Recognition of artwork page

phlit · December 1, 2011

Hello,
I would like to know how to make a page presenting a photo recognized as an Artwork item (i.e. Document type : illustration). As an example, this is what is done on Flickr, but I can figure out what is presented on these pages, that make them recognized as such.

I would appreciate help on that point !

Thanks !

ajlyon · December 1, 2011

Flickr has a dedicated translator. If you have a page that you'd like to enable the same feature for, see the exposing metadata documentation.

phlit · December 1, 2011

Hello, and thank you for your answer. Do you mean i'll need to write my own translator ?

adamsmith · December 1, 2011

either that, or follow the exposing metadata link - unAPI would work, too, for example.

phlit · December 7, 2011

Thanks for your replies. So for example using Dublin Core "type" element :
<meta name='DC.Type' content="Artwork"/>
only affects the "website type" field for Zotero.
So I guess I should be using another DC field, or more generally another RDF vocabulary. Am I right ? Any idea of which rdf to use ?

Thanks again !

ajlyon · December 7, 2011

No, the RDF type should be http://vocab.org/biblio/schema#term-Illustration, but that can't be expressed using the basic meta tag approach. I recommend using RDF or BibTeX with unAPI; that means you'd need to create the metadata file, put it on the server, and reference it using the unAPI instructions.

phlit · December 7, 2011

Thank for your reply. So that means i need to create a metadata file for each of the illustration (and for other item types as well) i have, right ?

ajlyon · December 7, 2011

For some item types, you can use COinS, but that doesn't support the artwork type, which is why I directed you to the unAPI approach, which is the most flexible at present.

If you are using a content management system on the server, it should be possible to have it generate the BibTeX / RDF on request and insert the unAPI code in the item pages. If these are manually generated pages, you'll need to do that more or less manually.

phlit · December 8, 2011

Thank you for your answer.
So i tried to implement an unAPI server. Here are how it show formats availables :
<?xml version="1.0"?>
<formats>
<format name="rdf" type="application/xml" docs="http://www.w3.org/1999/02/22-rdf-syntax-ns#"/>
</formats>

And the result for a given image :
<?xml version="1.0"?>
<rdf:RDF
xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns:res="http://purl.org/vocab/resourcelist/schema#"
xmlns:z="http://www.zotero.org/namespaces/export#"
xmlns:ctag="http://commontag.org/ns#"
xmlns:dcterms="http://purl.org/dc/terms/"
xmlns:bilbio="http://purl.org/net/biblio#"
xmlns:bibo="http://purl.org/ontology/bibo/"
xmlns:foaf="http://xmlns.com/foaf/0.1/">

<bibo:Image rdf:about="http://phlit.org/items/show/2755">
<dcterms:title>« Corsets, boulevard de Strasbourg, 1912 »</dcterms:title>
<dcterms:date>1912</dcterms:date>
<dcterms:creator>ATGET, Eugène</dcterms:creator>
<bibo:uri>http://phlit.org/items/show/2755</bibo:uri>;
<dcterms:creator rdf:nodeID="n1"/>

</bibo:Image>

</rdf:RDF>

But Zotero still doesn't see anything here.
Any idea of what could be wrong ?

Thanks !

ajlyon · December 8, 2011

What code are you putting into the page? You need to indicate the unAPI server, and provide the id in the text. Example from the WordPress unAPI implementation below.

In the header:
<link rel="unapi-server" type="application/xml" title="unAPI" href="http://www.gimranov.com/wp/wp-content/plugins/unapi/server.php"/>
In the text, for each item on the page:
<abbr class="unapi-id" title="http://www.gimranov.com/wp/?p=1061"></abbr>

phlit · December 8, 2011

If you take a look at this item http://phlit.org/items/show/2755

you can see the unapi server address in header :

<link rel="unapi-server" type="application/xml" title="unAPI" href="http://phlit.org/items/browse/output/dcmes-xml/">

and id is given as
<abbr class="unapi-id" title="2755"></abbr>

Thanks for your help :)

ajlyon · December 8, 2011

Zotero understands the following formats:
"rdf_zotero", "rdf_bibliontology", "mods", "marc", "unimarc", "ris", "refer", "bibtex", "rdf_dc"

Set the unAPI format list response to send "rdf_bibliontology" and it should work.

ajlyon · December 8, 2011

But I thought that Omeka had its own robust Zotero metadata exposing feature... As another CHNM project, that would be pretty logical.

adamsmith · December 8, 2011

I don't know much about this, but if I read the unAPI translator correctly, you need to specify the type of RDF - i.e. in this case you'd want format name set to rdf_dc
From that translator:
var RECOGNIZABLE_FORMATS = ["rdf_zotero", "rdf_bibliontology", "mods", "marc", "unimarc", "ris", "refer", "bibtex", "rdf_dc"];

phlit · December 8, 2011

Well, you mentionned unAPI protocol for illustration item types, but I haven't seen any easiest way to do that.

Anyway, thank you very much : now it works !

adamsmith · December 8, 2011

I couldn't see anything on Omeka, no - they have a COinS plugin, which probably wouldn't do here.

ajlyon · December 8, 2011

No, COinS wouldn't do it. If your code is pretty easy to generalize, I'm sure the Omeka folks would be happy to see it posted somewhere for others to use when they need a solution more robust than COinS.

adamsmith · December 8, 2011

nice site, by the way!

phlit · December 8, 2011

Thanks for your comment. I'll consider writing sort of a recipe for Omeka.