Just tried RTF scan in a couple of other reference managers... If you do eventually implement this request, please don't copy Zotero. They have tried to achieve this functionality without requiring unique identifiers in the temporary citations (e.g. {Smith, 2001}) and the results are poor.

However in EndNote (and I expect all the other reference managers in my previous comment), unique identifiers are required, e.g {Smith, 2001 #456}. The results are excellent. I assume Zotero's scan works poorly because it is difficult to do without the unique identifiers.

Also, something that hasn't been suggested in previous comments below is that the Mendeley plugin for Word and OpenOffice should be able to convert back and forth between temporary citations (used by rtf scan) and formatted citations (used by cite-while-you write). This is a useful feature in the EndNote plugin for Word (see the "update citations and bibliography" and "convert to unformated citations" functions in the EndNote plugin of Word for Mac).

Just did a web search. Many reference managers have this feature: EndNote, Zotero, Papers, Sente, Bookends, Refworks, Refbase, Citavi, Biblioscape

robert.knight: Thanks I will look into this.

However, I'll note out that it is not a cross-platform solution and from what I have seen in my first foray into bibtex/latex in the last couple of days, other (latex-based) solutions seem a bit stuck in the 80s/90s for my taste and require an excessive amount of effort to use with no clear guidelines on the net about how to get started (for absolute beginners). So I still contend Mendeley should implement a built in RTF scan feature.

Mendeley has a bibtex syncing feature. I am not familiar with bibtex - is there any software I can use to achieve RTF scan functionality using the synced bibtex file?

Seems like a relatively high impact solution to the problem that many people's word processor / outlining tool of choice is not supported by cite-while-you-write plugins.

Both Zotero and Endnote have RTF scan.

Is it currently possible to search a date range?
year:1993-1995 (for example) does not seem to work, but I wonder if the required syntax might be different.

Also, please keep a search history so I can choose previously used search terms.

The logical NOT operator in the following term is not working
author:doe -author:smith

The search result is the same as
author:doe author:smith

Also, please ignore accents and other diacritical marks in searches. In case its not on your radar - there is a specific request for this:
http://feedback.mendeley.com/forums/4941-mendeley-feedback/suggestions/926129-ignore-accents-during-search-either-by-default-or

The current inability to order the search results is a major annoyance.

Being able to save searches as folders would be very useful too:
http://feedback.mendeley.com/forums/4941-mendeley-feedback/suggestions/250337-smart-categories-

I have not installed the 1.5 preview release myself, but someone has commented on the blog post for the 1.5 release that the journal abbreviations hack is no longer working.

Please get it working again before releasing the stable version of 1.5

Anonymous: Why do you need the citeproc-js folder?

Anyway you can get to citeproc-js by right clicking mendeley desktop.app and showing package contents or by using an app that is capable of searching within packages, e.g. http://itunes.apple.com/gb/app/easyfind/id411673888?mt=12

I am noticing some imperfect behaviour of the journal abbreviations feature when using copy and past (C&P) or drag and drop (D&D) from Mendeley desktop. When MD is launched, C&P or D&D initially ignores journal abbreviations. You first have to switch to another style and do a C&P or D&D using that other style before journal abbreviations starts to work (MD for Mac 1.1.2).

Following the last couple of comments, I hope when Mendeley get around to implementing journal abbreviations properly, they make it easy to switch between different abbreviation lists from within Mendeley desktop.

Elad, you don't need to use the dev version anymore. It works with the current stable version.

Thanks Gaston. I may wait until stable 1.0 before upgrading, but good to know your lists will be available.

Thanks for explanation Steve. Is my understanding correct that users will have to pupulate the default.txt file with abbreviations themselves? I have not yet upgraded to try myself and it is confusing that abbreviations was working for tamsa apparently before creating default.txt. Requiring users to populate default.txt would be frustrating. It would be more helpful if Mendeley provided base lists that the user could add to or tweak or override if needed. There should be base lists for various abbreviation schemes that are in use (e.g. like loadabbrevs.js lists provided for MedAbbr and IsoAbbr schemes by Gaston 1-2 months ago).

