Hi Adam,
I want to thank you again publicly for having done it, it's really
perfect (to be sincere I didn't expect such a response, and so quick)!
For a given @inproceedings, it includes a crossref for the
corresponding @proceedings.
This record:
http://dblp.uni-trier.de/rec/bibtex/conf/srds/ShafiSB03
shows both the @inproceedings and the @proceedings with extra info.
I honestly don't know how hard this could be, but it might be nice to
be able to import a crossref for a paper that has one, since it's
more complete to have both.
I did a search for shafi and came up with this result you linked
to. Double-clicking it showed crossref buttons in the editor, and
clicking one of those opened the crossref. If you can find a case
where the @inproceedings and @proceedings parent aren't downloaded
together, let me know, since the parent entry is definitely required
for these and it's supposed to be present.
That's definitely a crucial feature. There is something I am not sure
to get about the use: is the @proceedings entry supposed to be
imported automatically when I click on the import button of the
@inproceedings one? Because I don't think that happens with me.
Another remark is that new lines and spaces make strange things in the
bibtex entry. For example the last entry appears a bit clumsy like this:
Raptor: Integrating Checkpoints and Thread Migration
forCluster Management
One small question: are the DBLP groups present by default in the
Searches menu? I guess it shouldn't be kept as a secret feature.
The next milestone I guess could be to extract URLs not only from ee
fields, but also from the results of DBLP Webcrawler (especially when
no ee field is provided). DBLP Webcrawler is a quite funny and
surprisingly efficient way (more than Google Scholar I find) to
retrieve PDFs that are hosted on personal homepages for example. You
can try it by making a search in http://dblp.l3s.de/, there is a small
magnifying glass icons after each title which launches a DBLP
Webcrawler search. I can try and get some more details about this kind
of mystical feature, but it seems very interesting.
Some theoretical considerations can be found here
http://www.springerlink.com/content/c63w80ngm110k6x2/
. You may not have access to it, but here is a quote from their
motivation:
One of these libraries is the Digital Bibliography
Library Pro ject (DBLP). From a researcher point of view however it
would be an added value if not
only bibliographical metadata were supplied but also free and direct
access to
the content itself would be possible through the web service. DBLP
faces this
challenge by supplying links to electronic editions where the
information is avail-
able. These links mostly point to publisher webpages where the users
have the
opportunity to buy the desired full text. If the user is a
member of the publishers digital library, e. g. SpringerLink, ACM-DL
or IEEE
digital library, then the full-text is accessible. But most of the
publications how-
ever are not only available through publisher websites but also on
homepages
of research institutes or the researchers themselves. The question
arises if it is
possible to automatically identify relevant full-texts for
bibliographical records
which would lead to an added value for researchers.
Alexandre
-
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
___
Bibdesk-users mailing list
Bibdesk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bibdesk-users
