On 10/2/07, James Howison <[EMAIL PROTECTED]> wrote: > > On Oct 2, 2007, at 9:08 AM, Michael McCracken wrote: > > > On 10/1/07, Simon Spiegel <[EMAIL PROTECTED]> wrote: > >>> I just added a scraper to support Google Scholar in the Web Group. > >>> > >>> Here are a couple of notes about it: > >>> - you need to change your google scholar preferences to add the > >>> "Import into BiBTeX" link. You can do this from within BibDesk, > >>> it'll > >>> stick. > >>> > >>> - If your language is not english, the parser won't work! We can > >>> localize this, though - we'd just need to look for the 'lang' > >>> param in > >>> the URL. > >>> > >>> - It's not as easy to get a useful URL for these links, because > >>> there > >>> are many results on one page and the markup is not really easy to > >>> decipher. > >>> For starters, it's invalid in a very big way - 700 errors or so, > >>> according to the W3c validator. > >>> > >>> So I haven't tackled that. > >>> > >>> Otherwise, it seems to be working pretty well. I'm interested in > >>> hearing feedback. > >>> > >>> Thanks, > >>> -mike > >> > >> > >> This looks interesting. I only played with Google Scholar a bit since > >> this is the only page I use, but it looks nice. > >> > >> A question: I realize that this a bit outside the scope of this > >> project but would it be possible to turn this into a browser plugin? > >> While surfing inside BibDesk is nice, most people probably preferred > >> to use Safari for this. So what would be ideal IMO is a Zotero-like > >> approach. Have a plugin for your Webkit browser which will tell you > >> if a website contains bibliographic data - two clicks and the data is > >> imported into BibDesk. As I said, I know this is quite a request but > >> it seems like the natural way to do this, especially if more formats > >> get added. > > > > I sort of like this idea, and I believe that it will be getting easier > > to write such plugins for safari in 10.5. > > > > I only say 'sort of' because I don't use safari all the time, I > > believe Adam is an OmniWeb user, and I think this is representative - > > we would have to write plugins for many browsers. > > > > I also think that as the web group feature matures, there will be some > > advantages to having the browser be part of Bibdesk, such as providing > > special actions to download PDFs and file them automatically... > > > > However, there is no reason a plugin couldn't be written (as in, not > > by me :)) that would use applescript to send items to Bibdesk. > > > > -mike > > Perhaps it could be added to Operator, the microformats parser? > Yeah, these are not microformats per se (must ufs be html?), but > these bibtex links are a hell of a cowpath.
I'm not sure what you're suggesting. What could be added to Operator? microformats do have to be html, and in these cases, we're not generating or parsing much structured HTML. You know, if you want to propose a microformat that had a chance of being useful soon, try "rel-bibtex" - just give us a link to raw bibtex source, and mark it up with rel="bibtex". Done and done. I'd suggest that on the uf-discuss list if I thought anyone would take it seriously. -mike -- Michael McCracken UCSD CSE PhD Candidate research: http://www.cse.ucsd.edu/~mmccrack/ misc: http://michael-mccracken.net/wp/ ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Bibdesk-develop mailing list Bibdesk-develop@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-develop
