- From: Corey Harper <corey.harper@gmail.com>
- Date: Mon, 21 Jan 2013 10:52:27 -0500
- To: "Young,Jeff (OR)" <jyoung@oclc.org>
- Cc: Adrian Pohl <pohl@hbz-nrw.de>, public-schemabibex@w3.org
This is the point I was trying to make before, though I guess I didn't manage to convey it well. I also agree with you about string identifiers, but the fact is: many of them remain, and many organizations are terrified of minting URIs for these things. I think we're going to have to build out infrastructure that supports this as well. I also don't see why the same pattern, based on the medical codes example, cannot be used for both string-based & for URI-based identifiers... Thanks, -Corey On Mon, Jan 21, 2013 at 10:19 AM, Young,Jeff (OR) <jyoung@oclc.org> wrote: > I agree that the pattern already exists and should be generalized instead of reinvented. The essential pattern is ultimately SKOS. Schema.org is making a mess by not realizing it. > > "String" identifiers are buggy whips. > > Jeff > >> -----Original Message----- >> From: Adrian Pohl [mailto:pohl@hbz-nrw.de] >> Sent: Monday, January 21, 2013 4:59 AM >> To: public-schemabibex@w3.org >> Subject: Re: Re: alternate identifier draf >> >> Coming back to discussing the proposal itself: First, I wonder, why one >> needs a new property called schema:issuedBy. Why not use >> schema:publisher? >> >> This would mean only adding three new properties and one new class to >> schema.org: >> >> - schema:Identifier as the class of all identifiers >> - schema:identifier as the property which links something to an >> identifier >> - schema:inStandard to indicate where the identifier comes from. >> >> But why add anything new at all? >> >> As I said before, this design is very close to the practice of >> expressing a medical code for a disease or drug with schema.org. [1] >> Instead of schema:identifier they use schema:code and instead of >> schema:inStandard schema:codingSystem. I encourage re-using these >> properties or developing a general approach for both use cases instead >> of coming up with a new but very similar proposal. >> >> One might argue that a controlled vocabulary like MeSH is quite >> different from a set of identifiers issued and that these should be >> made clear in schema.org. I'd say for the high-level markup schema.org >> provides, these can be treated as equivalent as some body issues >> identifiers for some things - may these things be publications or >> drugs, diseases etc. What type of thing the identifier belongs to (e.g. >> a schema:Drug or a schema:Book) is clear from the context, anyway. >> >> - Adrian >> >> [1] >> http://www.w3.org/community/schemabibex/wiki/Identifier#Adopting_approa >> ch_from_medical.2Fhealth_extension >> >> >>> On 20.1.2013 at 22:52, Niklas Lindström<lindstream@gmail.com> >> wrote: >> > Nice! I just fixed some minor HTML syntax issues, and also changed >> the >> > full URL in the first @typeof to just the term "Book" (which, like >> the >> > second @typeof="Identifier", is bound to the given >> > @vocab="http://schema.org/"). >> > >> > (Also verified the output with the latest RDFLib.) >> > >> > Cheers, >> > Niklas >> > >> > On Sun, Jan 20, 2013 at 10:31 PM, Adrian Pohl <pohl@hbz-nrw.de> >> wrote: >> >> I started to work on your Notation3 example before having read this >> thread >> > to the end. While being at it I put Ivan's example into the wiki. [1] >> But I >> > left out the schema:publisher truple as this wasn't part of the >> original >> > example and doesn't make sense. >> >> >> >> >> >> - Adrian >> >> >> >> >> >> [1] http://www.w3.org/community/schemabibex/wiki/Identifier-2 >> >> >> >>>>> Karen Coyle <kcoyle@kcoyle.net> 20.01.13 16.40 Uhr >>> >> >> No surprise that my examples suck :-) Should