John Shott shott at
Fri May 14 11:34:55 PDT 2004


I'd be happy to tell you a bit more about Coral.  I'm a bit surprised,
however, by the claim that Cornell is running it ... while a couple of their
folks have a version of our remote Coral client running on their computers
and while I fully expect them to actually begin to run Coral in their
laboratory over the summer, to my knowledge they don't actually have a
version of Coral running in their laboratory.  So, I want to make sure that
we are talking about the same thing.

Basically, Coral provides the ability to:
	1. Maintain lists of qualified users on each piece of equipment
	2. Make reservations on each piece of equipment
	3. Track usage of equipment by enabling/disabling that equipment
during usage (with 	optional hardware interlocking of the equipment.
	4. Allows staff members to recharge their time to other lab users or
projects or 	accounts.
	5. Report equipment problem and shutdown conditions.

This ia a Java application and, as a result, it is reasonably portable to
different platforms although Solaris and Linux are the platforms upon which
existing installations are running.  At the moment we support both Oracle
and Postgres databases.  It has been developed predominantly by personnel at
Stanford although we now are getting two folks at MIT who have been spending
a portion of their time over the last year in helping to develop this

At the moment, we have been running Coral here at Stanford "in production"
since Jan 1, 2000.  MIT has been running it for a little over a year.
Minnesota is the first place where non-developers have tried to install and
configure it ... and, to be honest, we found that it is not as easy to get
installed and configured as we would have liked.

As a result, we are working on repackaging it so that it is easier to
install and configure and are hoping to have that repackaged version
available during the summer.  At the moment we have a waiting list of one
other lab at Stanford, 3 other labs at MIT, and several of the NNIN sites
(including Cornell) who are interested in running Coral in their

Because NSF has supported (at least indirectly) much of the effort
associated with developing Coral, we have been planning on giving Coral away
to other US universities ... well, at least to other NNIN sites.  By the
same token, we are not interested in letting for-profit organizations have
it or run it.  To be honest, we haven't really considered a policy relative
to national labs ... and that is clearly something we would have to
consider.  Part of our quandry is that, at present, we really have no direct
support for distributing this elsewhere ... and yet are doing our best to
try to be good citizens in this regard as it is our belief that Coral does
useful things in laboratories of this type and that there is, to our
knowledge, no commercially available software that does what Coral does.

I hope that helps to answer your questions ... let me know if you would like
any further information.



More information about the coral mailing list