Thanks for all of the responses to my original question. They have given me much food for thought. I will throw out several comments that have come to mind as a result.
1. With regards to what Dean and Tony have said about the kind of "linkage" that taxonomists want:
My hypothesis is that taxonomists (and taxonomy) "want" linkage like this: taxon name ==> correct authority information ==> correct bibliographic metadata ==> actual original publication (full-text)
Although a linkage to a full text version of the publication would be nice, it's not really what I wanted/needed when I posed the original question. As a non-taxonomist, I've absorbed the point which has been made repeatedly that in an Identification which simply provides a taxon name is inadequate and that we should be specifying a taxon name usage (i.e. name sensu publication, or name sec. publication). So I'm more interested in a globally unique and reusable identifiers for names, publications, and their intersections as TNUs. As such, I don't care that much whether there is a linkage to a pdf of the publication as long as I can refer to the TNU by a GUID and somebody else knows exactly what taxon usage I'm talking about. I don't have a high degree of confidence in using text strings like "Andropogon virginicus L. sec. Radford et al. (1968)" as identifiers for TNUs because all it takes is one missing period, no parentheses, or an extra space and then I have a different string from somebody else. If the GUID is going to be a URI, then I'd really like it to be dereferenceable to RDF. In the event that the Linked Data world comes together, that gets me out of the business of thrashing with all of the taxonomy stuff that I'm not interested in doing myself. I believe that this is exactly one of the points that Pete was trying to make.
From this standpoint, what Paul illustrated in his example of http://biodiversity.org.au/apni.taxon/118883 is exactly what I had in mind: a URI for the taxon/TNU/concept with RDF links to the URI for the name and the URI for the publication. The "fundamental problem" (recognized by Pete with his asterisk) is that most of these URIs don't yet exist and it would be counterproductive for a bunch of different people to start "minting" them on their own. I certainly don't have the interest or ability to do it and I doubt that Paul has time to create them all for the rest of us on the planet at biodiversity.org.au . This should be large scale/community effort. I was disappointed to see that although http://citebank.org seems to be positioning itself as such a large-scale effort, I can't see any evidence that it is planning to create "Linked Data-ready" URIs that are subject to content negotiation (or did I just miss it?). I think that is probably a mistake. Making a GUID that could be used in the LOD world doesn't force anybody to subscribe to the LOD model, but making a GUID that is not suitable for LOD will cause those who are interested in Linked Data to look elsewhere.
2. On the subject of DOIs, I discovered what Paul mentioned in his post: there doesn't seem to be any easy way to search for a DOI or to know if a DOI exists for a publication. Another problem is who would pay for the DOIs that would need to be created for all of the obscure, out of print publications that would need to be put into the system? (I believe that there is some per-item cost for DOIs, right?) Generic HTTP URIs seem like an easier solution.
3. Another issue which I think should be mentioned in the context of this discussion is that I don't think that it is a good idea to blur the distinction between a URL pointing to an information resource (such as a pdf or jpg file) and a URI that is serving as a GUID. If an information resource URL were considered a GUID, then I believe that it would be bad practice to ever change even a single byte of the particular file to which that URL is pointing. It would also be difficult to achieve content negotiation (i.e. to provide either a human-readable representation or a machine-readable RDF file when requested) if a specific file format is specified in the URL. So even if there were a PDF available for a publication online, the URL of that PDF shouldn't be used as a GUID. It would be better to represent the publication as a non-information resource and use a seeAlso link to the pdf representation. I think what I've said here has bearing on Dean's message (directly below). In particular I think that the statement
"a GUID for the curated, corrected bibliographic metadata as something distinct from a GUID for the actual publication."
makes an artificial distinction between a publication and its metadata. I think the correct approach is to consider that the GUID represents the publication as a conceptual entity (i.e. non-information resource) which has properties that can be described by the bibilographic metadata, and which can have a representation as a pdf, jpg image, etc. With that approach, the GUID can be an unchanging identifier for the publication while the metadata and representations associated with it can be free to be corrected, updated, or improved.
