It would be nice to browse databases directly from Zotero using the z39.50 protocol just like some commercial software - if you see what I mean. Great Job : I love Zotero.
z39.50 implementation may happen (though I'm skeptical), but it would require a substantial addition to Zotero, since it's a search&retrieve standard and Zotero doesn't provide for direct querying of databases. In many ways, the whole point of integrating Zotero into a browser was to make going through z39.50 unnecessary. Instead of querying a library or pubmed or whatever through z39.50 you can just use the catalog and the URL bar importer.
Having seen this at work in Endnote, I'm not sure I find the speed/efficiency gains very impressive and worth the significant time this would take to add (and Zotero would, if I understand that correctly, still need to add a specific filter for every database it wants to query via the standard).
If I write a z39.50 translator, we could use it to essentially add any major library or database to the current lookup system, which only uses OpenSearch at present (and OpenSearch is not a major standard for the library world).
right - but no search results, which I thought was what people really wanted. But talking about library standards
Have you seen this?
http://www.loc.gov/standards/sru/
I have seen SRU. I think that z39.50 and SRU should both be explored as possible future translators... And then, presumably, we would build a plugin (later to be merged into the core) that let people send an arbitrary query to a number of enabled lookup services. I imagine this as something implemented on a chrome:// URL, like the CSL test pane or the translator test status screen.
z39.50 implementation may happen (though I'm skeptical), but it would require a substantial addition to Zotero, since it's a search&retrieve standard and Zotero doesn't provide for direct querying of databases. In many ways, the whole point of integrating Zotero into a browser was to make going through z39.50 unnecessary.
This.
I don't see the point. And adding it would require new UI and backend code; not just translators.
Having just recalled that Z39.50 isn't traditionally built on standard HTTP, I retract my statement that this could be done in translators alone. SRU might still be a valid target.
Upgrade Storage
A searching option in z39.50, as an importing using UNIMARC, would be important function for Zotero.
I hope for an implementation.
Having seen this at work in Endnote, I'm not sure I find the speed/efficiency gains very impressive and worth the significant time this would take to add (and Zotero would, if I understand that correctly, still need to add a specific filter for every database it wants to query via the standard).
Have you seen this?
http://www.loc.gov/standards/sru/
z39.50 gives us MARC21, which we are pretty good at reading already, and SRU gives plenty of interesting data formats too (http://www.loc.gov/standards/sru/resources/schemas.html).
I don't see the point. And adding it would require new UI and backend code; not just translators.