- From: Jeff Mixter <jeffmixter@gmail.com>
- Date: Thu, 9 May 2013 12:44:50 -0400
- To: Dan Brickley <danbri@danbri.org>
- Cc: lazarus@lazaruscorporation.co.uk, Guha <guha@google.com>, W3C Vocabularies <public-vocabs@w3.org>
- Message-ID: <CAC=429BH-ZxE0goQqeZRmJO4zhtKQmnpg0Ooi2m6hD8faH0gWQ@mail.gmail.com>
Dan, (I know this is a repeat for you but I forgot to reply to all with the previous email) The example that you put together is really interesting. I just wrapped up work on my masters thesis that more or less dealt with this exact problem. In my study, I developed a new proposed data model for VRA 4 that incorporated Schema.org as much as possible. Like you, I had trouble when it came to certain aspects, such as saying something is a Sculpture. More problematic was mapping over existing XML measurement attributes such as depth, area etc. (this was also a problem in other areas of the VRA data model). To overcome the problem, I had to create custom "vra-p" classes/properties but I positioned them as sub-classes or sub-properties of existing schema elements. For example I created a custom vra-p:area object property that has a given range of schema:quantitativeValue. My thinking was that search engines might not care about my custom VRA element but they would care that it connects a schema:CreativeWork with a schema:quantitativeValue. Obviously my way of overcoming the issue is not ideal, but I think it serves as a nice stop-gap. I do whole heartily agree with your suggestions though and as Jeff Young and others have commented the combination of schema and productontology will help resolve many of these issues. If you are interested in seeing some other examples of how I worked around mapping the relatively complex VRA Core 4 model to schema.org feel free to check out my website for the project. The thesis as well as the tools used/developed during the study can be accessed and downloaded. I would welcome any comments or suggestions. http:purl.org/jmixter/thesis/ <http://purl.org/jmixter/thesis/> Thanks, Jeff Mixter 440-773-9079 On Thu, May 9, 2013 at 11:12 AM, Dan Brickley <danbri@danbri.org> wrote: > On 7 May 2013 23:43, Guha <guha@google.com> wrote: > > I agree. This is a good idea and a simple addition. > > > > If there aren't objections, we will include it in the next draft. > > http://www.w3.org/wiki/WebSchemas/VisualArtwork > > Ok, I have transcribed from the Wiki page into a candidate RDFa+RDFS > config for schema.org. It is pretty straightforward. Note that > width/height/depth already exist; we are here just attaching another > expected type to the list of included domain types. And a reminder > that 'domain' and 'range' in schema.org are weaker than RDFS's domain > and range; we'll rename/alias them to rangeIncludes / domainIncludes > at some point to clarify this. > > The draft lives in the W3C WebSchemas repository, > > https://dvcs.w3.org/hg/webschema/file/default/schema.org/ext/visualartwork.html > > Mercurial reminder, the 'raw' link shows you HTML, > > https://dvcs.w3.org/hg/webschema/raw-file/default/schema.org/ext/visualartwork.html > ... but for these files, the detail is best viewed in the raw HTML > source. > > Feeding that URL to Ivan Herman's RDFa 1.1 distiller - > http://www.w3.org/2012/pyRdfa/ - gives the Turtle summary below; seems > readable. > > A couple of points, by which as usual I mean at least 3: > > (1.) I didn't see a short description for the type itself, so I > invented one - 'A work of art that is primarily visual in character.' > (2.) It would be good to find wording (and/or hierarchical) that > acknowledges that some visual arts can be tactile too; e.g. that blind > people can appreciate http://schema.org/Sculpture ... hence the > 'primarily' here. Suggestions anyone? Would we say Sculpture is a > sub-type of VisualArtwork ? > (3.) I have omitted 'edition' for now, partly because it seems > something that ought to be handled for creative works, journals, > comics etc consistently; but mostly because I didn't quite understand > "For multiples such as prints, the number of copies in the edition" > (did someone mention > > http://en.wikipedia.org/wiki/Functional_Requirements_for_Bibliographic_Records > ?) > (4.) This transcription might not do complete justice to the various > points raised in the rest of this thread; but hopefully it helps firm > things up. > > Dan > > > > @prefix dc: <http://purl.org/dc/terms/> . > @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> . > @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> . > @prefix schema: <http://schema.org/> . > > schema:VisualArtwork a rdfs:Class; > rdfs:label "VisualArtwork"; > dc:source < > http://www.w3.org/wiki/WebSchemas/SchemaDotOrgSources#source_VisualArtworkClass > >; > rdfs:comment "A work of art that is primarily visual in character."; > rdfs:subClassOf schema:CreativeWork . > > schema:artform a rdf:Property; > rdfs:label "artform"; > schema:domain schema:VisualArtwork; > schema:range schema:Text; > rdfs:comment "e.g. Painting, Drawing, Scupture, Print, Photograph, > Assemblage, Collage, etc." . > > schema:depth a rdf:Property; > schema:domain schema:VisualArtwork . > > schema:height a rdf:Property; > schema:domain schema:VisualArtwork . > > schema:materials a rdf:Property; > rdfs:label "materials"; > schema:domain schema:VisualArtwork; > schema:range schema:Text; > rdfs:comment "e.g. Oil, Watercolour, Acrylic, Linoprint, Marble, > Cyanotype, Digital, Lithograph, DryPoint, Intaglio, Pastel, Woodcut, > Pencil, Mixed Media, etc." . > > schema:surface a rdf:Property; > rdfs:label "surface"; > schema:domain schema:VisualArtwork; > schema:range schema:Text; > rdfs:comment "e.g. Canvas, Paper, Wood, Board, etc." . > > schema:width a rdf:Property; > schema:domain schema:VisualArtwork . > > -- Jeff Mixter jeffmixter@gmail.com 440-773-9079
Received on Thursday, 9 May 2013 20:22:52 UTC