Organizing background reading without post-it notes?
October 28, 2011 8:29 AM   Subscribe

Software recommendation filter (Windows): I think it's called "reference management software". We're a research collaboration, need to organize existing background reading documents (academic papers, company reports, etc) so 3-5 people can read, comment on their relevance to the project, and do searches to find the useful papers again later. No need for content editing, check-in/check-out, versioning, etc. these are all final-version documents. Boss made a recomendation; I'm looking for a second opinion before I agree.

I just got tapped as a new collaborator on a (now) 3-person R&D project. The main guy has a collection of pdfs and word docs (of academic papers, general articles, company reports, spec sheets for tools and parts we're using, etc), and one of my first tasks is to organize that, so as to educate myself on the topic and provide future collaborators with a crash-course. Most reference-management and document-management software is geared towards writing a book or a giant paper in a collaborative way: check out and editing a document, versioning, and easy input of bibliographic info that can then be linked into your big paper in any of the standard bibliographic styles, etc. I don't actually need any of that since our end goal is an object with documentation, not a big disseratation. The information is going into our heads, not into a future document we're writing. In fact, we'll pretty much never need to edit any of these documents, just read them and edit the commentary and tags.

What I really want is a list of what's in our file library, each document with my/someone's note on why it's relevant: "good info on [keyword] + faint connection to [keyword]", "follow-up to [link to previous paper]" (can I have links?), "total fluff but good pictures"). I'd like to avoid entering everything by hand (author, title, journal, dates, keywords, etc) but most key is that the comments don't get treated as secondary useless information - I'd like to be able to do searches not just by author/keyword/etc, but include searching my comments, and ideally the entire document text.

Document files are currently on a shared drive (windows) and probably won't need to be used outside of people who can access that shared drive. I want to avoid anything that involves uploading them to a web server with a 5-second pause after I change each piece of information. First threat from management was "but we're using Sharepoint already, do we really need new software?" - I find it absolutely infuriating; everything uploaded needs to be checked in individually, too much check-in/check-out all around, very clunky interface to add information, slower than molasses. If there's a server-type application that is infinitely superior to Sharepoint I'd consider it, but that's made me leery of web-based interfaces.

Project manager had previously worked with a team (in Europe) that used Citavi, so (when we point-blank refused to use Sharepoint) immediately said that's what we should use. Without disparaging management in a public forum, let me just say that I would like to hear some voices of experience before I get us locked into a particular software package. I get a lot of hits for "reference management software": is this the kind of choice I need to be careful of, or is it one of those situations where pretty much anything would be adequate, so I don't have to stress?
posted by aimedwander to Computers & Internet (11 answers total) 1 user marked this as a favorite
 
EndNote is somewhat of a standard in (the science part of) the academic world. It's got hooks into other software, so it's easy to insert references (from a database of sorts) and prepare reference lists as you require. Disclaimer - I don't use it (most of my colleagues do), but mostly because I'm old and started out submitting papers I wrote on a typewriter ... I should really upgrade to the 20th century ...
posted by zomg at 9:50 AM on October 28, 2011


I use Aigaion to manage a bibliographic database. It is open-source and Web-based (so you can throw it on a server, and many people can access it). It can keep track of both bibliographic data and the PDFs themselves.

It mostly works OK, but I'm not in love with it. It has some big problems dealing with authors with special characters in their names, and import/export from weird formats doesn't always work right. You occasionally have to be comfortable with MySQL and PHP to get it to do what you want.

A couple of years ago I wrote a plugin for the Trac project management system that searched for BibTeX files in your version control repository, and let you cite them in the Wiki and produced nice bibliographies. I never cleaned up and released the plugin have always meant to. Trac is pretty great and I'd check it out regardless of what you use for the bibliography.
posted by miyabo at 10:15 AM on October 28, 2011


This Wikipedia chart has been helpful in the past. Not sure if your biggest concern is managing a lot of documents, or simply making those documents available for any and all who need them...
posted by Wretch729 at 10:17 AM on October 28, 2011


