D. J. Bernstein
Notes on writing papers

Don't publish with IEEE!

Before you read this page, you should understand (1) authors putting papers online to benefit readers, (2) commercial publishers using copyright to limit #1, and (3) authors dedicating papers to the public domain as one way to stop #2. I have a separate page discussing these issues.

It turns out that, in response to #3, IEEE is overriding its scientific referees and flat-out refusing to accept public-domain papers.

I learned about this from a UIC graduate student who had submitted a paper to a conference whose proceedings were to be published by IEEE. After the paper was accepted, IEEE notified the student that a copyright transfer was required. The student declared his intention to put his paper into the public domain. The IEEE Intellectual Property Rights Manager threatened the student with non-publication of the paper. Faced with this pressure, the student capitulated and, rather than eliminating the copyright, transferred it to IEEE.

When I heard about this incident, I asked the IEEE Intellectual Property Rights Manager to explain his overall assertion that IEEE refuses to publish public-domain papers:

Works from government authors are in the public domain, and I find it difficult to believe that IEEE refuses papers from government authors. I see that the IEEE Copyright Form has a special section for government authors.

How many public-domain papers does IEEE actually publish? Surely you have the exact figures for each year. Is it fair to say that, in fact, IEEE publishes many public-domain papers?

The IEEE Intellectual Property Rights Manager refused to answer these questions.

In his messages to the student, the IEEE Intellectual Property Rights Manager had asserted that ``IEEE needs to be the owner of the work ... by assignment.'' Obviously this is impossible for papers in the public domain: copyright assignment isn't possible when the copyright no longer exists. I looked at the IEEE Copyright Policies and found that public-domain papers were clearly exempted from the copyright-transfer requirement:

Such transfer shall be a necessary requirement for publication, EXCEPT for material in the public domain.
(Emphasis added.) I asked the IEEE Intellectual Property Rights Manager to explain the contradiction between his statements and the IEEE Copyright Policies. His only response was the rather idiotic comment that ``IEEE policy requires authors to submit an IEEE Copyright Form in order for publication to occur''; needless to say, the IEEE Copyright Form is written by the IEEE Intellectual Property Rights office.

In his messages to the student, the IEEE Intellectual Property Rights Manager had explained IEEE's alleged need for being ``the owner of the work'' as follows: ``We can put it into the Xplore database and license it to others as one of our ongoing electronic distribution of IEEE publications.'' I asked the IEEE Intellectual Property Rights Manager for clarification:

When there is no copyright, IEEE is completely free to do these things. It can distribute the work as widely as it wants, and it can authorize others to do so. All IEEE gets out of a copyright is the power to _stop_ the distribution of the information.

If the public domain is a problem for IEEE, how does IEEE publish works from government authors? Do you have a better explanation for your desire to be the owner of the work? Is it fair to say that IEEE actually _does_ want to stop the distribution of scientific information?

Naturally, the IEEE Intellectual Property Rights Manager refused to respond.

The IEEE Intellectual Property Rights Manager had also devoted some effort to trying to fool the student into believing that papers could not simply be dedicated to the public domain. I asked for clarification:

The Ninth Circuit Court of Appeals has commented that ``It is well settled that rights gained under the Copyright Act may be abandoned.'' The standard way to abandon copyright is by a clear written dedication of the work to the public domain. One example of a public-domain dedication is http://creativecommons.org/licenses/publicdomain; surely you're aware of attorney Lawrence Lessig's Creative Commons effort.

You stated that ``there is a specific legal process for placing your work in the PD.'' When the student asked you for details, you refused to answer. Instead you said that you were ``dubious about the idea of simply declaring one's intention to inject a work into the public domain,'' and that IEEE needed to be able to ``prove'' its rights.

If IEEE has trouble ``proving'' its rights to publish a public-domain paper, then how does IEEE ``prove'' alleged authorship of a paper whose copyright is allegedly transferred, and how does IEEE ``prove'' an allegation of government employment? Is the ``legal process'' you mentioned something more than making a clear written dedication of the work to the public domain? If so, what exactly is the process? Why didn't you answer the student's question regarding this process? Is it fair to say that the IEEE Intellectual Property Rights Office is trying to intimidate authors into transferring copyright to IEEE?

The IEEE Intellectual Property Rights Manager refused to respond.

The bottom line is that IEEE is refusing to accept public-domain papers except from government authors. IEEE has no justification for this position. IEEE's action is a blatant attempt to maintain control over papers that would otherwise have been freely available to the public. Unfortunately, at least in this student's case, the attempt succeeded: a paper that was accepted by IEEE's scientific referees, and that would have been in the public domain without IEEE's pressure, is now part of IEEE's copyright portfolio.

Consequently, I am blacklisting IEEE here. I recommend that authors find another publisher. Springer, for example, tacitly (although quite unhappily) allows public-domain papers, and AMS explicitly does not require a copyright transfer.