Steve
From: tdwg-content-bounces@lists.tdwg.org [tdwg-content-bounces@lists.tdwg.org] On Behalf Of Dean Pentcheff [pentcheff@gmail.com] Sent: Wednesday, 5 January 2011 10:02 AM To: tdwg-content@lists.tdwg.org Cc: Chris Freeland Subject: Re: [tdwg-content] GUIDs for publications (usages and names)
A wrinkle on this is that the authority information (name and year) are not the same thing as the actual original publication. Nor is the bibliographic metadata for a publication the same thing as the publication.
My hypothesis is that taxonomists (and taxonomy) "want" linkage like this: taxon name ==> correct authority information ==> correct bibliographic metadata ==> actual original publication (full-text)
In the general case, you can't determine the authority date by inspection of the actual original publication. You often need external information regarding publication mechanics. (Yes, I know that the date printed on the publication is the correct date of publication for most papers, but anyone who's dealt with taxonomic literature knows that there are many, many exceptions to that.)
I'm not sure that they actually want: taxon name ==> actual original publication (full-text)
Which is why I start thinking about a GUID for the curated, corrected bibliographic metadata as something distinct from a GUID for the actual publication.
-Dean
Dean Pentcheff pentcheff@gmail.com dpentche@nhm.org
On Tue, Jan 4, 2011 at 1:01 PM, Peter DeVries pete.devries@gmail.com wrote:
Hi Chris, You are missing one of the main benefits of Linked Data. Users do not want to curate their own bibliographic databases and related RDF, they want to simply link to a globally unique, resolvable identifier for that citation. For example: Felis concolor Linnaeus 1771 Linnaeus 1771 <= What specific publication is this? "Felis concolor" hasOriginalDescription http://www.biodiversitylibrary.org/citation/234123412 If the BHL exposed semantic web URI's for each publication, then content experts could search the BHL, find it, and then link to it with a simple resolvable URI. The alternative is that Steve, et al., will create their own identifiers for specific publications, code their own local bibliographic databases. In essence, each group or individual then continues to duplicate the efforts of others. Right now, I have duplicated your functionality in my species concepts. Ideally, I would have done this: <SpeciesConcept> hasOriginalDescription http://www.biodiversitylibrary.org/citation/234123412
<SpeciesConcept> hasRevisionalDescription <http://www.biodiversitylibrary.org/citation/234124356> Not only are my concepts then linked to your citations, any other data sets that link to your citations are findable. For instance, who else has data sets that link to this citation? To see how this works on the live LOD cloud check out. http://bit.ly/fChHwJ Browse through the related Linked Data Sets - in particular the has close match links. - Pete
On Tue, Jan 4, 2011 at 2:31 PM, Chris Freeland Chris.Freeland@mobot.org wrote:
Ooops, to clarify my very last example, we actually would support "Pallas 1767" if properly parsed:
http://www.biodiversitylibrary.org/openurl?aulast=Pallas&date=1767&f...
What we don't yet support & need to is linking at article citation level. That's where our newly (quietly) launched CiteBank http://citebank.org comes in, and what we're hoping to receive funding to expand.
Chris ________________________________ From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Chris Freeland Sent: Tuesday, January 04, 2011 1:57 PM To: Peter DeVries; Steve Baskauf Cc: tdwg-content@lists.tdwg.org Subject: Re: [tdwg-content] GUIDs for publications (usages and names)
Steve, Pete, et al.,
BHL has an OpenURL resolver that can accept a variety of input criteria & return matching records with responses in JSON (with or without callback), XML, HTML, or a direct link. Documentation is here: http://www.biodiversitylibrary.org/openurlhelp.aspx
And linked from our broader documentation here: http://biodivlib.wikispaces.com/Developer+Tools+and+API Here's an example, referenced in the documentation, for querying on a monograph/book:
http://www.biodiversitylibrary.org/openurl?genre=book&title=Manual+of+No...
You can also query based on common abbreviations, like Sp. Pl.:
http://www.biodiversitylibrary.org/openurl?stitle=Sp.%20Pl.&date=1753&am...
