Scientific research, by nature, should be open. All knowledge accumulated via generations of scientific research is the common wealth of all human beings and should be shared and passed on for our future endeavors. As such, the practice of open standards in sciences, at least basic sciences, should be in the heart of every scientist.
Unfortunately, many other concerns and issues seem to be diverting sciences from keeping the very nature of openness. Concerns related to national security, intellectual properties, copyrights, and credits, are all valid, yet they should not become barriers to block open access to science.
Here are the list of things, I think, we scientists can do better to make science more open among ourselves and to the general public. I hope to discuss some of them in more detail in my later posts.
Open arXiv-like publication service with overlay open journals.
This is really the heart of open sciences. In the information explosion era, a well organized repository for original research works and the corresponding dissemination will do wonders for the scientific advancements. The digital e-print server, arXiv.org, has done an amazing job so far on these aspects, at least for the research in physics.
However, it is far from a complete service. There are still no arXiv-based overlay journals. Many components in research dissemination and publication are not open or transparent yet. For example, the journal review process is still closed within a few individuals. Fraudulent papers keep popping up in top scientific journals from time to time. The ideal service will be something driven and regulated openly by all members of the scientific community instead of a select few. It could be realized by extending arXiv with a well-thought review/credit system.
Open funding opportunities for individual scientists.
Money talks, not only in a business world. The scientific community can indeed learn some from the business world. For a healthy capital economy, we need not only steady giant companies but also more vibrant startups. As a matter of fact, many recent economic growths were initiated by some startups. Many of these startups failed in the end. But a few of them survived and blossomed into future’s giants. The risky nature of the startups also offers the innovation needed for real economic breakthroughs.
We do have large collaboration groups in sciences similar to the business giants. But where are the “startups” in our community? Do the newly hired tenure-track professors count? Maybe, they are indeed smart people. But our evaluation system is not really open. Why a person got the tenure-track position may be strongly related to where she/he is from, who sent the recommendation letters, and the foreseeable future success. Do any universities want a risky hire? Do any new hires get encouraged to risk their careers before they get tenured? See, our current evaluation/funding system tends to keep these smart young people spending their best time in a safe route.
One possible solution would be a funding source regulated openly by all members of the scientific community. Any member can vote for any risky proposals by other members. Hopefully it will mimic how the venture capital fosters the startups in the business world.
Open data to be shared within the scientific community.
This is critical for the integrity of sciences, in particular for certain disciplines. Any publication of original results should have their data openly accessible to other members in the field. It could be via a centrally hosted data server. Either the institutions of the authors or the funding agencies should ensure such an infrastructure for data storage and access. Any publishing service should make open data as a requirement for publication.
Open source codes for scientific computing tools.
Computer software developers have done a much better job in this respect than other scientists. For instance, linux is one of the best examples as a community-driven platform replacing proprietary systems. Physicists have started doing a better job in sharing their computational tools. A lot of large scientific software tools have become open-sourced. Now we just need to make it a standard. For example, if a scientist uses her/his own code for a publication, she/he has to make the code open source. In this way, we don’t have to re-invent the wheels all the times. If people of the open-source community identify bugs later on in the code, we know to how to modify or understand the results from the previous publication. That should be how we forge forward in sciences.
Open facilities for proposals rated by the community.
Most of user facilities currently adopt a policy of using a Program Advisory Committee (PAC) who rate all the incoming proposals and decide who get the time at the facilities. It is not a bad model. But it is certainly not open enough. Could we have open proposals instead? The community could rate/review the open proposals together. This will prevent an elite circle taking over most of facility time behind closed doors. PAC could still play a role (e.g., soliciting reviews from other experts) like editors for journals. Another important aspect is that some part of the facility funding should be tied or allocated to selected proposals. In other words, the funding agencies should fund the facilities more to carry out the experiments, at least partly. This can ensure good proposals will be better funded. If there are too many demands and the best facility could not do them all, other similar facilities could take on some of unselected proposals without invoking (repeated) new round of writings and reviews. That is a more efficient way to utilize the community resources.
Open meetings with topics and keynote speakers voted by the community members.
The funding agencies could establish portals for organizing meetings with all community members. All members can propose topics and nominate keynote/invited speakers. The top rated topics and speakers should be automatically considered in the meeting. The organizing committee may still keep some of its power to select other (sub-)topics and speakers. The idea is to have best results to be reported in the meeting and to prevent an elite circle from taking over all the talking powers. Another issue is how to make sure that the hottest topic does not dominate all meetings. One way to solve it is to introduce a rating system by taking into account how often the topics are addressed in recent meetings. This can ensure a warm enough topic can have its turn for its own meeting.
Epilogue
The biggest strength of human beings is our openness to each other. Our economy prospers because of the openness of capitalism. Openness involves dialogues, communications, and mutual understandings leading to a healthy society. On the contrary, many negative phenomena like wars, hatred, and injustice are somewhat related to closeness or secrecy. Physics has told us that we need an open world to reduce entropy and disorder. Now it is time for the scientific research community to set an example on how to build a truly open world for our society.
No comments:
Post a Comment