Example of integration of BibDesk, Skim and DokuWiki

I realize this isn't directly relevant to Zotero, but I thought you might be interesting in this - we also share the same ideas about open source, open sharing, semantic data etc! I investigated a bunch of citation mgmt systems, and ended up with BibDesk. One of the reasons I didn't use Zotero (which I like a lot), is that there isn't an easy way for outside programs to interface with the database right now (I know it's _possible_, but not easy). Compared to BibDesk, whose canonical database is a bibdesk file in a very widely recognized format (with libraries for almost all languages to parse, etc).

I wrote a bunch of small Ruby/AppleScripts to tie together BibDesk, Skim, my Kindle (automatic export and import), and DokuWiki in Chrome. It keeps track of my PDFs, my citation metadata, and my notes, and allows me to share it easily with others (I edit offline, and use rsync to update my server).

I've tried to document the stuff I've done here, and there's also a short screencast showing my "academic workflow": http://reganmian.net/wiki/researchr:start

I would love to make all the info on my wiki readable by Zotero and others. I am already exposing the full bibtex entries on the article pages, but Zotero doesn't seem to catch that. I'd love to provide other microformats etc, but looking at the options on Zotero's wiki didn't make me much smarter. I'd love to find a library for php or ruby which can take a bibtex string and transform to RDF or BIBO or cito or whatever... (Or even a CSL string, so that every citation manager that can use CSL, including citeproc-ruby, could generate these microformats).

I also wish the Zotero web interface was more usable for people with other citation systems. If I see somebody's shared library on Zotero web, and want to import their publications to BibDesk, I don't see any option - no way to view BibTex, or any other formats? I think Zotero is great, but an open citation infrastructure needs to build on common standards, where it doesn't matter which citation system you use, it shouldn't depend on everyone adopting Zotero.

Stian
  • Stian -
    Zotero recently activated and documented a read/write web API
    http://www.zotero.org/support/dev/web_api
    the API for the local database has existed longer:
    http://www.zotero.org/support/dev/client_coding/javascript_api
    So interfacing seems to work pretty standard relatively well documented - that there isn't a comparable ecosystem of parsers etc. as for bibdesk is pretty natural - bibdesk has been around much longer - there are already several people working on various projects interacting with the Zotero database through the API (and, of course, gnotero, which accesses the mysql directly, I believe).

    On your wiki you already have your data as bibtex you can just used UnApi
    http://unapi.info/
    to serve it to Zotero (which will recognize and import it) - seems pretty simple. I don't really understand why you'd even want to supply other micro-formats - doing one and doing that well seems perfectly sufficient
    See this site for more if you haven't already seen it:
    http://www.zotero.org/support/dev/exposing_metadata

    The web-interface is currently being overhauled entirely - the current online display was never intended to be particularly useful - you can't do anything meaningful like search, order etc. - it is my understanding that the new web interface will include standard export options among many other things.
  • BibDesk just uses BibTeX last I checked. That's a feature for some people. a liability for others (certainly for me).
Sign In or Register to comment.