Guys, it’s time to talk about copyright and decide whether or not we want to change things in the academic publishing process. There’s been grumbling about this for years, but in CS the situation has started to garner some real motion towards change.
The situation: academic publishers place a copyright on work they publish that limits the distribution and republication of the author’s work.
The problem: this limits access to our research. Many researchers have serious ethical issues with this. I get uppity about this because I am funded in part by the National Science Foundation (thanks for the fellowship, guys!) and I don’t think that the American public pays for me to do my research so that my work can sit behind some publisher paywall. They fund that work so that it can be made accessible to the public and that future startups and companies can use the ideas to start new businesses, create new jobs, etc (well, that’s assuming I do anything useful).
Why we usually don’t worry about it: In CS, we have some pretty flexible exceptions to this: authors are typically allowed to post a copy of the camera-ready version of their work to their homepage, and classroom republication is usually okay. Further, most of us doing CS research are either employed by highly profitable companies, or very well-funded university departments who can afford to pay for licenses to all the major publisher’s work. Further, we often just blatantly ignore the copyright policies and so far nobody has sued us.
Why not worrying about it isn’t a good idea in the long term: I’ll go through the exceptions I listed one by one (I know the list isn’t exhaustive but hopefully it’s representative):
- Homepages don’t last forever. Look at all the famous Computer Scientists who died this last week. They won’t keep homepages anymore.
- Universities are cutting costs everywhere; licensing fees are expensive. Poor schools and young startups never had legal access in the first palce.
- Just because no one has sued us yet for blatant violations doesn’t mean they won’t in the future.
Okay, so so far I’ve summarized a bunch of things a lot of other people have already said [1, 2, 3, 4], but here’s the thing I want to draw attention to with this post: change is happening and we should all get on board. Pockets of the CS community are starting to push for new policies, and you should know that your pocket is not the only pocket doing so. It’s time to walk down the hall and talk to that guy in the office who works on that thing you don’t understand and ask him what his sub-community is proposing, or email that friend at another university and see what they’re talking about there. While each sub-community should set its own policies, we need to gather a sense of what we should be pushing for the ACM/IEEE/Etc. to do for our community as a whole.
Here’s what I’ve observed from my Systems, Berkeley-centric corner of the world:
- The security community at the IEEE Security and Privacy Symposium voted to reject the IEEE copyright policy in favor of the USENIX policy this year.
- The ACM SIGOPS meeting at SOSP last week also voted on the issue, a report from Ari Rabkin: There was a near-unanimous vote saying “we’d like ACM to let authors keep copyright.” There was also a vote on whether this mattered enough that we’d be willing to take SOSP outside the ACM and perhaps have USENIX sponsor it. That failed, about 2/3 to 1/3. But we are on record as saying we want a change.
- Princeton University as a whole has banned researchers from handing over copyright unless they receive a waiver from the university.
- UC Berkeley’s EECS department met Friday for our annual “Town Hall”, in which graduate students and faculty get together to discuss the issues we think are most important in the department. Copyright policy was one of only three issues discussed at length, a student group has formed to advocate for policy change, and a followup meeting specifically on copyright is being scheduled to get the faculty and students on the same page and move forward. (Hey student-friends, take notes, we started asking questions and our faculty jumped on board with us!).
- Some senior researchers are committing to only review for conferences and journals that provide open access to the published work.
What discussion is taking place at your university, institution or within your CS sub-community? Have they taken any tangible steps towards a more flexible and open copyright policy?
Update: Peter Bailis adds w.r.t. university-wide open access policies: Harvard has an open-access policy (http://cyber.law.harvard.edu/node/3462), but, in my limited experience, few people submit the addendum (or the waiver you’re supposed to request if you don’t submit the addendum). Worse, the ACM has flatly rejected the Harvard addendum (http://mybiasedcoin.blogspot.com/2009/04/acm-does-not-support-open-access.html). In my experience, the ACM leverages the waiver option against authors (quote: “I am aware that Harvard’s Scholarly Publication Office has previously waived the use of this Addendum”), and authors either waive their rights or waive their publication’s right to inclusion in the ACM Digital Library. Springer-Verlag accepted one of our addendums without protest (or questions).