krj@na.toronto.edu (Ken Jackson) (04/29/89)
From: nagmag%ukc.ac.uk@NSFnet-Relay.AC.UK
Date: Fri, 28 Apr 89 12:25:03 EDT
Subject: NAgMAg V89 #10
NAgMAg Friday, April 28 1989 Volume 89 Issue 10
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% The official electronic digest of the NAG Users Association %%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Today's Topics
Re: [NAgMAg V89 #9]
sparse mathematical optimization
TSA installation
NAGMAG #10 Contribution from NAG
%%
%% A shortish issue to take us into the Bank Holiday -- I'll
%% expect to receive an avalanche of articles on Tuesday when
%% everybody has returned from their day in the garden.
%%
%% SOS BOARD
%%
%% 1) I know this is a complete long shot but I've been mailed
%% by
%% bsatzer@gonzo.eta.com
%% to be included on the mailing list. All mail to this addressee
%% gets bounced with an 'Unknown user' message.
%%
%% Anyone out there know an alternative address.
%%
%% 2) Also somebody (from Germany I think) offered to redistribute
%% nagmag to a statistics mailing list -- I had a slight accident
%% with the mailer and lost the original message -- if you're
%% reading this please can you contact me again!
%%
%% Have a sunny week-end!!
%% Tim
--------------------------
Date: Sat, 22 Apr 89 16:33:31 BST
From: MRO2@phoenix.cambridge.ac.uk
Subject: Re: [NAgMAg V89 #9]
I would like to reply to Tim's comments in the last issue on the
new NAG Help.
1) The new NAG Help does have all its I/O in separate routines. The
prototype Unix implementation is written entirely in Fortran, except
for 3 routines to handle CONTROL/C which are written in C. It works
on at least one Unix so, if what Unix hackers tell me is true, it
should work on any old Unix, shouldn't it?
[%% Oh! The naivety of the man -- he BELIEVES unix hackers!!
Of course it will port:-) Tim]
2) New HELP is a hypertext system of sorts, depending on exactly what
you mean by that term. Quite a lot of thought has gone into future
development of a windowing interface to it, and we will turn our minds
seriously to that once the new system is working satisfactorily in
'line mode' on a reasonable number of systems. The problem with more
than one HELP system is that database maintenance is very labour
intensive - in fact it was the prime motivation for writing a new
program! Ultimately, when the specification is stable, it will be
possible to publish the format of the database so that users can
develop their own or convert what is supplied.
3) Following David Sayers' comment, I now think it would be unwise
to supply the example programs as part of the HELP system because
of implementation-dependent problems, and also because users often
want to modify and run them. However, the HELP system should
certainly be able to indicate their location on the host system,
and something could be done to make this more visible to the user.
---------------------------
Date: Tue, 25 Apr 89 10:40:41 MEZ
From: ZZENDE@dhvrrzn1.earn
Subject: sparse mathematical optimization
Some of our users have to solve sparse mathematical optimization
problems. Several of them can do that with a package, and we offer
MINOS to them. Others would prefer to call some subroutines
from their FORTRAN program. I am aware that in principle MINOS can
be stripped down to do that job, if you get permission from
Stanford.
I wonder whether other NAg users have similar problems and whether
NAg is going to provide something in that field.
Peter Endebrock, RRZN Hannover
---------------------------
Date: Thu, 27 Apr 89 14:48:33 MEZ
From: ZZENDE@dhvrrzn1.earn
Subject: TSA installation
I have been asked to inquire through NAgMAg whether anybody has
successfully installed TSA, the time series analysis package
distributed through NAg.
The installation that has asked me has tried to get a complete,
compilable and working version (in coded form), to install it on
their CYBER system running under NOS/VE (for NAg information: that
is DHI in Hamburg, Germany). Until now, nothing could be achieved.
Can anybody give information? Peter Endebrock
---------------------------
Date: Thu, 27 APR 89 18:20:16 GMT
From: DAVID@vax.nag.co.uk
Subject: NAGMAG #10 Contribution from NAG
Communication with NAG Staff
NAG is on a number of networks and many staff welcome and prefer
this means of communication. Some systems would prohibit the use of
the network for commercial exploitation and so, for example, we
would not wish to abuse the very legitimate use of the JANET network
to answer technical queries and pass legitimate information (like
this) by encouraging a use that could be considered as commercial.
In short if the rules of the network allow it we are willing to
receive messages; otherwise we would not wish to compromise our
position by laying ourselves open to the charge of commercial
exploitation.
[%% Would you like to be a little more explicit and provide some
e-mail addresses for people in Nag willing to field different
sorts of enquiries? -- Tim]
Workstation Versions of the NAGHELP System
Of the people I have spoken to within NAG about hypertext, the view
was that we are not closed to the possibility of looking at the
results of Tim's experiments. We want users to use our software and
to do that it is important to give them the right information in a
form that they find most convenient.
Cray XMP Availability
Mike Geary has returned from the USA with the news that the
implementation is essentially complete. A master tape has to be
written and the supporting documentation printed, but we do not
expect this to take much longer.
David Sayers 27 April
---------------------------
%% For further information about the NAG Users Association please contact:
%% Janet Bentley, Administrator NAGUA,
%% Shore Lane Farm, Blackstone Edge Old Road,
%% LITTLEBOROUGH, Lancashire, OL15 0LQ, UK.
%%
%% Replies or submissions to nagmag@uk.ac.ukc
%% Distribution changes to nagmag-request@uk.ac.ukc
%%
%% END OF ISSUE
-----------------
Reposted by
--
Kenneth R. Jackson, krj@na.toronto.edu (on Internet, CSNet,
Computer Science Dept., ARPAnet, BITNET)
University of Toronto, krj@na.utoronto.ca (CDNnet and other
Toronto, Canada M5S 1A4 X.400 nets (Europe))
(Phone: 416-978-7075) ...!{uunet,pyramid,watmath,ubc-cs}!utai!krj