Collaborative Wikis and Research

Over at Machine Learning (Theory), the Learner points to a Scientific American article on Science 2.0 which discusses various efforts in bringing scientists into the 21st century, and scientists reluctance to openly discuss their research in progress in public forums. Sometimes I wonder what would happen if I started blogging about my own research. First of all, I'm pretty sure it would bore a large number of people into a deep comatose sleep from which they would never emerge. On the other hand, I'm not a very smart guy, so exposing my work to the vast power of the intertube's collective neuron pool seems like a smart way to advance my own personal empire building, err... I mean research.

In a related note, I have recently started using a wiki not just for my group's webpage (see here.) In addition, I have also started a different wiki where I am storying working papers for collaboration. By combining mediawiki with this extension I can put LaTeX documents on the wiki, and compile them from the browser to pdf. So far the system isn't optimal, but I do like it. I keep the pages password protected, so no, I'm not open science-ing at all. But it is nice to have an online forum for sharing documents. The one thing I like about it, as opposed to a versioning software solution, is that it is always nicer to have the structure of a wiki and the ability to edit in the browser. The one real hangup is that because you just enter LaTeX into a regular wiki entry box, your text editor is basically pretty crippled (oh how I depend on you Mr. Syntax highlighter.) Another issue is offline use: it would be nice to be able to check out the entire document when needed.

Okay, well back to the wiki to finish writing that grant.

More like this

The one real hangup is that because you just enter LaTeX into a regular wiki entry box, your text editor is basically pretty crippled

Try this. Not only does it give you a more Wiki-like (Markdown, specifically) composition syntax, it exports to LaTeX with the click of a button.

And it has other interesting features.

I have actually considered trying to do something like this, partly because I have such an enormously difficult time finding people interested enough in my work to give it the time of day, let alone a thorough reading and critique (notwithstanding the one person who shall remain anonymous at this point, but who does read my papers fairly carefully). I'm not certain if it would help, though. Note that to some extent, Quantalk.org is aiming to do something similar to this via their open peer review (OPR) process.