On Jan 17, 2008, at 5:19 PM, Andy Green wrote: > I wrote a while ago about getting z39.50 support working with NASA > ADS, and you guys were great. Unfortunately, I found out from ADS > that they are not planning to support z39.50, and may remove it.
Well, that sucks. Their z39.50 support is lousy, but at least z39.50 / is/ an established, documented library standard, and suited to network queries. Why are they considering removing it? > However, the new web search feature has tempted me, as I see more > potential there. ADS is able to return standard bibtex entries as > search results. So I got it working such that I could search, get a > results page (all in BibTeX, which I ignored) in the first pane, and > then the results ready to import in the second and third pane. But > now that doesn't seem to work. I'm guessing that the first two lines > of the results (which aren't valid bibtex) are confusing the parser, > but I'm not sure. If you can provide a link, that would help...but your diagnosis is likely correct. The web parsers are all one-off jobs for each site. Not hard to write, but each one requires custom code and runs a chance of breaking. We might want to reconsider allowing plugins for them... > Seems it would be easy to get ADS working similarly to Google > Scholar, or similar. Also, it might be better/easier to use the ADS > Tagged Format instead of BibTeX, since it includes the abstracts. I > can send more details if that helps. Well, the ADS tagged format is documented, and not really hard to parse. But it still sucks. Why does every site feel a need to a) write their own bastard server/search tool b) use their own bastard query syntax and c) write their own bastard tagged data format? I know that's not your fault, but it really irritates me. > On a related note, it would be nice, now that the web kit is > installed, to be able to visit a linked web page from within BibDesk, > and even nicer, if clicking on a pdf (or maybe even other types) link > on that web page would automatically download and link the file to > the BibDesk entry. (Someone showed me this feature in Pages today, > and I was very impressed.) Yeah, I'd like that too. Hopefully someone will implement it :). One issue is that there's often not a single entry to link it to (e.g. in Google Scholar results). I haven't looked at Papers (assuming that's what you meant?) lately to see how they handle this. -- adam ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