Response by poster: I was just looking at the EndNote website, and I must be thinking with the wrong vocabulary, because I couldn't see what part of the product description said you could save comments about papers. But I can't see how it would be super-popular if you couldn't do that.
To me, the most important part of this software is not to be able to export a bibliography into a MS Word document according to the Chicago style guide, or to do a search and locate J. Smith's paper from the early '90's, but to look at all the papers by J. Smith and see which one has the virtual post-it note that says "Colleague A supplied this, interesting graph of X trends over time, but paper mostly talks about irrelevant Y and Z - (me, nov2011)"
Is this something that any reference management software out there has, thus there's not a column for "stores notes well" in the Wikipedia chart?
posted by aimedwander at 11:15 AM on October 28, 2011


Give Zotero a try... free and open source
posted by mfoight at 11:50 AM on October 28, 2011


Mendeley is my favorite thing. It lets you share reference libraries among collaborators, and notes added to those references are shared and saved. I don't know if you can link to other papers in those notes.

Tags/keywords work great for sorting papers and finding them again later, and there's also a folder system if you like folders. Searches work not only on title/journal/author etc, but on your comments and on the text of the document (assuming the document is OCRed).*

When you add papers, Mendeley is usually good about scraping the metadata and filling in the bibliographic information automatically. Again, this works much better with an OCRed PDF than with an ancient one that's a flat image.

Mendeley Desktop runs on your local library of papers (stored on your own server) and can also sync those papers to the web. (It syncs your bibliographic information to the web regardless, but you can opt to have the actual files synced as well.) I've never been slowed down by the web sync. I'm not sure how to set it up so multiple users can access a Mendeley library on your own server as opposed to Mendeley's server, though.

Oh, yeah, and it's free, unless you want to buy extra web storage from them.

*There's two commenting systems in Mendeley. You can leave notes about the paper in general (they appear in a sidebar); these are searchable. You can also leave annotations in line with the text of the PDF; these are not searchable as far as I know.
posted by pemberkins at 12:48 PM on October 28, 2011


Zotero allows you to search notes, citation info (like author), tags, and many other properties, which you can also manipulate. It allows you to share a library across multiple stations, and has support for multi-user shared libraries if you don't want to have everyone use one account. It allows you to store files locally, and to sync them to either your own webDAV or zotero's limited size free one. I think that might be an important feature for business since it will allow you to keep the files off public servers, though I do not know if it handles encrypted transport (I suspect it does).
posted by a robot made out of meat at 1:44 PM on October 28, 2011


Example screenshot
posted by a robot made out of meat at 1:53 PM on October 28, 2011


Response by poster: Thanks, all - this has been helpful. I'll check out the free versions of a couple of options, as well as Citavi, so I can form a reasonable opinion before I ask for money to buy the full version of anything.
posted by aimedwander at 9:57 AM on October 31, 2011


I find it absolutely infuriating; everything uploaded needs to be checked in individually

That is not true. That depends on the specific configuration of each specific document library - if you want, you do not need to require check-in's at all - yet you can still have versioning, metadata, etc. You can bulk-edit properties for documents. You also realize that there are far more collaboration features that are built-in, right? So - you can setup alerts (for yourself or your team) to be notified when someone changes a document, you can setup a workflow to enfore your processes. You can expose your document library via an RSS feed. By default both document contents and metadata are searchable. You can synchronize a document library with your Outlook client for quick offline use.

Do you even know about SharePoint's WebDAV (who do you think assisted with that standard and has the most-compliant implementation?) and/or "Explorer View" capabilities? Or heck - if you are still concerned with speed, but using 2010 - there is the SharePoint Workspace client, which allows you to do everything locally (and offline) (including edit metata) and then synchronize with a SharePoint site.

Hell - if you are using 2010 - there is even the concept of "Document Sets", which can aggregate a group of disparate documents and then provide over-arching metadata, process, etc.

It sounds like both yourself and your IT/management team don't understand the capabilities of software you already own.

So - instead of buying yet another tool to maintain, support and learn - why not learn the tools you already own? I am fairly certain there are courses, vendors and partners you could find locally to assist you.
posted by jkaczor at 6:51 PM on November 10, 2011


... And if it is taking 3-5 seconds to respond, then it is either broken, or under-resourced (which doesn't bode well for any server-side solution)...

You realize, that there are valid business reasons to store important data on servers, right? It is not "your" data.
posted by jkaczor at 7:00 PM on November 10, 2011


« Older My body decided it hates food. My mouth disagrees.   |   Help me pick boots please. Newer »
This thread is closed to new comments.