wouldn't use that domain name in the example.
Also, what *is* the correct pl;ace to get DwC? Peter was using http://code.google.com/p/darwincore/wiki/Taxon; but I've been using http://rs.tdwg.org/dwc/index.htm. Now I'm thoroughly confused....
Rich
-----Original Message----- From: Kevin Richards [mailto:RichardsK@landcareresearch.co.nz] Sent: Thursday, June 10, 2010 11:22 AM To: Richard Pyle; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
I think my main point here was the fact that in most schemas we (TDWG, et al) have created, we have not really provided an ID field for (2). As you said (2) is the "concept definition" but there is no ID field (that I have come across), for referring to it explicitly.
My thought would be to have something like:
http://zoobank.org/taxonconcept/12345-ABCDE
that returns data for the "whole" taxon concept (ie 2), not just the Name + Reference
I think it is really a data/technical issue - ie the way the schemas/models are defined, a Taxon Concept ID includes, and only includes, a Name ID and a Reference ID. This is based on my understanding of TCS - perhaps DwC is different??
Kevin
-----Original Message----- From: Richard Pyle [mailto:deepreef@bishopmuseum.org] Sent: Friday, 11 June 2010 9:03 a.m. To: Kevin Richards; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
This is something that has been slightly confused over the
years, ie
there
seems to be 2 ways of defining a "taxon concept":
- A Taxon Name (nomenclatural data) + Literature Reference
- ie Name
X as
defined in article Y
- As you have said a grouping of data that define a taxon concept
(Name +
Reference + Synonyms + Type Specimen + Protologue, .)
I don't think of these as two different ways of defining a concept. I see #1 as a way of *pointing to* a taxon concept definition, and #2 as the concept definition itself. Basically, #1 (usage instance) is effectively a container or an identifier for the taxon concept definition.
However, there is somewhat of a dichotmy in the way that taxon concepts are defined - one is by included members (i.e., specimens, presumably including at least one name-bearing type specimen, from which a name-label is derived), the other is by properties (i.e. characters -- morphologic, genetic, or otherwise). In practice, most concept definitions include both. But I think the "definition" of the concept (i.e., the circumscription boundaries) is the same for both -- it's just that those boundaries can be articulated in different ways (i.e., by examplar members, and by purported properties).
1 has been covered quite well with the various schemas we
have come up
with over the years, but I think these schemas have failed
to capture
2 very well (the data fields are there, but the encompassing ID is not),
ie
Agreed -- sort of. I think the schemas are there, but have not been organized appropriately (yet). See below.
TaxonName ID = N1, FullName = "Aus bus" Reference ID = R1, Citation = "Richards, how to define a
taxon concept"
TaxonConcept ID = C1, NameID = N1, ReferenceID = R1 BUT, the taxon concept C1 does not encompass all related data that defines
that concept (synonyms etc)
No, but it could, through a network of linkages, as I tried to describe in one of my recent posts.
To do that we need more Concept Ids and relationships between these
concepts, eg
Exactly! And we need a schema-based process to capture the relevant information (diagnoses, etc.), anchored to the Concept Ids. At a basic level, Plazi/TaxonX does this. However, it usually only goes as far as the text-blob. To parse the text blob, we need to either look towards SDD (for character-based concept definition stuff) or DwC/Occurrence (for specimen-based concept definition stuff).
ConceptRelationship ID=CR1 ConceptFromID=C2, ConceptToID =C1,
RelationshipType='has preferred name'
Yes, I agree we need this as well! But again, I see this as a way of networking pointers to taxon concept defintions, not describing the definitions themselves.
Man, these conversations really hurt my brain.... :-)
Aloha, Rich
Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
Sorry - I meant to post this URL: http://rs.tdwg.org/dwc/terms/
Rich
-----Original Message----- From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Richard Pyle Sent: Thursday, June 10, 2010 11:36 AM To: 'Kevin Richards'; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; 'Jerry Cooper' Subject: Re: [tdwg-content] Name is species concept thinking
Well, for starters, zoobank is not in the business of dealing with taxon concepts -- so I wouldn't use that domain name in the example.
Also, what *is* the correct pl;ace to get DwC? Peter was using http://code.google.com/p/darwincore/wiki/Taxon; but I've been using http://rs.tdwg.org/dwc/index.htm. Now I'm thoroughly confused....
Rich
-----Original Message----- From: Kevin Richards [mailto:RichardsK@landcareresearch.co.nz] Sent: Thursday, June 10, 2010 11:22 AM To: Richard Pyle; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
I think my main point here was the fact that in most
schemas we (TDWG,
et al) have created, we have not really provided an ID
field for (2).
As you said (2) is the "concept definition" but there is no
ID field
(that I have come across), for referring to it explicitly.
My thought would be to have something like:
http://zoobank.org/taxonconcept/12345-ABCDE
that returns data for the "whole" taxon concept (ie 2), not
just the
Name + Reference
I think it is really a data/technical issue - ie the way the schemas/models are defined, a Taxon Concept ID includes, and only includes, a Name ID and a Reference ID. This is based on my understanding of TCS - perhaps DwC is different??
Kevin
-----Original Message----- From: Richard Pyle [mailto:deepreef@bishopmuseum.org] Sent: Friday, 11 June 2010 9:03 a.m. To: Kevin Richards; 'Peter DeVries' Cc: tdwg-content@lists.tdwg.org; Jerry Cooper Subject: RE: [tdwg-content] Name is species concept thinking
This is something that has been slightly confused over the
years, ie
there
seems to be 2 ways of defining a "taxon concept":
- A Taxon Name (nomenclatural data) + Literature Reference
- ie Name
X as
defined in article Y
- As you have said a grouping of data that define a
taxon concept
(Name +
Reference + Synonyms + Type Specimen + Protologue, .)
