Talk:Comparison of reference management software/Archives/2014

"Evil"
I understand that lock-in is a bad thing for many (maybe even most) users, but it doesn't seem neutral to call that "evil". As WP:NPOV says: "The tone of Wikipedia articles should be impartial, neither endorsing nor rejecting a particular point of view." Presumably the authors or publishers of software that locks in users would not agree with its characterization as "evil". Why not just describe the policy (lock-in) rather than take the position that it is "evil"? --Macrakis (talk) 00:05, 23 December 2013 (UTC)
 * It also seems to violate WP:NOR. I plan to revert it out-right. --Karnesky (talk) 21:00, 24 January 2014 (UTC)

MODS support
Does someone have a reference (or even a screenshot) showing Reference Manager MODS XML export? --Karnesky 20:05, 25 April 2007 (UTC)
 * It is not documented anywhere, so it seems to have been a false entry. --Karnesky 20:09, 1 June 2007 (UTC)

Import formats
I wonder about the selection of import formats. Particularly I don't believe CSA format exists anymore, as all CSA databases are now part of Proquest, which doesn't have its own format. On the other hand it seems odd that MARC and MARCXML are missing? bastel (talk) 17:20, 8 February 2014 (UTC)

References for the import formats would be useful too. The product seems to conflate file import with database connectivity for PubMed/Medline/etc. --Karnesky 20:08, 25 April 2007 (UTC)
 * In particular, can someone cite bibtex support? The other thompson products can't import this & the ftp site shows no filter.  --Karnesky 11:44, 24 May 2007 (UTC)
 * Again, the program's own documentation and another comparison betray mistaken entries. --Karnesky 20:12, 1 June 2007 (UTC)

Export formats
Same as Karnesky, I can't find any support for exporting to BibTeX (with RefMan 12.3). I also couldn't find anything in the program's help or on their website. I'm going to correct this entry. --JonathanWilliford (talk) 15:31, 27 October 2011 (UTC)

Potential mistake
I am not familiar with the information presented here, but I think I hove found a mistake. "BibBase	Christian Fritz	2005	2013-07 (v3)	Free	No	proprietary	centrally-hosted website, intended for publication pages" This is one of the table entries. How can a program have a cost of free but be not free (look at the table if this doesn't make sense)? Again, I do not know the software but I believe it at least needs some clarification. Guy who couldn&#39;t get a username (talk) 17:20, 28 July 2014 (UTC) After looking back over the article, I found that there are several such entries. Can we work on a way to clarify this? Guy who couldn&#39;t get a username (talk) 17:22, 28 July 2014 (UTC)

Database connectivity
The database connectivity table is biased towards the needs of academics in the sciences and tech sectors and so is misleading as to the real capabilities of the applications listed. Going by the table, RefWorks' connectivity looks terrible because (except for PubMed) all it offers falls under the "other" column. Now, I'm not a fan of RefWorks. In fact, I found the Wikipedia article because I was looking at alternatives. But the table ought to be rethought. Here are two columns that could be added, for instance. --Kartavyam (talk) 13:21, 19 March 2008 (UTC)


 * It is currently limited, but I'd hardly call it 'biased.' We should eventually add MANY more online databases (in addition to ATLA & JSTOR).  Feel free to add these! --Karnesky (talk) 15:17, 19 March 2008 (UTC)
 * Well "bias" is not necessarily intentional. As for adding columns to the table, is the best thing to just add columns with a bunch of question marks to the article itself?  Or should new columns be staged on the discussion page and then moved to the article when they've been filled enough (e.g. > 50%).   BTW, the table fragment above was intended for staging, people should feel free to edit it and add what they know. --Kartavyam (talk) 10:57, 20 March 2008 (UTC)
 * Ideally, neither: someone would do the leg work of looking up references (such as the other comparisons that this article links to) so that they could complete the other columns. One should not rely on their personal experience testing applications for adding information anyway.  Rather than long tables on the talk page (such as the one above), a simple list of the databases that should be added might be more helpful.  Perhaps a table would be o.k. on a subpage, but it seems "heavy" for a discussion page & would be harder to track & is unusual to find on other talk pages.  If most of the rows can be completed, I think it is fine for their to be question marks in the main article. --Karnesky (talk) 13:32, 20 March 2008 (UTC)
 * Well, this ideal scenario is not something I can contribute to, for lack of time. As for personal experience, there's a difference between what somebody who relies on the marketing literature and documentation but does not use the software can tell you and what somebody who actually uses the software can tell you.  For instance, yes, RefWorks exports to BibTeX but what the RefWorks people don't tell you is that it does not treat accented characters in a safe way (BibTeX chokes on what is exported by RefWorks) and it creates utterly useless keys to use in \cite commands.  I've had to write my own python tool to clean up what RefWorks exports.  In effect, this makes the BibTeX export capability of RefWorks useless for people who can't or don't want to deal with cleaning up the mess.  This is a piece of information that could materially impact whether someone wants to use RefWorks or not but you don't get that from the literature provided by the company nor from people who perform a mere cursory inspection of the software. Only actual experience with the software will reveal this shortcoming.  --Kartavyam (talk) 14:25, 20 March 2008 (UTC)

Please refer to WP:NOR. I never said we should rely on marketing literature--in many cases, features will be documented by some third party (such as those we link to) that hopefully satisfy WP:RS. --Karnesky (talk) 14:37, 20 March 2008 (UTC)

I think the databases that are added should not be a function so much of what each software supports, or what its marketing says, but what's potentially useful to various scholars (probably a lot more contentious!). For example, the current list very biased to sciences (pub med, IEEE, arXiv) and not the needs of e.g. Humanities scholars (say ... JSTOR would be a good inclusion there). However until I finish writing my thesis I don't have time to add this so feel free to ignore me. When I'm done I may come back and adjust it if no-one else has in the meantime. (For the record I use Endnote, for my secondary sources only. I'm finding it a bitch especially as I deal with Classics, whose ancient authors need to be cited in a specific way, along with object catalogues and specific abbreviations for special collections, along with the modern secondary sources in the usual styles, but it just cannot support two or more styles of referencing succesfully (it claims to, but implementation wise it sucks)). GermanicusCaesar (talk) 04:50, 6 November 2010 (UTC)

API and developer extension points
I'm not sure if its what the previous commenter was after or not, but it would be really useful if this comparison could include whether or not each project offers api for working with the references/citations. For instance, I am trying to build an app which searches EndNote references and I learned that it only offers a C/C++ API (only desktop version of course). I was hoping the web version would offer some kind of web API for accessing the references, but no luck. Now I'm wondering if any of these other projects offer this capability. Other API/developer extension points could be interesting to catalogue as well. --Lmsurprenant (talk) 14:49, 2 October 2014 (UTC)