I use RefWorks and love it so far. It’s still a little sloppy in terms of interface for my tastes, but I *have* to have portability. I need to be able to login from anywhere and manage my stuff, so for now the only option for me is RefWorks.

* Author handling was particularly bad. Anyone with multi-part last names, titles, or suffixes would typically be broken incorrectly into First & Last Name fields. No authority control to allow for global repairs either; each entry would have to be fixed manually. (As near as I can tell, there is no way to fix author parsing in Zotero without editing the actual source code. In EndNote, I could’ve tweaked the filter.)

* Many of the reference types mapped incorrectly. I cannot remember which Zotero types mismatched with EndNote specifically, but I remember there were several occurrences.

* And, when there was a mismatch, the fields would likewise not parse correctly. It made a pretty big hash of it, as I recall. Lots of dropped info, and a fair amount of garbage in the Notes field of most EndNote records.

And yes, this was from RIS export from Zotero — I remember specifically looking over Zotero’s web pages for suggestions before I embarked on the project. I followed their instructions to the letter. I never notified Zotero’s developers of the problems I had.

I appreciate your fervor for Zotero. I wish I could share in it, but I don’t.

Which record types didn’t match & what format did you use for data exchange between Zotero and EndNote? Did you report that?

Unfortunately, EndNote has no support for a standard file format that describes how to import PDFs that I am aware of.

Zotero doesn’t import EndNote PDFs yet either, but there’s a description of working around an extension that is peculiar to EndNote to a standard file format that might enable it in the future.

I agree that Zotero has value. I’ve said several times that I hope Zotero eventually becomes the open source alternative to EndNote. Right now, though, I cannot recommend it to anyone who isn’t a significantly advanced user of such products. And even then, it has some big problems.

As an experiment, I tried to write a book chapter using nothing but Zotero, just last year. It did a top-notch job of gathering the reference metadata from the web, one area in which the software really excels. I was very pleased with the resources I was able to tap to put together what I thought was a very good bibliography.

However, Zotero mangled the output style so badly that I was forced to try transferring the records to EndNote instead. Most of the records transfered in the wrong format because the record types don’t match, fields got dropped or mismatched, and the PDFs I’d attached didn’t go at all. Ultimately, I ended up reconstructing the entire library by hand in EndNote. It cost me hours of work.

I’ve been using reference management software for over a decade now and I’ve never had as many problems with putting together a bib than I had when I tried to use Zotero. It just isn’t ready for prime time.

I think the numbers of citations lie. There certainly aren’t requests for 2500 different styles from Zotero users & I doubt that all 2500 Endnote styles get used much. I also know that some of those styles ine EndNote are more-or-less identical to one another, though with different names. This might be good for usability–end users shouldn’t have to be able to figure out the technical name for a style used by a journal they write for is. Maybe Zotero will need to adopt a similar model when they establish a better online repository of styles. Finally: Endnote uses the output styles to refer to file export formats too. Many reference managers (including Zotero) separate export formats from citation styles (which would add another half dozen or so to Zotero’s side).

Your patrons don’t need to “change browsers.” They are free to use IE for most browsing & launch firefox to use Zotero as if it was a stand-alone application. To say that EndNote and Reference Manager (and other desktop applications) “work with either browser” is just not true. One can’t go to Wikipedia or Amazon or the NY Times or even most OPACs & download citations to Endnote yet. You can go to some sites & manually download an RIS or ISI file & import it into endnote. But the same exact thing is doable with Zotero (even if you used IE to download that file). Maybe this suggests Zotero “works with either browser” too. It just happens to offer a better end user experience for internet users when they use Firefox (as compared to the desktop programs).

I’ve played with most reference managers out there & use more than one every day. Zotero isn’t a perfect reference manager or even the one I use most frequently. But I think you really sell it short. I think it is able to play with other reference managers better than the average alternative. In my own experience, Zotero has generally better exporting out-of-the-box than EndNote. Perhaps this is because Zotero has richer output (such as MODS XML) that I can adapt as I see fit–I don’t know. I’ve occasionally been able to somewhat easily build custom EndNote export formats that are better for my needs than anything that comes with either program, though.

I’ve found the Zotero devs FAR more responsive to user input than those at Thomson & there are EndNote bugs and annoyances which have been around for a LONG time without seeing patches.

I don’t know what others have complained about, but Ms. Hoberecht’s example just isn’t a very good one. The program that intercepts ISI/RIS files is configurable & I’ve seen complaints from people saying that Endnote was intercepting links they wanted to put into Zotero. Would you think that this is an example of EndNote being “hostile” to Zotero?

Agreed that making styles, either with EndNote or Zotero (or any other product) can be tedious. Doesn’t change the fact however that EndNote offers over 50 times as many styles, many for obscure or non-U.S. titles.

Also, the fact that IE isn’t developer-friendly doesn’t change the fact that 80 percent of my patrons use it. Suggesting that they change browsers for a single extension when there are better products than Zotero available to them would be laughable. And, in fact, EndNote, as well as Reference Manager and Ref Works, work with either browser because they are not browser-dependent and don’t use plug-ins. You can download records equally well with either browser.

Trying to export to EndNote from Zotero causes dropped and mismatched fields for many record types, both standard and non-standard. Also, any PDFs have to be transferred manually, which is tedious in the extreme. Frankly, the program is hostile to EndNote users, as evidenced by Ms. Hoberecht’s and others experiences. Zotero’s programmers would do well to recognize that there is a “bigger bear” in the room and try to find ways to play with it nicely.

* IE is not nearly as developer-friendly as firefox. You can (and some people do) only use firefox because of a few extension, such as Zotero. The browser issue is a red herring. Endnote doesn’t work with IE or firefox!

* Zotero has web services planned & these will presumably work with IE. You can already easily export to connotea, CiteULike, refbase, etc. What problems did you have with exporting to Endnote?

@Ms. Hoberecht:
There is a preference for whether Zotero should intercept Endnote files or not. This has been there since the beginning & it doesn’t seem that Zotero assumes you want to be locked into using it instead of other reference managers.