Document management for a lab group

Here’s how papers for scientific journals are written in most lab groups. Someone, most likely the person who generated most of the results, writes a draft. They then email the draft to the other authors. Changes are suggested and made. The paper bounces between email inboxes until everyone is satisfied and is then submitted.

This is a terrible way to do it for all sorts of reasons. I’ll highlight one major problem – nobody except the person working on the most current revision is able to keep track of progress. Very often an author is busy with other tasks, days or weeks go by, when asked they’ll say “yeah, I’m working on it”. What you really want is for all the authors to have access to the most current version and for them to know when it was last updated.

Now in an ideal world, we wouldn’t be bouncing multiple versions of a Word document between inboxes. We’d be writing in plain text, marking up using LaTeX and storing documents in a versioning system such as CVS, with online access. Most researchers in biological sciences don’t inhabit that world (though many in physical sciences do – go figure). However, they can handle the idea of using a web server to upload, download and view documents.

So – the point of this post is that I’m investigating document management software. It doesn’t have to be complex or feature-rich; it doesn’t even have to include versioning. It should just be a web-based application where authors can login, upload/download/view documents (mostly Word and perhaps PDF), see when a document was last modified and maybe leave a comment (“get on with it!”). I’ll be combing Google and the open-source repositories for possibilities. In the meantime, if anyone has experience or favourites that they’d like to share, feel free to leave a comment.

12 thoughts on “Document management for a lab group

  1. Deepak

    Great question. I come from a generation where document management systems didn’t exist. The biggest challenge will be bibliographies. I suppose you want a good check-in/check-out system. Unfortunately I have only used commercial or home-made DMS’

    You might want to take a look at this. At worst, there is a list of open source DMS options.

  2. Pedro Beltrao

    I have been using Google Docs in my current collaboration work. For me it has almost everything required. The only thing that I am missing is a chat box (like the have in google spreadsheets) and a way to deal with references. Zoho also has a online document management tool.

  3. nsaunders Post author

    I have been using Google Docs

    That’s a pretty good idea. You know, I’ve only just signed up for a Google account. When I last checked (a long time ago), their accounts were much more limited. Now it looks like you can do “Google everything”.

    Now I’ll probably spend several hours importing my life into my Google homepage. Thanks Pedro ;)

  4. alf

    Yes, Google Docs is pretty good for collaborative editing. For citations/references, I’d like to be able to invite a bot – connected to an online bibliography database/manager – to collaborate on a document, and have it automatically format the references.

  5. Deepak

    I’ve actually used Google docs quite a bit for document collaboration, but how do you handle citations and other material? If one could do what alf suggests then you’d be in good shape.

    Neil, my life is already my google personalized homepage (with multiple search engines :) ). Welcome to the club!!!

  6. Andrew Perry

    I’ve looked into this from time to time also, since I hate messing about involved in the email-edit-email cycle that goes on when collaboratively writing a paper. Rather than just having another way of passing back and forth documents (ie shared storage), I’d really like proper collaborative editing.

    I think a wiki is ‘theoretically’ a good solution which could be very powerful for collaborative writing of scientific papers, but typically wikis don’t integrate cleanly with Word/Endnote etc. Older academics are unlikely to change their ways, and even if they did they would certainly want decent offline editing. One project that is at least attempting to solve this problem is the UniWakka wiki (http://uniwakka.sourceforge.net/HomePage), which has OpenOffice import and export, and the ability to import a BibTex database of references and cite them. I think in reality, users would have to be very disciplined with their use of OpenOffice so as to avoid creating documents that don’t import to UniWakka correctly.

    I’ll be watching closely to see what you find .. :)

  7. chris

    Probably overkill, but I’ve been considering using an internal manuscript tracking system – like journals do. They track versions, support up/down load of files, etc.

    Ultimately, the main question you have to ask is whether you want to support parallel editing. If so, you need a text-based versioner of some sort. Tex format can be mostly converted into rtf, so perhaps the best solution is to use that? I’m assuming that trickier aspects of the document, such as display items, will be collated by the lead author. Even if they are not tex-savvy, there may be higher-level tex editors which will help them put these things together.

  8. Matthias Steffens

    Neil, thanks for this post, it hits a critical nerve. I agree that a plain text markup language such as LaTeX combined with a server-based versioning system and good doc comparison would be ideal for collaborative writing of scientific documents. Unfortunately, this isn’t an option for many people since, as noted in a previous post, the whole process “feels” to techy and is daunting and too complex for many scientists. However, I think that a wiki-like web-based software (that uses a simple markup language) could lower this barrier substantially.

    With regard to refbase, we have plans to better support the process of collaborative manuscript creation (which is a frequently requested feature). People could attach new manuscript versions to a database entry (which would be invisible to the public until its final) and include their comments. This would be more of a shared & commented manuscript repository that would be always accessible by all group members, but keeping it simple will help to get everyone actually using it. So manuscript editing would still be done offline (using your favourite tools) and the type of documents (.tex, .rtf, .doc, etc) could be freely chosen. Ideally, I’d like to integrate something like this with a versioning & document comparison system.

  9. nsaunders Post author

    Matthias: that all sound excellent. You have a project there that just keeps growing and developing. I’ll keep an eye on the latest features.

  10. Simon Baker

    Hi Neil – although its not open source, we’ve just started using “Basecamp” (http://www.basecamphq.com/?ref=free) for a project on surfactin. Unsurprisingly the idea for using this came from the physical scientists and not from me! The free version provides a web-based platform for what is essentially a closed chat room. I’m still getting used to it, but the functionality is quite good. I’ve yet to use its document system in anger, but I’ll let you know how things progress. However, there is still the problem of trying to get people to actually work on the document (perhaps there should be a web-button that allows connection of their desk chair to the nearest mains supply?) or a system for dealing with the deliberately obtuse…

Comments are closed.