unAPI formats (RDF vs RIS)

Zev.Goldstein

So I'm trying to make this website work with Zotero. library.case.edu/digitalcase
I've already set up an unAPI server, and I'm trying to determine which format to use. Our data is naively stored as oai_dc, but i'm also going to make it compatible with endnote, so i will have RIS available as well.

It looked like oai_dc and rdf_dc are similar enough that converting might be pretty simple, and it seems like Zotero can do more with RDF than it can with RIS.

I guess I want to know what are the benefits of using RDF over RIS, and are they worth the extra effort of converting oai_dc to rdf_dc.

noksagt

UnAPI allows you to serve both formats. The best description for the different fields the formats accept is the translator code.

Zev.Goldstein

I do know i can serve both. I'm serving RIS regardless. I want to know, why do they recommend MODS and RDF over RIS in the first place. Is it worth the effort to serve both or not.

As for your second sentence, I don't really understand it. Aren't the descriptions of the fields the formats accept in their specs? What do you mean by translator code?

noksagt

I want to know, why do they recommend MODS and RDF over RIS in the first place.

Because they're richer, hierarchical formats. You can express things in MODS and RDF that you simply cannot in RIS

Is it worth the effort to serve both or not.

The effort isn't really all that great, but are you happy with RIS or not? If there's no piece of data that you want to give to clients that can't be transmitted via RIS, perhaps it is good enough. Or perhaps you may think that it is important enough to transmit things like the language the resource is in, which is simply not possible using RIS.

What do you mean by translator code?

https://www.zotero.org/trac/browser/extension/trunk/translators/RDF.js
https://www.zotero.org/trac/browser/extension/trunk/translators/RIS.js

Zev.Goldstein

Thank you. That is all really helpful. :)