>To say that I disagree with that summary wouldn't be exactly accurate
-- I
> think it would be more accurate to say that I find it irrelevant.
It's
> like saying that there's no evidence to date that giving away free
books
> would lead libraries to stop paying for them.  Of course there's no
> evidence for that proposition -- how could there be?

Rick, This seems a bit strong given the 15 year history of arXiv.

> Similarly, if and when easily-searched IRs provide immediate,
permanent OA access to the full
> content of a given journal, it's crazy to suppose that subscriptions
to
> that journal wouldn't drop.

I agree.  But since Stevan's critique states "it is possible, indeed
probable, that self-archiving will cause some cancellations,
eventually," I am not sure how you are disagreeing with him.

> Stevan seems to doubt that librarians will know when a journal
becomes
> completely self-archived, but I suggest that he underestimates the
> increasing budget pressures that are shaping our subscription
behavior.

Given the volume of his output, I'm sure I'm missing something, but in
rereading the critique, I don't see this.

>We'll start asking publishers to send us lists of those titles for
which they allow
> self-archiving.

Aren't these permissions generally set at the publisher, rather than
title level?  Isn't this information publicly available?

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

>And where we can cancel paid subscriptions, we will.  Would everyone
cancel every
> title in this scenario?  No.  But many of us would cancel many
> subscriptions -- we'd be irresponsible not to.

I absolutely agree, but I'd be interested in hearing what others
think.

Bill

--

William Walsh
Head, Acquisitions Department
Georgia State University Library
100 Decatur Street, SE
Atlanta, GA 30303
Phone: 404.651.2149
Fax: 404.651.2148
Email:  wwalsh@gsu.edu