I can see how minting an additional term "dwcattributes:example" has a certain consistency given the namespace is already in use but, as a close alternative, and following best practice of not reinventing terms (and thus making it easier for 3rd parties trying to process the RDF directly), I think skos:example is also worth considering. I am not familiar with "iao:example of usage" but checking the definition [1]:
A phrase describing how a class name should be used. May also include other kinds of examples that facilitate immediate understanding of a class semantics, such as widely known prototypical subclasses or instances of the class. Although essential for high level terms, examples for low level terms (e.g., Affymetrix HU133 array) are not (definition looks truncated)
it refers to classes rather than properties. The majority of DwC terms are properties. So, can "iao:example of usage" be applied to them?
Éamonn
[1] http://purl.obolibrary.org/obo/IAO_0000112 http://purl.obolibrary.org/obo/IAO_0000112
From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Markus Döring Sent: 20 March 2015 14:35 To: John Wieczorek Cc: TDWG Content Mailing List Subject: Re: [tdwg-content] Darwin Core Proposal - term content recommendations to comments
Thanks John,
mostly repeating what I said earlier I would prefer to coin the new example term in the existing dwcattributes namespace.
It feels more consistent, is very straight forward and as you say we can always declare same as relations elsewhere if that is really useful to anyone.
I cannot see any immediate advantage of reusing either the skos or the iao term over minting a new one in the single dwcattributes namespace that helps documenting our dwc terms.
Markus
On 20 Mar 2015, at 14:14, John Wieczorek tuco@berkeley.edu wrote:
Dear all,
I would like to summarize my understanding of the state of the proposal to put term content recommendations in comments consistently, rather than in the definitions where they can sometimes be found. There has been universal support for this idea.
A second part of the proposal was to take this opportunity to separate the examples from the comments using a distinct property to do so in the rdf definitions of the terms. There has also been universal support for this idea.
Discussion revolved around how to accomplish the second part of the proposal. I originally proposed a new attribute "dwcattributes:example", which would be included once for every distinct example of term usage - with potentially many per Darwin Core term. There was universal support for the idea of enabling this capability, though there was mixed opinion about using an existing term instead of minting a new one in the dwcattributes namespace. Specifically, "skos:example" and "iao:example of usage" were discussed. Potential problems with "skos:example" were pointed out. Similar concerns about "iao:example of usage" were addressed, leaving this option still open for consideration.
It is not entirely clear technically (to me, anyway) how the annotation property "iao:example of usage" would be imported into the normative rdf document. I'm unaware of an OWL to RDF import mechanism. That doesn't mean much, because I am not an expert. If there is not, I suppose we would have to define an rdf file ourselves for the iao term and import that into dwcattributes or into the normative Darwin Core rdf file(s) directly. I don't think this would be a rigorous solution, but it would demonstrate our intent fairly well, especially if there were comments to that effect in the rdf files. Some guidance here might help us make a final decision on this topic.
If there is no clear and rigorous solution to the rdf import problem from IAO, then it may just be easier to coin "dwcattributes:example" as originally proposed, at least for now, and worry about it's equivalence to "iao:example of usage" on the ontology side (BCO).
I would really like to get the broader issue resolved soon, as we do already have consensus on making the contents of the definitions and comments consistent. If we can't resolve how to separate the examples (even though we agree that it is a good idea), I will propose that we forget that added part of the proposal and just deal with the consistency issue first.
Comments encouraged, as always.
Cheers,
John
On Thu, Feb 5, 2015 at 10:01 AM, Ramona Walls rlwalls2008@gmail.com wrote:
As far as I know, iao:example of usage also allows both literals and URLs as the range. However, I fail to see a serious problem with that, and to me, the benefits of re-using existing properties far outway the benefits I could see from having a separate set of properties for literals. Also, I don't know anyone who reasons over annotations properties, although I am sure there are those who do.
In response to Paul Morris's comment, simply using iao:example of usage does not import all of IAO, and therefore does not include any of the "baggage" of using IAO.
Ramona
------------------------------------------------------ Ramona L. Walls, Ph.D. Scientific Analyst, The iPlant Collaborative, University of Arizona Research Associate, Bio5 Institute, University of Arizona Laboratory Research Associate, New York Botanical Garden
On Thu, Feb 5, 2015 at 10:48 AM, Richard Pyle deepreef@bishopmuseum.org wrote:
Like others, I like the idea in principle, but Im not well-versed in the implications of alternate approaches to implementation to weigh in on that. Whatever technical solution is adopted, I would like to hope that it supports the representation of more than a single example; as sometimes it is useful to show alternate forms of acceptable content.
Aloha,
Rich
From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of John Wieczorek Sent: Thursday, February 05, 2015 7:22 AM To: Paul J. Morris Cc: TDWG Content Mailing List; Ramona Walls Subject: Re: [tdwg-content] Darwin Core Proposal - term content recommendations to comments
Does anyone have similar concerns about iao:example?
On Thu, Feb 5, 2015 at 2:09 PM, Paul J. Morris mole@morris.net wrote:
On Thu, 5 Feb 2015 11:17:58 -0500 Bob Morris morris.bob@gmail.com wrote:
But skos:note and its subproperties (including skos:example) can take literals or references [1]. To me, that weighs more than the baggage of minting two new terms.
Also, SKOS, unless care is taken to import the Owl-DL version, brings you into Owl-Full, with undesirable consequences for those who wish to do reasoning. In early versions of dwcFP, we did include SKOS terms, but removed them because of the consequences for reasoning.
SKOS has some nice terms, reuse is a nice idea, but it comes with significant knowledge engineering consequences.
-Paul -- Paul J. Morris Biodiversity Informatics Manager Harvard University Herbaria/Museum of Comparative Zoölogy mole@morris.net AA3SD PGP public key available
_______________________________________________ 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