Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!samsung!brutus.cs.uiuc.edu!lll-winken!csustan!news
From: rsc@altair.uucp (Steve Cunningham)
Newsgroups: comp.edu
Subject: Re: Why are CD-ROMs so expensive?
Summary: The cost isn't that outrageous
Keywords: Programming costs money
Message-ID: <1989Dec28.191540.24116@csustan.CSUStan.Edu>
Date: 28 Dec 89 19:15:40 GMT
References: <829@tijc02.UUCP>
Sender: news@csustan.CSUStan.Edu
Reply-To: rsc@altair.csustan.edu (Steve Cunningham)
Organization: CSU Stanislaus
Lines: 36

> In article <829@tijc02.UUCP> pjs269@tijc02.UUCP (Paul Schmidt) writes:
> In doing some research on my Master's Thesis I turned to
> Mathematic Review...[remarks on missing MathSci]..Because it cost $3000!
> 
> This is an outrageous price for the disk.  A CD-ROM costs
> about $10,000 to master and $2/disk to reproduce.  If you
> would assume that it cost another $10,000 in manpower to
> create these disks (about one man month) then they are
> making a killing on these disks!

As I understand the nature of MathSci and Mathematics Reviews, there are
three cost factors involved.  The first is the cost of running MR's (as
we called them when I was in the MathBiz and reviewed for them) -- this is
quite high, and MR's probably ran a deficit in the American Mathematical
Society publications budget (personal guess).  The second is the cost of
actually preparing the CD-ROM, and I'll bet there's some fairly good
indexing and direct access programming on it -- I have *serious* doubts
about the "one man month" assumption!  This is programming -- and that's
never cheap.  The third is a personal guess that there is a third party
involved in the works somewhere, that this is a commercial concern, and
that they don't work free.

At any rate, the implication of the original posting is that the AMS is
ripping off the mathematical public (THERE's an oxymoron!) by charging
too much for this disk.  I strongly doubt this, since the AMS is a
professional organization with open books, and such organizations are
very sensitive to the benefit/cost ratio.

Oh, do I think that $3000 is a lot of money?  You bet!  We certainly do
not have this disk (or, to put it more correctly, this *service*, since
it probably involved updates as well, and it is the service that's being
paid for).  But I don't think it's as simple as the posting suggests.

Disclaimer:  I am not now a member of the American Mathematical Society,
             am not associated in any way with this project, and may not
             know all the details I suggest above.