MOBOT's Tropicos uses the OpenURL resolver to link to protologues, as in this example: http://www.tropicos.org/Name/2735114
With Tropicos we have an authority record for each journal or monographic title. We match Tropicos' TitleID to BHL's TitleID & use that as a more direct link to the appropriate reference, but still send collation & other info to get to the appropriate page, as in this link:
http://www.biodiversitylibrary.org/openurl?pid=title:626&volume=5&is...
I know that this is insufficient for zoology & other natural sciences beyond botany, where we need to be able to support citations like "Pallas 1767", which may or may not be preparsed into appropriate fields. A known problem, for sure, and one that we're eager to address, pending funding from NSF.
Chris
Chris Freeland | Director, Center for Biodiversity Informatics | Missouri Botanical Garden 4344 Shaw Blvd. | St. Louis, Missouri 63110 | 314.577.9548
From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Peter DeVries Sent: Tuesday, January 04, 2011 1:33 PM To: Steve Baskauf Cc: tdwg-content@lists.tdwg.org Subject: Re: [tdwg-content] GUIDs for publications (usages and names)
Hi Steve, I have been lobbying the BHL for this for some time. Most recently in this blog post.
http://www.taxonconcept.org/taxonconcept-blog/2010/8/5/why-linked-open-data-...
What I have realized is that, for many works that are out of copyright, Google books has already scanned and converted them to PDF. You can use these, even extract the relevant PDF pages as long as you keep the Google watermark. This is what I have done for the Cougar. http://lod.taxonconcept.org/ses/v6n7p.html
For more recent works you may be able to link to the article PDF. As in this spider example. http://lod.taxonconcept.org/ses/2mqjL.html
We also need a URI for to uniquely identify authors and in the absence of a better solution, I have been using, and in some cases creating, entries in Wikipedia which currently has over 5,000 taxonomic author profiles. This results in usable RDF via DBpedia. As you can see in the RDF in this example. http://lod.taxonconcept.org/ses/v6n7p.rdf
And in the Knowledge Base
<http://lsd.taxonconcept.org/describe/?url=http%3A%2F%2Flod.taxonconcept.org%...
That said, I have also been experimenting with this. http://lod.taxonconcept.org/people/sci_people_1700.rdf
About: Carl Linnaeus http://bit.ly/gLgElf
- Pete
On Mon, Jan 3, 2011 at 7:47 PM, Steve Baskauf steve.baskauf@vanderbilt.edu wrote:
I was reviewing some of the previous posts on taxon name usages in an attempt to understand them better. I have learned that the Global Names Index is an attempt to catalog taxon names and that it is possible to generate a URI that points to a name there. Is there a parallel effort to do the same thing for literature references? In other words, if I want to describe the TNU: Andropogon virginicus L. sec. Radford et al. (1968) I think I could find a URI GUID for the name Andropogon virginicus . But is there some place where I could find a unique identifier, or better a URL, or best a URI providing RDF/XML for Linnaeus 1753 (the author and publication for the name) and for Radford et al.1968 (the author and publication that expresses the usage I'm intending).
I suppose that this question has previously been answered in the in the many posts on taxon names, concepts, etc. However, since usually my brain goes numb and my eyes glaze over in those threads, I probably missed it.
Steve
-- Steven J. Baskauf, Ph.D., Senior Lecturer Vanderbilt University Dept. of Biological Sciences
postal mail address: VU Station B 351634 Nashville, TN 37235-1634, U.S.A.
delivery address: 2125 Stevenson Center 1161 21st Ave., S. Nashville, TN 37235
office: 2128 Stevenson Center phone: (615) 343-4582, fax: (615) 343-6707 http://bioimages.vanderbilt.edu
tdwg-content mailing list tdwg-content@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-content
--
Pete DeVries Department of Entomology University of Wisconsin - Madison 445 Russell Laboratories 1630 Linden Drive Madison, WI 53706 TaxonConcept Knowledge Base / 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 TaxonConcept Knowledge Base / 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
tdwg-content mailing list tdwg-content@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-content _______________________________________________ tdwg-content mailing list tdwg-content@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-content .