Thursday, June 19, 2008

Is there a one stop shop for (good) science videos?

My advisor has recently become enamored with the idea behind SciVee, which is essentially a place where you can view video blurbs of people's research, but is convinced that nothing short of YouTube will catch on. His plot for world science domination is for all of us graduate students to tape ourselves talking about our papers for a few minutes and upload the videos to YouTube, with dramatic increases in paper readership sure to follow.

While it's undeniable that YouTube is by far the most popular video server and thus wins by breadth and pure viewership, one could argue that SciVee and JoVE provide a service by being specific - one features research blurbs, the other features video explanations of protocols and experiments. You can be assured that the videos will be high-quality and reputable, at the very least. But there are also many videos that may not fall under those two categories but are still interesting to scientists, or those interested in science.

The Inner Life of a Cell, various TED talks, and this demonstration of cornstarch physics come to mind as some science-related videos that I've enjoyed recently. Maybe even science humor. What YouTube has going for it is precedence and near monopoly, but quality control is dismal and it can be impossible to find good, engrossing science videos (their "Science and Technology" category is mostly dominated by technology - software, hardware, gaming, etc - and the science offerings are hardly scientific).

What I'd like to see is a way to aggregate high-quality science-related videos, categorized by type (protocol, experiment, research/paper promotion, cinematic, humor, etc - how would one categorize the TED talks?). Because sometimes a video is worth a hundred readings of a paper or protocol, because we're all just curious about so many things, and because we all need a bit of reinspiration every now and again.

Is there anything like this that exists now?

Probably old news by now, but JoVE's got this going for it - published videos will be indexed on PubMed! Will the day come when we have citation rates for videos, and can list number of hits as a bullet point in our CVs?

Wednesday, June 11, 2008

Would open science profit from a non-profit?

In my first foray organizing a formal meeting (the PSB workshop), I've learned that pretty much everything comes down to money. Having a successful meeting requires getting people to attend, and getting people to attend often involves money. Getting money to allow people to attend can even require money (for example, publishing costs for conference proceedings).

One idea Cameron's mentioned a few times during our fundraising discussions is the Open Science Collective, and though he hasn't fully described what this is to me, I get the impression that it would be some abstract entity comprising individuals, organizations, resources, and activities - some larger body that would provide support for open science-related endeavors. At the very least, I think it would provide a means to fund raise separate from any particular event such that we (in the collective sense) could act independently. By this I mean that the OSC could support individuals to go to meetings, sponsor meetings like the PSB workshop, or even organize its own meetings.

The idea hasn't really been taken further yet, with whimsical t-shirt ideas pretty much our only tangible revenue strategy so far. But to give the OSC a step towards reality, perhaps it's time to start talking about building a non-profit. From a quick glance through of one website's guide to starting a non-profit, it appears that we need to create a mission statement, obtain a board of directors, and eventually file for either incorporation (plus maybe other things, like tax-exempt or tax-deductible status). This is about where I start getting fuzzy on the details, so if anyone has experience with non-profits or foundations (I don't even know what the difference is) please feel free to enlighten me!

At the most basic level, it would be nice to have some external entity with a bank account into which we could funnel funds that we (again, the collective we) could apply towards open science activities. If that t-shirt shop ever sells anything, the proceeds should go into that account. What the best way to accomplish that is, I'm not quite sure. If it is to start a non-profit, then perhaps the first step is to see who else is interested and start drafting a mission statement together?

According to, a mission statement should cover the purpose, the business, the values, and the beneficiaries of the organization; i.e. the what, how, why, and who/where. This can be accomplished in one line but can also be expanded into paragraphs. But let's start small - here's a stab at a one liner mission statement:

"The Open Science Collective is an international and interdisciplinary [non-profit] organization that promotes open exchange and collaboration in science, and provides resources and support for the advancement of open science."

So some concluding questions: Does it make sense to have something like the Open Science Collective? If so, what should it be and how do we get there? What would be the mission statement of the OSC? If not the OSC, what do we need?

