The download is the bibtex file. This is the intended behavior.

Https is now working. You need to provide the additional parameter "protocol=https". With this it is now possible to use Zotero groups, as long as you know the group id. For instance, for the group with id 168779, here is the bibbase page:  

http://www.bibbase.org/cgi-bin/pyBibBase/pyBibBase.cgi?protocol=https&bib=api.zotero.org%2Fgroups%2F168779%2Fitems%3Fformat%3Dbibtex%26limit%3D99

You can still use all other bibbase parameters, incl. group0=year, etc.

-- Christian

Also, you may want to use

\~{n}

{\~n}

Both are legal latex, but I haven't yte implemented the second notation, which is why the entry below is not formatted quite right.

@inproceedings{DBLP:conf/flairs/Munoz-AvilaAJKM10,
  author    = {Hector Mu{\~n}oz-Avila and
               David W. Aha and
               Ulit Jaidee and
               Matthew Klenk and
               Matthew Molineaux},
  title     = {Applying Goal Driven Autonomy to a Team Shooter Game},
  booktitle = {Proceedings of the Twenty-Third International Florida Artificial
               Intelligence Research Society Conference, May 19-21, 2010,
               Daytona Beach, Florida},
  year      = {2010},
  ee        = {http://www.aaai.org/ocs/index.php/FLAIRS/2010/paper/view/1307},
  bibsource = {DBLP, http://dblp.uni-trier.de}
}

@inproceedings{dehghani2008,
  title={Order of Magnitude Reasoning in Modeling Moral Decision-Making}
  author={Morteza Dehghani and Emmett Tomai and Ken Forbus and Matthew Klenk},
  journal={Proceedings of the 22nd international workshop on qualitative reasoning},
  year={2008},
  url={http://matthewklenk.com/papers/QR08-MoralDM.pdf}
}

Jeroen,

If you can find your groupid (different from the name, e.g., 168779), then you should be able to just use the following URL for bibbase: (with the groupid replaced)

https://api.zotero.org/groups/168779/items?format=bibtex&limit=99

You can find the groupid by opening the group settings, and then looking at the URL.

This just for now. And please let me know if this works and if you have a specific group with some useful content to try this on. I'm still planning to write a proper extension for bibbase to talk to the Zotero API, so people can also show their Zotero collections using bibbase.

UPDATE: sorry, no, actually this won't work yet. I haven't yet implemented support for https and that's what Zotero is using for their API.

Good idea! I've had a very brief look at their API this morning and it may actually be almost trivial. I'll keep you posted.

Yeah, I was wondering how that would affect the use of Mendeley. Frankly, Mendeley's API was never really great, so if people were to abandon it, it would make things easier for me.

Yes, I'm aware of this issue. I believe it only happens for groups, so if you have a collection containing all the same papers then it should work fine. There are some issues with the Mendeley API that I need to work through to fix this.

This should be fixed now. Please let me know if you notice any more issues with this. The UBC page seems to be doing some server-side caching (probably of the php request), so there it isn't working yet as of right now (7:48pm on pi-day), but should start working soon.

Thanks for pointing this out!

(setting status)

This is a great suggestion! Thanks for the pointer. It is indeed a goal of bibbase to do the SEO (search engine optimization) for its users. So features like this rank highly on my to-do list. I'll look into it.

Update (see comments): this is a non-issue. As for Mendeley papers, we have to respect Mendeley's robots.txt just as much as Google does.

Google respects the robot.txt instructions set forth by web sites. Mendeley seems to disallow bots to retrieve PDFs: http://www.mendeley.com/robots.txt. BibBase won't -- and shouldn't try to -- circumvent this. If you would like your PDFs to be picked up by Google, you can download them from Mendeley and put them on your personal web site.

I just looked at the Google scholar instructions. For those PDFs that appear on a bibbase page and which are hosted on a web site that Google's bot can access, this should already work.