OPINIONS SUBMITTED ELECTRONICALLY I think SICOMP is a better place for the special issue. (I no longer am an editor for SICOMP so I think I am unbiased here.) Allan Borodin ====================================================================== I think it is appropriate that the special issue is in SICOMP. I just heard that the subscription fee for JCSS is $1670 for institutions! Mike Goodrich ====================================================================== My opinion is that there is no point in having STOC/FOCS special issues at any journal devoted to theoretical computer science at large. This is in contrast to special issues on some area (to which one wants to focus attention) or to commemorate some special event. For further ellboration, see URL http://www.wisdom.weizmann.ac.il/~oded/on-SpecIss.html Still, if the community insists on continuing the pointless tradition of STOC/FOCS special issues, then I'd prefer that these special issues appear with SICOMP. Oded Goldreich ====================================================================== I have read the materials on the web site and am swayed, both by the arguments that we should not continue to support commercial publishers who are pricing themselves out of the market, and also that the long-standing relationship with JCSS should not be terminated in haste but only after due deliberation. I only want to add two points to the arguments already made: 1. The question of whether to have a special issue at all needs to be separated from the big issue concerning dissemination of scientific work and the role of publishers in that process. It has clearly been beneficial to JCSS to have had a pipeline to top STOC papers each year. One benefit to STOC is that it has facilitated the publication of STOC papers on new topics for which there were not enough well-established journals willing to consider them. Another benefit is that it has hightlighted STOC as a premier theory conference. Indeed, the collaboration with JCSS may well have been a contributing factor to the high esteem in which STOC is now held. 2. I am having increasing concerns about the future availability of work that is published only in digital form, whether by commercial publishers or other organizations. Companies and organizations come and go. We are used to relying on our libraries for the long-term preservation of information, but the move away from ownership of printed materials to subscription-based access does not bode well for the future. What happens to the archive when a publisher goes out of business or decides it is no longer in its business interests to continue providing online access? I'd like to think that some other organization could/would move in to ensure the continued availability of scholarly materials, but I have no reason to believe that that would necessarily take place. Problems are both technical and legal. Even if the defunct publisher is willing to let someone else take over its materials, it's not so easy to absorb materials from someone else's proprietary database, and it's not clear who would cover the costs. But more likely, the database would be considered to be the intellectual property of the defunct company, and others would not be permitted to take it over. I've seen this happen to copyrighted software from failed software companies, where even after the company is gone and the software has no commercial value left, it still is not released to the public because there is no one who is able or willing to assign the rights. This is a big problem that has to be approached on many fronts. But a starting point would be to insist on copyright agreements that require continual public availability of the work at nominal cost and that automatically terminate when availability is withdrawn. This would at least open the doors to someone else moving in and taking over an archive when the original "owner" was no longer interested in or able to provide access. If the scientific community can come up with a satisfactory "uniform copyright agreement for scientific publications", it could well have the clout to insist that all major publishers adopt it. Michael Fischer ======================================================================