Has there been any talk about more readily connecting Omeka collections to the linked open data movement/stable uri's etc? It seems the two could be a match made in heaven. I'm currently thinking about how I can incorporate linked data into my collection. I would think a plugin could facilitate this?
Linked Open Data is very much on our minds, especially given the use of Omeka by Europeana and the brand new Digital Public Library of America. It is in our long term plans, but we have not yet created a strategy for how to implement it.
In the meanwhile, working on a plugin might help us see the needs and wants of the Omeka community for LoD.
Thanks for the quick reply. Glad to hear Omeka thinking about LoD. Here is +1 vote for Omeka developing an LoD plugin or something.
Has there been any advance in the thinking about LoD (or even plugin development)? Omeka n00b here, so sorry if I am missing anything.
thanks in advance
I am reasonably new to Omeka too, but the Item Relations plugin takes care of intra-Omeka relationships via proper ontologies. I hope others can shed more light on this, and maybe comment on plans at the Omeka team but here is my summary - I could be wrong about some of this as I am very new.
You can link items internally using a decent set of ontologies via Item Relations. You can't search interactively for items to which you want to link, you need to know an item ID.
You can't pick a label for the relationship, eg being able to link a creator to a text and use a particular variant of their name.
It doesn't support external relationships, except (as someone noted elsewhere in the forums) by creating a local proxy for the external URI, eg in a 'links' collection. This is not necessarily a huge limitation and may actually be good practice, it would be nice to have good UI support for it though, so people can enter external relations easily.
And as far as I can tell, there is no way to import relationships via CSV, or add them via the API. I'm very interested in attacking this issue as I have a project where all the data is in available in several CSVs (people, texts, places etc), with cross-references via IDs. (One hack I have thought of is to put HTML links in the cross reference fields.)
I hope others can shed more light, and maybe tell us about plans for Item Relations and external relationships