Hi Maximilian,

Thanks for these links. The second one didn't work for me (I couldn't get the .bib), but the first one was very helpful for hardening BibBase. It turns out that the file had some issues, e.g., the use of invalid bibtex entry types (e.g., @conference and @url). While that is a problem with the input, I still want BibBase to be robust enough to handle that, and debugging with this file helped me accomplish that.

So now your file renders just fine---after a while ;-), it is pretty big.
http://bibbase.org/show?bib=https://raw.githubus...

Thanks for the praise :-)

Regarding the feature you are requesting: I have been thinking about something like that. As you know it is possible to mark papers as read, and I was considering storing the date when that happens and then enabling sorting by that. That should definitely be possible.

Regarding that 4MB file: I never thought about large collections like that. I'm a bit surprised that it doesn't work. Could you send me that file or give me a link to it so I can see where it might be failing?

Regarding donations: that's very kind of you. But what I need more than money right now is a way to get more users, usage, and overall traction. So perhaps rather than accepting your offer, would you mind if I added you to my list of "power users" who I can run ideas by every once in a while? There are some possible future directions I'm contemplating, and there is nothing more valuable than feedback from your users to see what they would find valuable.

The link you provided is not the URL of a bibtex file. It's a regular html page that has a download link. But it's not the .bib itself. Here is what I see when I open the link.



You need to use the download link with BibBase, since that download link is the actual URL of your bibtex file. That's what I did in the following link and it seems to work just fine:

http://bibbase.org/show?bib=https://dl.dropboxuser...

Levente,

Thanks for the bug report. I found a different bug and fixed it. The link now seems to work.

Mendeley is not well supported at this point (as you can tell), but I'm working on a new implementation right now that should be more robust.

-- Christian

Walter: Thanks for reporting this and my apologies for this bug. Given your report I was now able to find the real problem (some permissions had to be changed on the server). Your page now seems to be working again.

BTW, when you are embedding this page into your homepage then this problem might go away all by itself, since you are probably specifying the character coding as utf-8 in your homepage already. This is set in the head of the html. BibBase pages don't have a head, so that they can be embedded easily. But this headlessness also means that I can't set the encoding in the bibbase page itself.

Eliseo: it looks fine to me (see screenshot below). What most probably is happening is that you are nor using unicode (utf-8) as your character encoding in your browser, or the font you are using as your default fault in your browser simply doesn't support utf-8.



In Chrome you can change the font and encoding in the advanced settings:

Stéphane, thanks for the pointer and the clarification. That is indeed too bad that Mendeley treats some pdfs as copyrighted when in fact they are not.