sh> after striking out those passages (which lately include attempts to sh> make incoherent distinctions between permitted archiving on a "personal sh> server" and forbidden archiving on a "public server" -- absurd because sh> all "personal servers" on the Web are public! These distinctions are sh> based on paper publishing categories that simply have no counterpart on sh> the Web; hence the mootness of the older agreements).
Although I have been following this debate with great interest over the last few years, I rarely comment. I was intrigued, however, at the distinction between "public" servers and "private" servers attempted by commercial publishers. It's easy to see what they mean - "Nature" or Medline are "public sites" ie places one would go to seek out literature that one was hitherto unaware of, while my Departmental home page is a "private" site and one would perhaps only find my paper there if one knew it was there to begin with, or if one stumbled across it fortuitously. The trouble is, as you point out, it's a very tenuous distinction. I would love to know how they would view a paper which was archived (physically) on a "private" server, but which was pointed to by a link on an abstract residing on a "public" server. The reader would of course, neither know nor care where the document in fact resided, as long as there was a hyperlinked trail from the public site. Does this mean that "privately" archived papers are illegal if they are referenced by a hyperlink from the existing "public" literature? And how would one know or enforce it? What if one publishes the hyperlink to one's "private" site in the abstract of the "public" version? Or if one's article is indirectly hyperlinked (deliberately or unwittingly) by an intermediate paper? Would one ban hyperlinked references in published papers just in case they led to an "illegally archived" paper? What about papers in the future which exist not as discrete documents but as database entries in a number of parts (possible distributed across servers) which are assembled "on-the-fly" in different forms for different users? The situation rapidly becomes ludicrous. Of course there could well be technological solutions to the situation, given a unified effort on the part of commercial publishers. Take, for example, the copy-protection/piracy battle over the last 2 decades, or the current MP3 controversy. It would have been naive to say 20 years ago "but files can easily be copied, so protection of software is impossible". And on the Web, one can never say "never"! _________________________________ Dr Ian L. Ross PhD Centre for Molecular and Cellular Biology The University of Queensland AUSTRALIA 4072 Ph 61-7-3878-8618 or 61-7-3665-4447 fax 61-7-3878-8663 or 61-7-3665-4388