I don't think of these as two different ways of defining a
concept. I
see #1 as a way of *pointing to* a taxon concept definition, and #2 as the concept definition itself. Basically, #1 (usage instance) is effectively a container or an identifier for the taxon concept definition.
However, there is somewhat of a dichotmy in the way that taxon concepts are defined - one is by included members (i.e., specimens, presumably including at least one name-bearing type specimen, from which a name-label is derived), the other is by properties (i.e. characters -- morphologic, genetic, or otherwise). In
practice, most
concept definitions include both. But I think the "definition" of the concept (i.e., the
circumscription
boundaries) is the same for both -- it's just that those boundaries can be articulated in different ways (i.e., by examplar
members, and
by purported properties).
1 has been covered quite well with the various schemas we
have come up
with over the years, but I think these schemas have failed
to capture
2 very well (the data fields are there, but the
encompassing ID is
not),
ie
Agreed -- sort of. I think the schemas are there, but have
not been
organized appropriately (yet). See below.
TaxonName ID = N1, FullName = "Aus bus" Reference ID = R1, Citation = "Richards, how to define a
taxon concept"
TaxonConcept ID = C1, NameID = N1, ReferenceID = R1 BUT,
the taxon
concept C1 does not encompass all related data that defines
that concept (synonyms etc)
No, but it could, through a network of linkages, as I tried to describe in one of my recent posts.
To do that we need more Concept Ids and relationships
between these
concepts, eg
Exactly! And we need a schema-based process to capture the
relevant
information (diagnoses, etc.), anchored to the Concept Ids. At a basic level, Plazi/TaxonX does this. However, it usually only goes as far as the text-blob. To
parse the
text blob, we need to either look towards SDD (for character-based concept definition stuff) or DwC/Occurrence (for specimen-based concept definition stuff).
ConceptRelationship ID=CR1 ConceptFromID=C2, ConceptToID =C1,
RelationshipType='has preferred name'
Yes, I agree we need this as well! But again, I see this
as a way of
networking pointers to taxon concept defintions, not describing the definitions themselves.
Man, these conversations really hurt my brain.... :-)
Aloha, Rich
Please consider the environment before printing this email Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not
read, use,
disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails. The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
tdwg-content mailing list tdwg-content@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-content
On Jun 10, 2010, at 3:37 PM, Peter DeVries wrote:
i.e. The name is the concept.
No, it's only a label.
Also since there are several name variants for each "species", how do you distinguish which of these nameID's are the same species and which are different?
When you want to compute over taxonomies this will need to be determined through an ontology. That's how we make the distinctions in TTO (Teleost Taxonomy Ontology).
-hilmar
I should have probably been more clear that what I had stated previously was a bit of a rhetorical question.
There had been a species concept id in an earlier version of the DarwinCore.
I was checking out this page and noticed that it was gone.
http://code.google.com/p/darwincore/wiki/Taxon
http://code.google.com/p/darwincore/wiki/TaxonIt seems to have been either deleted or replaced with what I call "nameID's"
I have been working with the EOL Woods Hole group looking at ways that names can be connected to concepts as part of the GNI.
We have a sample set of about 70,000 species that have the following kinds of mappings:
1) basic mapping to lexical variants in the GNI 2) basic mapping to other related semantic web identifier that have a "similar" meaning 3) Basic mapping to foreign key identifiers in various databases 4) A place in the RDF for mapping the species concept to various publications
In the future, I would like to add additional information that will allow a human or machine to determine how well a particular specimen matches a particular concept.
These are not in the current version of the RDF but their is a placeholder for this kind of data.
These species concepts are not the only potential kind of concept, but they are a data set that can be used to try out and see what works.
What I was hoping was that there would be some field in the DarwinCore that would allow a user to map their specimen to the URI for a species concept.
These would not have to be only for these particular species concepts, but something that worked in a similar way.
Here are some examples of these species concepts that are live on the Linked Open Data Cloud.
The links on this page will take you to Sig.ma, a semantic web portal that allows one to view what the Linked Open Data Cloud "knows" about these species concepts.
http://www.taxonconcept.org/example-taxa/
Here are some example SPARQL queries that can be run live against the data in the LOD cloud
http://www.taxonconcept.org/example-sparql-queries/
I have a small set of sample occurrence records in this data set that are not currently in DarwinCore but I hope to modify in the near future.
This first blog entry provides a brief introduction to this particular conceptualization.
http://www.taxonconcept.org/taxonconcept-blog/2010/6/9/introductory-blog-ent...
http://www.taxonconcept.org/taxonconcept-blog/2010/6/9/introductory-blog-entry.htmlI think that I should have probably stated my initial question a little differently.
*Is there still a field that expect a URI and allows concepts like mine to be linked to species occurrence records?* * * *On first glance, it appeared as if this functionality had been removed.*
Respectfully,
- Pete
On Thu, Jun 10, 2010 at 2:37 PM, Peter DeVries pete.devries@gmail.comwrote:
I was looking through the latest DarwinCore and comments related to the TaxonConceptID.
Since the name serves as both a unique identifier and a phylogenetic hypothesis, you are effectively saying that observations labeled
*Aedes triseriatus*
and
*Ochlerotatus triseriatus*
Are separate species concepts, and should therefore be treated as separate things.
i.e. The name is the concept.
Also since there are several name variants for each "species", how do you distinguish which of these nameID's are the same species and which are different?
- Pete
Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 GeoSpecies Knowledge Base About the GeoSpecies Knowledge Base
participants (6)
-
David Remsen (GBIF) -
Hilmar Lapp -
Kevin Richards -
Mark Wilden -
Peter DeVries -
Richard Pyle