Monday, June 9, 2008

Epernicus - 2 steps closer to a science exchange?

This morning I received an email from Mikhail Shapiro inviting me to join Epernicus, a science networking site which he helped launch. He'd visited the blog and (quite rightly) deduced that I would be interested in trying out a web community that aims to do some of the very things discussed here regarding a science exchange. So after the slightest hesitation (so many networking sites, so little time...), I decided to give it a go. I populated my profile (inspiring me to update my CV), uploaded a picture and a shameless plug for the Open Science Collective shop, checked out other people's profiles (has anyone invented the term "profile envy" yet?), clicked on a little link next to the "Launch BenchQ" icon for "What is BenchQ?", and generally muddled around until I ran out of narcissistic things to do.

So far, I'm rather pleased with Epernicus. It has a nice, aesthetically pleasing interface (as do SciLink and Laboratree, which I've mentioned before) and I didn't run into any bugs like others have experienced with other networking sites. When you first sign up and fill out your affiliations, it automatically populates your network with members of Epernicus with the same affiliations - in my case, 2 students in my PhD program and 60 or so people from my university. Once you've signed up, you can create groups, send people messages, and invite others to join - fairly standard social networking fare. The two distinguishing features of Epernicus, however, are Assets and BenchQ.

As part of your Epernicus profile, you can list your "assets" - topics (Phosphoprotein signaling, gene expression analysis, etc), materials (anything from chemicals, cell lines, equipment, and kits), and methods (protocols, algorithms, techniques) in which you can claim knowledge or experience ranging from novice to expert. Through their search bar, you can identify people who know something about your topic of interest. Taking this one step further is BenchQ, their version of a discussion board, which lets you post a question to a person or network. The questions can be geared towards methods ("How do I stain live cells?", "What is the best software for hierarchically clustering microarray data?") but can also be requests like "I need to do assay Y on protein X. Does anyone have purified protein X available?" or "My [equipment] is broken, can I use someone else's this weekend?"

I posed two of the major issues that came up during the science exchange discussion to Mikhail. Issue 1 is Motivation - what is the incentive to help in science? Issue 2 is making connections rapidly and effectively - given the passive nature of BenchQ, were people getting answers to their questions? Issue 1 can be mitigated by having networks of people who are more likely to help each other because they have an actual professional relationship. Regarding issue 2, Mikhail said that he'd seen a number of successful interactions within the MIT and Harvard networks, and that almost every BenchQ question gets answered, which sounds like a great start. He also mentioned that an algorithmic notification targeting system is being developed, sort of like what human supernodes might do, which I'm interested in seeing when it comes out.

One thing Epernicus does not seem to support yet is some kind of "Projects" feature.* This is kind of a loose concept but generally it could be described as a place where you could list projects you work on and include pages with more information or interactive areas for each project. Project management, sort of. A mini-blog type thing could also be a useful feature, though at some point you need to decide what features contribute to the overall scientific mission and which ones distract. If Epernicus is more about science networking and less about the actual collaboration and project management aspects, maybe it's fine for it not to go down that road.

For a site that does go down that road, Laboratree tasks itself as a research management system, so in addition to the shallow networking aspects, it also allows you to upload documents, write blog entries, and create projects with group members, messages, and project-specific blogs.

In my limited experience, I've been most impressed with Laboratree and Epernicus out of the various science networking sites I've seen (though Laboratree does have some bugs of its own - notably the search bar, and empty Help page). They independently provide different aspects of the science exchange idea and do so with style without being over the top. A question now is whether we want to have everything but the kitchen sink in one website? Or do multiple networking sites (like Laboratree for research management and Epernicus for question-answering) serve their different functions? Is it simply an issue of critical mass?

Check out my public profile at Epernicus.

* Update: After posting this, Mikhail pointed out that the "Group" feature does accommodate some project management through document uploads and a "whiteboard". Thanks for the clarification and apologies for the lack of research on my part!