@Robert,
Can you elaborate on the nature of the "slightly better solution" coming in the next dev preview please? The hack posted by Lochana & Gaston has been working very well for me and this is the reason I have refrained from installing the recent dev preview release. Specifically, I am interested to know if Mendeley's solution is going to be similar to the current hack (i.e. using abbreviation lists) or if it will involve the addition of an abbreviated title field to each item that the user will have to make sure is populated for each item in their library. The former seems better to me because it allows for switching between the different abbreviation conventions (IsoAbbr, MedAbbr etc) that different journals may request when submitting a paper, but judging by the comment from tamsa it seems you may be going for the latter option.

Gaston: Here is an example of a PMID for an article from a journal with accented characters in its full name: 21516330. I don't know if it is of any note, but for epub ahead of print articles from this journal, the journal name obtained from PMID search does not contain accents (e.g. currently this is ahead of print: 21597929).

Gaston: Thanks very much. This seems to work very well (I have tried both versions of the runcites.js)

I posted a comment with another question a few hours ago that does not seem to have shown up on this forum. I've sent a support request to ask why this happened but the gist of it was, is it possible for this method to ignore diacritical marks (accents etc) in the full journal name? I have come across an instance where the PMID search pulls down a full journal name that contains accents but laodabbrevs.js does not contained accented version of the journal name.

Gaston: Thanks very much. The update is working well.

A different issue I have is this: There is a journal that I frequently cite that contains accented characters in the full name pulled down by Mendeley via the PMID search (even though it does not contain accented characters in this list: ftp://ftp.ncbi.nih.gov/pubmed/J_Medline.txt).

In case of interest, the name of the journal I am referring to, as pulled down by PMID search, is: Experimental brain research. Experimentelle Hirnforschung. Expérimentation cérébrale

Like the list I pointed to above, the loadabbrevs.js file does not contain accented characters in the entry for that journal and consequently it does not get abbreviated in the bibliography. I have no idea if there are other journals similarly affected but I would not be surprised if there are.

Is there a way to make this method ignore diacritical marks in the full journal names?

Lochana & Gaston: Thanks very much for the workaround.
Gaston: The 'loadabbrevs.js' file that you have provided contains IsoAbbr abbreviations. Is there a version with MedAbbr abbreviations?

@Jason,
They shouldn't turn off importing the full title. Although uncommon in your field (I assume a bioscience - based on pubmed use), some citation styles do require you cite using the full title. Journal entries should have a full title field and also an abbreviated title field (zotero has this and I would be very surprised if any of the other major reference managers don't have this). The citation style being used to create your bibliography should then speficy whether the full title or the abbreviated title is used in the style. In fact, I suspect this is already supported in citation style language (the common open language used by mendeley and zotero to create citations and bibliographies within your documents). If that's true I don't know what Mendeley needs to do other than adding an abbreviated title field to the journal entries - which seems like it should be a fairly trivial task.

Also, I think it would be useful if you could allow me to choose which collections/folders contribute to your overall list of receomendations for me. I don't want the recomendations to be based on papers I was reading 5 years ago that are no longer relevant to my current research.

Since posting this suggestion I have noticed that in my dashboard mendeley provides a list of recomended papers based on my whole library.

However, as per my original suggestion, please also provide me with recomended papers based on individual collections/folders. This would provide a broader list of recomendations than those based on individual papers, and a more focussed list than those based on my whole library.

On the blog post linked to at the end of this comment, you invite us to

"not only upload the final, peer-reviewed copies of publications for which they maintained rights to their Mendeley profile, but also to upload the pre-publication versions as well"

I would love to do this. But without proper management of attachments (so that each attachment can be clearly identified on my profile as published version, pre-publication version, supplemental data etc) it's pointless advice.

Why has this idea not even got a status? It's so obviously required if you want us to upload our work and make it available on our profiles.

http://www.mendeley.com/blog/academic-features/on-sharing-research-and-the-value-of-peer-review-mendeleys-response-to-sopa-and-the-research-works-act/

On the mendeley web page for one of my papers, the pdf preview displays the supplementary material instead of the main paper: http://www.mendeley.com/research/lack-otolith-involvement-balance-responses-evoked-mastoid-electrical-stimulation/

This makes my profile, and Mendeley web, look a bit shoddy. Surely, the simple & necessary solution would be to allow us to distinguish attachment types in Mendeley desktop.

Although not a proper solution to this problem, why not allow us to manually rename pdfs once they are already in mendeley? Then I could rename a pdf that has been auto-named to whatever I want.

Yes, either a dedicated supplementary files section or a better naming convention for supplemental files (or perhaps disable auto-renaming on supplemental files).