I did not know that Elsevier had changed it policies in that regard. I did know that if you asked they would send you a different copyright transfer form asking only for a permission to print, but you had to know to ask for it and my last publication in PAA was a dozen years ago. And a colleague of mine got a "lawyer's letter" from some journal demanding that he remove a paper from his own web site. Although I am glad (I suppose) to hear that they have yielded on this point, my basic objection remains. Michael On Sat, 9 Jun 2007, John Baez wrote:

Michael Barr wrote:

...

After thinking about it, I cannot restrain myself from responding to Ross's message that the procedings of CT07 will be published by Applied Categorical Structures. [...]

Indeed! It's a pity that the proceedings of the main international conference on category theory is going to be buried in this journal.

Why not publish it in TAC?

...

ACS is published by Kluwer (now a subsidary of Springer). Kluwer is one of the "gang of five" publishers that are sucking all the life (not to mention money) out of mathematical publication. The journal is not subscribed to by McGill nor by any other university in Montreal. I would actually be surprised if any university in Canada or more than a small handful in the US subscribe. It is no wonder since they charge, as far as I can tell, in the neighbourhood of $3 a page so that the annual subscription of nearly 100 pages costs nearly $3000. The author of a paper published there is legally enjoined from posting it on his own web site.

Is that still true? If so, that's terrible. Even most Reed-Elsevier journals allow you to keep your papers on your own website - and more importantly, on the mathematics arXiv.

However, Reed-Elsevier only officially accepted these practices recently. Before that, it worked like this: if you demanded the right to keep your paper on the arXiv, they'd give in and let you do it. I think they were trying to avoid public battles, to keep from looking bad.

So, if anybody feels compelled to publish in a Springer/Kluwer/Reed-Elsevier journal for some reason, they should simply refuse to give away the complete electronic rights to their papers. If necessary, amend the copyright form to say you have the right to keep your article on your website and the arXiv. Journals are unlikely to turn away papers for this reason after they've already been accepted for publication.

You can read the copyright transfer forms for some math journals here:

http://front.math.ucdavis.edu/journals#copyright

and many more here:

http://www.sherpa.ac.uk/romeo.php

Unfortunately, the information about Springer seems a bit contradictory.

Best, jb