I waste an inordinate amount of time, but today I decided that if I couldn’t focus on work I’d at least do something useful, so I made a template for publications. Of course I’m using CSS, so the style is completely configurable to your colorscheme preferences. Thanks to John Owen’s advice page for the reference to CMU’s Robotics Institute style, which I stole and CSS-ified. I’m a scripting ignoramus, but I think with a little work I’ll be able to make a PHP script which will let you enter the information in and then generate the HTML for you automatically. Naturally, it’s not the perfect format for all people, but it has enough in its barebones-ness for the “busy researcher.”

John’s point is well-taken, which is that if you want people to read your work, you should make it easy to get and provide them with things like the BibTeX (a citation tool used with the LaTeX document processing system) and the abstract in text. Another advantage is being able to give supplemental information online that can’t fit in the paper, such as code, extra graphs and figures, and even an annotated bibliography. I would be so happy if every paper with simulation results in it provided the code as well. I’m convinced that a large percentage of researchers aren’t willing to put their balls on the line by providing their code (what if it’s buggy?), which is why it’s not done more often. I think that if the paper is written so that the results can be replicated then the code should be given to make that replication easier.

Research is not the fast and exciting world of financial investments, where fortunes turn to ashes in the course of a few hours and then rise again like an ascendant monetary phoenix. Taking the 10 minutes to provide a nice presentation for your work will faciliate its spread and make new work easier. I’m going to package up the stylesheet, template, and whatever else I come with and maybe evangelize a bit. I’m sure if it’s no harder than maintaining their existing publication page then people would do it.