Re: [xml-dev] XML Conference archives

[John Cowan <cowan@mercury.ccil.org>]

Henry S. Thompson scripsit:

> Ah yes -- I use it regularly, but somehow didn't think of it in the
> context of producing the references for an academic paper.  I wonder
> what the acceptability of Archive references will be in times to
> come. . .  

Normally we cite a paper to its publisher, not to the (physical) library
that holds a copy of it.  The IA should be no different in this respect.

> <uninformed_possibly_misinformed_speculation>The Internet Archive has
> quasi-NGO-status by now, in that I believe there is some more-or-less
> official agreement in place giving them some form of special status
> wrt copyright, possibility in return for being willing to curtail
> access in some cases on
> request. . .</uninformed_possibly_misinformed_speculation>.

No, they have no special status, and have been sued for infringement.
They normally settle such cases by removing the documents.  They do
the same on request, and they apply a site robots.txt file not only at
crawl time but also retroactively: if a new robots.txt specifies that
a document already crawled should not be, they hide it.  Their policy
is that they don't want to archive anything that the copyright owner
doesn't want archived.

In the one case that's been adjudicated, IA content and timestamps were
held not to be admissible evidence in a civil suit.  On the other hand,
the USPTO allows them to provide evidence of prior art.

-- 
Almost all theorems are true,                   John Cowan <cowan@ccil.org>
but almost all proofs have bugs.                http://www.ccil.org/~cowan
        --Paul Pedersen