This document represents a concise summary of brainstorming on the relation of Web 2.0 to future scientific progress. See brainstorming notes.
The internet is undergoing a major change - from an original environment in which individuals post static information to a new environment where anyone can dynamically and collaboratively create, edit, and disseminate content. New search, aggregation and publishing tools are making it easier to find and contribute to the collective of information available on the web. This shift has been described as the switch from Web 1.0 to Web 2.0.
As these new technologies develop, science can both use these new tools directly and also apply similar concepts by analogy. A key example of how Web 2.0 technologies can advance biological research is OpenWetWare. OpenWetWare is a wiki on which researchers can share expertise, information and ideas in biological science and engineering. Inspired both by Wikipedia and MIT OpenCourseWare, OpenWetWare seeks to create a useful resource that relies on a community of users to keep the content accurate and up-to-date. We hope that OpenWetWare will foster enhanced collaboration among community members as well as provide a useful reference source for researchers around the world. The expectation is that these new technologies will change how scientists communicate their work and the way in which research is done. Such changes should accelerate the pace of scientific discovery and technology development.
In the biological sciences, the primary mechanisms for sharing work have traditionally been reference books, journal publications and personal communications via conferences and invited talks. There is a divide between those who have access to the equipment and expertise to carry out experiments and those who do not. Such a situation contrasts with computer science which has taken advantage of and benefited tremendously from more democratic platforms for dissemination like the world wide web. OpenWetWare represents an initial effort to decentralize and lower the barriers to information exchange among all researchers, be they professors, students or research scientists. It seeks to help forge a culture in which researchers openly share their experiences thereby reducing needless duplication of effort and improving the quality of the work.
Initially called Endipedia, OpenWetWare began in May 2005 as a means of sharing and archiving information among members of the Endy Lab and Knight Lab. Its utility in sharing protocols and providing a forum for brainstorming ideas quickly became apparent. In June 2005, Endipedia was renamed to MIT OpenWetWare in an effort to include more research labs at MIT. The hope was that the wiki would provide yet another mechanism for interaction with local labs. By August 2005, there was sufficient interest from individuals and groups outside MIT that the wiki was renamed to OpenWetWare such that any researcher or lab in biological science and engineering would be encouraged to participate.
Current Status & Facts as of 10 Nov 2005
- Users = 265
- Administrators = 29
- Pages = 2725
Today, there are approximately 15 labs on OpenWetWare, each with varying degrees of participation. Some labs such as the Grossman Lab only use OpenWetWare as a place to have their website. Other labs, such as the Silver Lab, Endy Lab, and Synthetic Biology, have taken fuller advantage of its capabilities by using it as a website, collaborative idea generator, and a place to share and standardize protocols and equipment usage. For example, the synthetic biology website is just a dynamically updated page that just takes content from the wiki page. Looking at their site you can find a number of discussions that they are having (e.g., BioBricks abstraction hierarchy and Semantic Web Ontology), initiatives they are running (e.g., The BioBricks Foundation and Synthetic Genomics Study), projects they are working on (BioBrick Parts for Plasmid Engineering and Rebuilding_T7), classes/competitions they are running (e.g., iGEM, BE.180, etc.), and resources/news they have compiled (e.g., Online Tools and a FAQ). In addition, other labs are beginning to disseminate and possibly standardize different scientific protocols (e.g., see DNA Ligation). Finally, people are beginning to put usage notes, simple control experiments, and data interpretations concerning laboratory equipment online as well (e.g., see the Endy Lab's Victor3 Plate Reader).
If the first 6 months are a barometer for future success, then we have much to look forward to. First, continuing advances in wiki technology will allow it to be a more useful tool. For example, static versions of history pages can assuage those worried about vandalism. Generally, many of the features seen in Web 2.0 technologies are inherent in the motivation and form of OpenWetWare. For instance,
- the network effects of having a large base of users
- value generation by the users themselves
- only some rights being reserved
- perpetual beta status and ease of use of the site to reduce barriers to content creation
- having a community of cooperation
If OpenWetWare follows path of other Web 2.0 successes, we can expect that as the site becomes useful to increasing numbers of people, the site will become ever more useful, and thus draw even more people. However, the greatest uncertainty in the future of OpenWetWare is likely not technological, but in determining whether such cooperative efforts actually increase the production and pace of scientific knowledge.
Responses and experiences
Our experience with OpenWetWare has been overwhelmingly positive. In general, labs have been excited about participating and have taken good advantage of the customizable nature of the wiki. Some labs utilize the site as simply an easily maintainable website, others post lab meeting notes, journal club discussions, and even online lab notebooks. In particular, the wiki format democratizes contribution, rather than centralizing information distribution through the PI or webmaster.
When describing OpenWetWare to labs we have also encountered some reservations. Most commonly, people are concerned about negative editing or vandalism of their postings. However, the fact that we have not had any such incidents thusfar and the accountability that comes with assigning logins for users, have assuaged many of those concerned. Additionally, there is a strong resistance to posting research progress and topics prior to their publication due to fear of being 'scooped'. This is an unfortunate limitation based on academic community norms, rather than a limitation of the technology. We hope to start to change these norms by demonstrating that sharing information about research in progress will be far more likely to lead to collaboration than to competition.
Moving forward there are many areas where Web 2.0 approaches can dramatically enhance the rate and scope of research in biological science and engineering. We have touched on how OpenWetWare can be used as a tool for generating ideas, sharing know-how and techniques to generate that data, and then analyzing the results. What we have not talked about is how OpenWetWare and other Web 2.0 technologies could be used to better publish and evaluate scientific work. Currently, the mechanism for distributing information in science is very rigid. While there is certainly a place for peer-reviewed, brief, and highly polished research summaries there may also be a place for more rapid though less reliable information streams. For instance, while papers typically present results 0.5-2 years old, a 'lab RSS feed' could stream results on a weekly or daily basis. Secondly, we can implement mechanisms to better evaluate published works by adopting an Amazon style reader review process, where people post and rate each others reviews. We can imagine many other ideas that could be implemented by adopting Web 2.0 approaches. Finally, with respect to the biological sciences, we do not think that technical limitations are preventing increased cooperation. It is worth considering how the reward structures in biology may slow the adoption of cooperative research.