Hi Michel,
> I came across Citeline and it's very nice. Well done!
Thank you!
> However, I can't see how to update the bibliography exhibit without going again through the process of uploading the new version of the bibtex file to create a new exhibit and then configure the way it's displayed again from scratch, and finally downloading to my site. I am correct or am I missing some point?
>
You are correct. That was our intent, since the users we designed this
for (i.e. faculty and researchers) are using various personal
bibliographic managements systems to edit their citations -- e.g.
Endnote, Refworks, Zotero -- and we don't want to force them to do
double editing. We've made this very easy with open source tools like
Zotero, it's just one click to update your Exhibit. But you're right
that in general you have to repeat the process each time you want to
update your Exhibit.
> I tried to translate the bibtex file via Babel to JSONP format and then change the source of the downloaded HTML file to point to my local JSONP file instead of the data.jsonp file on your server, but it didn't work anymore.
>
We could probably diagnose the problem if you sent the HTML, but could
you say a bit about why you wanted to do this? Was it to maintain your
bibliography in JSONP rather than the original bibtex file?
> Without any simple way of keeping the system working whenever a new reference is added to the bibtex file, I'm afraid the Citeline system won't become popular.
We certainly considered that, but writing a system to edit and manage
citations would be orders of magnitude more complicated that what
Citeline does. We want it to work seamlessly with existing, popular
editing tools and avoid having people maintaining this data in multiple
places. Does that make sense?
> Moreover, it seems contrary to the Exhibit philosophy of separating presentation from data. If whenever I change the data (the bibtex file), I have to go again through the same exhibition generation steps from start, then it seems the two are actually quite hard-wired...
>
Well, the system does support templates that can be updated
automatically from an updated bibtex file (e.g. in the Zotero export
case), so it can be as little as two extra clicks.
> Not to mention that keeping our bibliographical data on your website without any explicit statement in the citeline seems to contravene basic privacy good practice.
>
That is a convenience for users who don't want to store their data
locally and change their Exhibit HTML file. If you want to do that it is
possible -- you export the Exhibit JSON (in Exhibit), save it locally,
and edit the HTML file accordingly. But I do agree that some sort of
statement in the Citeline "about" file that explains this would be a
good idea. You are clearly more expert at how these tools might work
than our typical user!
I can also assure you that we are not storing any personal information
beyond the userid of the Exhibit owner.
> Thanks in advance for any clarifications!
>
You're welcome, hope this helped, and please let us know your reactions
to our thinking about this service.
By the way, since Citeline and Exhibit are 100% open source, we
anticipate that some institutions will decide to run their own Citeline
services rather than relying on ours. So if MIT's policies were
troublesome but you like the service you could consider that route.
Best,
MacKenzie
--
MacKenzie Smith
Associate Director for Technology
MIT Libraries
Received on Tue Sep 02 2008 - 12:03:46 EDT