Reading scientific publications
(This is a reprint of the original post)
Nowadays, I spend a lot of time on research tasks. On weekdays, I work as a Semantic Web researcher with a team. On weekends, I spend time on other projects, chiefly on the research for my dissertation; in this case, it is mostly an individual job. One of the most important tasks of any researcher is to read lots of books, papers and articles on the subject, and to keep track of them. I know there are some commercial software products that are designed for this. I haven't tried them. I use a simple BibTeX file managed with version-control (SVN). I know it is far from perfect, but it is handy. I used to manually edit the BibTeX file, but then I discovered JabRef, a fine GUI, and I changed my mind.
I always print the documents I want to read. After I've read them, I store them a filling cabinet. However, even if I try to do all my best to keep them organized, it is often quicker to find them on Google when I need them again.
I've noticed that I need to make a great effort to keep myself concentrated when I read these papers. From time to time, I find a breathtaking paper, but often they are just interesting. Of course, this is a subjetive vision, as most of them are not closely related with my research area, and therefore I tend to see them with a lesser interest. In addition, I'm not in the proper position to absolutely judge their scientific value (I still consider myself a newcomer). Anyway, I don't usually have any problem to focus on a technical subject (programming, for instance), but reading these papers is a completely different thing. To be successful, I need to be alone, stop the music and underline the main sentences of the text.