Yesterday I attended a discussion with Michael Nielsen and the GT faculty, where open science was the subject. We covered various venues for promoting collaborative science projects. Some were completely new to me. Let’s cover a few examples that were discussed.
The ASSET India Foundation aims to educate and provide better opportunities for Indian children of sex workers through technology training. However many parts of India, especially rural areas, have unreliable power supplies that are necessary for technology. They needed a solution to provide reliable power and networking. They used crowd-sourcing with InnoCentive to solicit the best solution. In the end, a software engineer with a life-long interest in solar power had the best solution. Crowd-sourcing allowed the foundation to exploration options outside of their domain and were able to find a person with a unique combination of skills that best suited their needs.
Perhaps the most popular examples among astronomers are SETI@Home and Galaxy Zoo. The first project uses idle CPU cycles of PCs to search for extraterrestrial signals and has been active for over a decade. The second project lets volunteers identify the morphology of various galaxies. Automatic detection of galaxy morphologies are not 100% accurate, and the human eye can sometimes pick out interesting objects better, especially when it’s something the algorithm is not expecting.
One very cool application of citizen science is FoldIt. It has transformed the science of finding the configuration of proteins that has the minimum energy into a game! The players’ scores increase as the protein’s energy decreases. Some of the best players have little science training but have a knack of recognizing spatial patterns.
Here we can see that open science that utilize both the human mind and their spare CPU cycles for the advancement of science. It just requires effort on the scientist’s side to develop a tool to leverage these resources.
Blogging and open-source software
Lastly, we discussed how ideas can rapidly mature in scientific blogs. The post can propose some problem, but the author only has a partial solution. (S)he then asks the community to give opinions and possible solutions. One person may not know the complete solution but an idea can spark to another idea and so on. This informal venue of developing ideas can be very productive and does not need the physical gathering of people, such as a conference.
You can clearly see this occurring in this blog post that discusses how such a collaboration would work. Throughout the comment process, the ideas are refined to produce a final set of guidelines for massively collaboration science.
This discussion then evolved into how blogging is perceived by scientific peers, especially in the hiring and tenure process. Authoring and commenting in scientific blogs don’t necessarily make it into CVs (yet). Is it best to spend your time blogging or writing a publication? Both have scientific merit, however the latter is mostly favored, especially in Physics. I believe this needs to change if we want to participate in more open science. I brought up a similar point in supporting open-source software through documentation, mailing lists, and IRC chat rooms, which has similar problems as blogging. We don’t necessarily hire people to support open-source software in science, and the burden usually falls onto the core developers to maintain the code’s presence and usability.
We also discussed the success of creating a focused wiki with the latest developments in a particular field. He highlighted several wikis that were failures in the sense that it did not obtain critical mass to develop the knowledge-base. This is a common problem that blogging, open-source software, and wikis share, and is vital for the success of collaborative projects and resources.
For more information, please read Michael Nielsen’s original post on the subject.