Re: Bruker Software/should be unix/linux

Eric Enwall (elenwall@chemdept.chem.ou.edu)
Thu, 11 Jun 1998 08:50:48 -0500

The cost of hardware is, I agree, today pretty trivial (though I run some
old stuff because of cost) and should not be the determining factor. I
would suggest that there are a variety of reasons why we'll see little
change and one why we might see a change to NY.

1. Most importantly for the status quo is the status quo. Most of the folk
who do the programming for B&V are unix geeks and are proficient in the
flavor of unix that the two are using now.

2. In conjunction with 1, corporate bean counters and software managers
know the (real or imagined) costs of retraining a computing staff and
making a major change in direction. The value of using legacy code is IMHO
much over rated, but the example of Varian's transition from V77 to 68000
systems shows how it can dominate software. Bean counters are not fond of
change.

3. Most large software outfits (B&V have pretty good sized groups) are
terrified, probably from experience, at the thought of relying on
"unsupported" development tools. Linux does not have a very good level of
contractual software support (although I think there are some independent
support companies for Linux or BSD.) If Linux is broken for your specific
use, how do you get it fixed? For Solaris or IRIX you at least have someone
to shout at. This is important in the .com realm.

4. Inertia. Why do we have integral data acquisition / processing programs
today? With multi-tasking cpus, the acquisition software and analysis
software could easily be running independently. Lord knows the acquisition
software could probably run on an Apple IIe if properly coded since the
hard/fast stuff happens in the acquisition console. In that case, bug fixes
could be independent. Of course then third party analysis programs might
make more inroads.

On the other hand, unless DOJ provides us with baby bills, the Jugernaut
will almost certainly result in NT as the operating system for all US users
who live in the .com and .gov world and for those who use software from
those places.

Only those of us with serious anarchistic tendencies or who are canalized
by other systems will be outside the blue screen of death.

Cheers.

--
Eric-Enwall@ou.edu
Department of Chemistry and Biochemistry, University of Oklahoma, Norman
Campus.  Norman Oklahoma 73019-0370
(405) 325-7571   (405) 325-2843  fax (405) 325-6111