Thanks so much for the comments and points. The Omeka team has been discussing these very things, and we appreciate your input. Let me respond to your points:

1) Experience tells us that individuals and small cultural institutions shy away from collection systems that emphasize esoteric schema. To balance their needs with the needs of advanced users, Omeka offers a generalized, easy-to-use schema that includes a permanent Dublin Core element set (the “flat fields”) that are common to all items, as well as customized item types to which users may assign customized metadata element sets (the “relational” metadata). This way, we feel, the average user will not be overwhelmed by strict metadata standards, while the more advanced users may customize their item types to whatever standards they wish.

2) We agree with you that this is a weakness, and we will be focusing more on localization in the future.

3) We have a ticket on Trac to add an Atom feed. And there is a RDF Dublin Core output that you may not be aware of (using ?output=dc). We intend to add more outputs as time allows.

We welcome any further comments or ideas. Please feel free to post on our forums:
http://omeka.org/forums/

Best,
Jim Safley
Omeka Developer
Center for History and New Media

@Bruce: You should raise these issues on the Omeka forums (though I’ll of course forward them to the Omeka team). The Zotero and Omeka teams are completely distinct, and so there perhaps has not been enough cross-pollination. But these seem like things the Omeka team can address quickly.