Wednesday, August 5, 2009

Tools for Research

I'm already excited to get going with my new area of research for my PhD. Although nothing is set in stone, I would like to work with augmented reality, and am interested in applying the technology to educational entertainment. This will allow me to take a more people-oriented approach and learn a bit more about the psychology of human-computer interaction. I've started to collect stories, links, and papers on the topic, but wasn't sure how to best organize them all.

I started by putting some links together in Google Notebook, which is no longer being developed, but still available for current users. This is fine for some basic, unchanging links, like lists of software, conferences, etc, but isn't very good for organizing academic papers. I tried this during my Masters, and it's hard to make notes and find info using this method.

So I set out on a search for research tools. I got a few ideas through some basic searching, and got some pointers via my Twitter contacts. For example, check out this detailed article at The Byte Baker. A lot of the promising software seemed to be Mac only (I use Windows), but I eventually stumbled upon a few that seemed to fit my needs.

I found the following software that is either for Windows or is cross-platform:
Some of these are not research-specific, but general note-taking programs. Evernote in particular looks really slick, and apparently works well on all devices, including iPhones. It's not quite what I was looking for in terms of organizing papers and such, but could be worth trying for other applications in the future.

JabRef is a reference database that I used during my Masters in conjunction with LyX to write my LaTeX papers. It runs in Java and is very easy to use. Highly recommended for making bibliographies, but not my first choice for general organization of potentially interesting papers.

Idea Rover "absorbs and crystallizes your research ideas and new sources into outline-structured notes, releasing your brain from monotonous switching and searching for relevant information." I haven't tried this yet because it seems that it would be more useful at a later stage, but I think I will give it shot eventually. At least it is targeted to PhD research specifically.

The one piece of software from this list that I have installed so far is Mendeley. It is "a research management tool for desktop & web." There are a few features that I really like so far. First, it has a PDF viewer that, while it seems to crash a little too often, does allow you to annotate your PDFs pretty easily. If you don't have the PDF, that's ok - you can still track the source and write notes about it. It connects to online citation sources to fill in missing info and can manage your PDF files if you want it to. You can synch your sources to your online account, making them available anywhere. There is a social network of academics that you can use to share your sources with if you want to (could be useful for a particular research group). Finally, it allows for some decent searching capabilities within your source list. So far a winner for me!

If you know any excellent tools for organizing PhD research and online PhD program research (at any stage of the game), tell me about it!


Stampie said...

The Firefox extension Zotero allows managing a lot of sources.

My favorite feature is that it integrates into the search processes very well, as new entries can be created about the current page very easily.

It also allows collaboration with others by using group libraries - but I haven't tried that feature.

Rick Davis said...

Zotero has made my life as a graduate student so much easier. It is easy to set up and use, and I can easily sync my citations between different computers. They also have an openoffice plugin for in-text citation.

It looks like Mendeley is very similar to zotero, only it isn't open source. I would be concerned about them changing their "free" policy in the future. So much of your work will be wrapped up in your papers that if they decide to start charging exorbitant amounts of money for the software, you wouldn't have much of a choice but to part with your dollars. This happens all the time in research software. VectorNTI was the latest program to suddenly stop being free and start charging hundreds of dollars per year for software that has always been "free".

rvidal said...

Hi redavis, as part of the Mendeley team I can assure you that whatever is currently available for free at Mendeley will remain free. So, feel free (no pun intended) to use Mendeley to organize your research articles :)
Let me also take this opportunity to state that the premium features that will come along at a later stage will not be priced with exorbitant fees.

Gail Carmichael said...

Thanks for all the comments so far. Zotero definitely sounds pretty popular as well.

Ricardo: What sort of premium features would Mendeley eventually look at having?

_ said...

I've been using a bit of a homebrew system that combines several tools throughout my Master's[1]. I don't claim that this is necessarily the best system (in fact, I'm sure it's not), but it's worked well enough for me, and it might be useful as a discussion point.

The key element of the system is a "universal ID" that I assign to every interesting paper I come across. I've used the format "Ref0001", and simply number the references sequentially. This ID is used to co-ordinate across the four places I keep info about the references:

1. Hard-copies of the papers get this number marked on the first page using pre-printed labels.
2. Citation information is kept in a plain BibTex file, maintained with gVim, with my ID used as the citation-key.[2]
3. PDFs of the article are kept in a single directory, named according to the ID.[2]
4. Links to the article source, notes about the article, etc. are kept in Evernote[3]. Each entry in Evernote is titled with the ID, and the body of the note is the article's title and abstract, followed by my comments. I also make liberal use of the tagging.

In practice, my "entry point" into the system is usually Evernote (or occasionally the hard-copy).

The really huge advantage of this system, I've found, is having a single, simple, unambiguous identifier for each paper. All of my notes use this system, so I never have to deal with finding "Note to self: check methodology section from Smith et al.", and wonder which one of Smith's paper's I meant. When it comes time to cite something, I don't have to do anything but type \cite{ref0000}. I've made my Evernote database accessible in read-only html format (a built-in feature), so I can use my IDs when discussing things with my supervisors.

The other advantage is flexibility. I'm not locked into anybody else's idea of "how things are supposed to work". I can save odd-ball formatted supplementary materials right next to the pdf. Additional links (PubMed ID, author's page, etc.) fit nicely in the Evernote entries.

The big disadvantage, of course, is that the info is a little more spread out, and not automatically co-ordinated. It takes a lot more than a single click to save a reference. It hasn't been a big problem for me, especially after getting used to the habit, but everyone's different.


[1] Technically, I haven't been using the system as described all the way through, since it's evolved organically over time.
[2] These files, along with the rest of my research and writing, are backed-up and synchronized to my home computer daily, using my Palm Lifedrive.
[3] Originally Google Notebook, until Google abandoned it.[4]
[4] Is it a sign that I've been in academia too long that I'm using footnotes in a comment to a blog post?

Hélène said...

I'm also a Zotero fan. I used Evernote for some time and liked it too.

Saskia said...

I'm not a PhD student but for my papers I like Citavi. With it's help I can search databases, organise thoughts, plan tasks and create a bibliography. It's very easy to use, too and looks nice.

jyotsna said...


I was fortunate enough to attend ur session on tips and tricks for keeping research organized(GHC09). I have begun to use Mendeley and its really helped me a lot. Could u tell about some tools for performing analysis e.g. , i am trying to develop a DPI based forensic analysis system for forensic analysis of network traffic(using grids) . so if i would like to evaluate that if the forensic analysis is done on grids how better it is in terms of cost/time as compared to the case if grid are not used for the purpose.

Gail Carmichael said...

jyotsna: It wasn't actually my session - I just blogged it. :) Also, I don't have an answer to your question.

jyotsna said...

Sorry,for the inconvenience caused due to some confusion. Thanks anyways.

Post a Comment

Comments are moderated - please be patient while I approve yours.

Note: Only a member of this blog may post a comment.