Your comments

Oleg, bibbase turns author names into links for all those authors who also use bibbase. I've noticed on your page that some of your coauthors who get linked don't actually use bibbase, but that bibbase currently incorrectly guesses the owners of some pages they link to. I'll look into that because it shouldn't happen.

More generally though it is not possible to suppress those links. If and when it is working correctly, I consider it a feature that bibbase automatically keeps track of the current page of your coauthors. Many authors like to keep a list of their collaborators and this is a feature that offers that with no effort on the author's side.
Stephane,

This seems to be a problem that is limited to some of your papers. I've checked some of the other BibBase-Mendeley users, and their papers download just fine.

Then I started checking papers on your page and noticed that at least this one does download fine:

Procrustean co-inertia analysis for the linking of multivariate data sets. Dray, S.; Chessel, D.; and Thioulouse, J. Ecoscience, 10(1):110-119. 2003.

Is there anything different about this paper from the others in terms of permissions or so? Did you perhaps change any settings in Mendeley recently?

The download URL starts with ../download/public, so it doesn't sound like the download should be restricted in general.

-- Christian
Hi Stephane,

Thanks for pointing this out. I'll investigate this further, but yes, on first sight it seems that they have changed the permissions where logged in users can only download pdfs in their library. But I can't say for sure. Elsevier has been getting more aggressive about sueing copyright violators recently. Maybe this has something to do with that (Elsevier bought Mendeley not too long ago).

If you are at all interested in alternatives to Mendeley, I recommend Zotero. It's API is a lot nicer and more reliable and the integration with BibBase works like a charm.

-- Christian

It sounds like a bug, but I guess this is really a feature request, right? If so, I'm happy to still look into that. Can you point me to some documentation of that CoinS you are mentioning? Is that a special markup of sorts?
It sounds like a bug, but I guess this is really a feature request, right? If so, I'm happy to still look into that. Can you point me to some documentation of that CoinS you are mentioning. Is that a special markup of sorts?

Thanks for your feedback, this is really helpful. I'll look into it.

Please see my last comment.

Gustavo,


I think the problem is not https but dropbox. There is a redirect. When I just retrieve the URL you gave me, I get the following response. The right URL is below.


> curl https://www.dropbox.com/s/ln4urie0mcnpbh9/cevolani-pub-prova.bib

<html>

  <head><title>Found</title></head>

  <body>

    <h1>Found</h1>

    <p>The resource was found at <a href="https://dl.dropboxusercontent.com/s/ln4urie0mcnpbh9/cevolani-pub-prova.bib?token_hash=AAEKvo7xMLwDigTIZRhjqMyfJDrCi7Cyqm4JIZxkuA6krA">https://dl.dropboxusercontent.com/s/ln4urie0mcnpbh9/cevolani-pub-prova.bib?token_hash=AAEKvo7xMLwDigTIZRhjqMyfJDrCi7Cyqm4JIZxkuA6krA</a>;

you should be redirected automatically.

<!--  --></p>

    <hr noshade>

    <div align="right">WSGI Server</div>

  </body>

</html>


If I use the URL that dropbox indicates as the redirect it works:


http://bibbase.org/show?bib=https://dl.dropboxusercontent.com/s/ln4urie0mcnpbh9/cevolani-pub-prova.bib?token_hash=AAEKvo7xMLwDigTIZRhjqMyfJDrCi7Cyqm4JIZxkuA6krA

This is a bug. https is supported and used to work. I'll look into this. Could you give me the https url you were testing with so I can test against that one as well?


Thanks.

Hi Amit,

This should still be possible. The dropdown menu doesn't contain this, but on your page you can still use the &groupby=keyword parameter in your bibbase URL. Please also give me your bibbase URL so I can verify that for myself.

Christian