I substitute these for >> >> mine? That's fine with me. >> >> >> >> kc >> >> >> >> On 1/19/13 11:19 PM, Ivan Herman wrote: >> >>> Karen, >> >>> >> >>> the example is not RDFa but microdata... Which is per se o.k., but >> the >> > microdata->RDF conversion will not create a blank node when it sees a >> >> > @itemprop="identifier/name" construct. Nor will RDFa, for that >> matter. The >> > result with RDFa will be something like (I just mechanically created >> an RDFa >> > file and ran it through my converter): >> >>> >> >>> @prefix ns1: <http://schema.org/identifier/> . >> >>> @prefix schema: <http://schema.org/> . >> >>> >> >>> [] a schema:Book; >> >>> schema:author "Claude-Michel Schonberg, Alain Boublil, and >> Herbert >> > Kretzmer."; >> >>> schema:bookFormat schema:Paperback; >> >>> schema:by "Hal Leonard "; >> >>> schema:description "For Piano/Vocal. Vocal Selections. "; >> >>> ns1:issued "Hal Leonard "; >> >>> ns1:name "HL.114335"; >> >>> schema:name "Les Miserables. (Selections from the Movie)."; >> >>> schema:numberOfPages " 88 pages."; >> >>> schema:publisher " Published by Hal Leonard " . >> >>> >> >>> which is not what you would expect:-( To do it right in HTML+RDFa >> (but, I >> > would think, in microdata as well, it is just that I am more 'fluent' >> in >> > RDFa) one has to something like >> >>> >> >>> <html><body> >> >>> <div vocab="http://schema.org/" typeof="http://schema.org/Book"> >> >>> <span property="name">Les Miserables. (Selections from the >> Movie).</span> >> >>> by <span property="author" >Claude-Michel Schonberg, >> Alain Boublil, >> > and Herbert Kretzmer.</span> >> >>> <span property="description">For Piano/Vocal. Vocal >> Selections. </span> >> >>> <l!nk property="bookFormat" >> > href="http://schema.org/Paperback">Softcover</a>. >> >>> <span property="numberOfPages"> 88 pages.</span> >> >>> <span property="publisher"><span property="identifier" >> > typeof="Identifier"><span property="issuedBy">Hal Leonard</span>, >> (<span >> > property="name">HL.114335</span>)</span></span> >> >>> </body></html> >> >>> >> >>> yielding >> >>> >> >>> [] a schema:Book; >> >>> schema:author "Claude-Michel Schonberg, Alain Boublil, and >> Herbert >> > Kretzmer."; >> >>> schema:bookFormat schema:Paperback; >> >>> schema:description "For Piano/Vocal. Vocal Selections. "; >> >>> schema:identifier [ a schema:Identifier; >> >>> schema:issuedBy "Hal Leonard"; >> >>> schema:name "HL.114335" ]; >> >>> schema:name "Les Miserables. (Selections from the Movie)."; >> >>> schema:numberOfPages " 88 pages."; >> >>> schema:publisher "Hal Leonard, (HL.114335)" . >> >>> >> >>> >> >>> Which is not exactly the same, b.t.w., because the publisher field >> will also >> > include the identifier stuff... >> >>> >> >>> Sorry...:-) >> >>> >> >>> Ivan >> >>> >> >>> >> >>> >> >>> On Jan 20, 2013, at 24:44 , Karen Coyle <kcoyle@kcoyle.net> wrote: >> >>> >> >>>> I have added a second identifier proposal [1], much simpler than >> the first >> > (and possibly too simple -- it is a first draft). >> >>>> >> >>>> kc >> >>>> [1] http://www.w3.org/community/schemabibex/wiki/Identifier-2 >> >>>> -- >> >>>> Karen Coyle >> >>>> kcoyle@kcoyle.net http://kcoyle.net >> >>>> ph: 1-510-540-7596 >> >>>> m: 1-510-435-8234 >> >>>> skype: kcoylenet >> >>>> >> >>> >> >>> >> >>> ---- >> >>> Ivan Herman, W3C Semantic Web Activity Lead >> >>> Home: http://www.w3.org/People/Ivan/ >> >>> mobile: +31-641044153 >> >>> FOAF: http://www.ivan-herman.net/foaf.rdf >> >>> >> >>> >> >>> >> >>> >> >>> >> >> >> >> -- >> >> Karen Coyle >> >> kcoyle@kcoyle.net http://kcoyle.net >> >> ph: 1-510-540-7596 >> >> m: 1-510-435-8234 >> >> skype: kcoylenet >> >> >> >> >> >> >> >> >> >> >> >> >
Received on Monday, 21 January 2013 15:52:59 UTC