Well along the vein of "you asked for it, you got it," I'm posting the full
text of the compilation that I prepared. Please excuse my use of excessive
bandwidth. There really was enough demand to post it in its entirety :^)
From: Steve Thompson VADMS Computational Molecular Genetics
Date: April 21, 1993
To: All concerned parties
Re: International support for VADMS' philosophy
This document attempts to summarize some of the current "world-view" on the
utility and worth of maintaining an interdepartmental computational molecular
biology facility such as VADMS in academia. The enclosed information is
particulary timely as VADMS, as well as much of academia, is currently
undergoing a budgetary re-evaluation with the powers-to-be in Central
Administration. It appears as if functions designed exclusively for teaching
and research missions have lost favor; only those directly designed to bring in
the almighty dollar are to remain in full working condition. All others risk
the danger of being sliced and diced beyond recognition. Especially in our own
case, anything less than the whole is almost nothing. I, a molecular
geneticist who has learned how to USE and TEACH the available systems for
sequence analysis, can do little without Susan's expertise in maintaining and
updating the system and writing and modifying the code when that is required.
Likewise, often my searches and alignments provide the starting point for many
of Susan's collaborative modeling projects. The VADMS type of biocomputing
facility is often much more than the sum of its parts. It's growth needs to
be encouraged in order to keep American universities competitive, not stifled.
The following is an edited and abridged compilation from several of the
responses received in reply to my annotated posting to the BIOSCI/BIONET boards
Molecular-Evolution and Biological-Software and LISTSERV board INFO-GCG of the
"Appeal for Help" memo prepared by my supervisor, Susan Johns (see Appendix I).
All opinions expressed herein are solely those of the identified authors and do
not constitute the policy of the employing agency or University. SMT 4-21-93.
David Morgan at Brandeis University was the first to reply:
Sounds bad .... I can believe the situation (when the budget axes get
sharpened, almost anything can happen), but it doesn't make any sense.
Bruce Roe, Professor of Chemistry and Biochemistry
at the University of Oklahoma explains:
Just a short note to lend my support to your efforts to maintain
a campus wide computer facility to aid molecular biology research.
I've read [the] memo and support your efforts to maintain a
state-of-the-art facility that includes programs, databases, training
and support of both teaching and research ....
In these days of wide spread use of molecular biological techniques
in many disciplines, it is critical that readily accessable computer
facilities be available at universities for both teaching and research.
I applaud your efforts to maintain such a facility and hope that your
administration will as well ....
One of the major purposes of the "Oklahoma University Genetic Computer
Group" is to unify the biotechnology and genetic engineering groups
through a central computer facility which eliminates expensive
duplication of resources .... Widening use of this centralized
computer facility will improve the communications among the various
State-wide research groups, will facilitate the training of personnel,
and provide the other services prerequisite to fostering the growth of
the strong, vital research environment necessary for further scientific
and economic development ....
One of the major goals of the OUGCG is to provide the support personnel
necessary to train researchers to both access and use OUGCG, as well
as to serve as a centralized resource to answer questions and to help
the researcher implement the available programs ....
He replies to our plans in a subsequent message:
Go for it. Give'm hell and all that stuff. I have no problem
with your broadcasting any of the comments I sent [except to maybe
cleaning up any obvious typos/spelling errors/and vulgar words :-) ]
Cheers and hope all turns out ok for you......bruce
Rick Westerman, system manager of the AIDS Center Laboratory for Computational
Biochemistry (ACLCB) at Purdue University, recounts a tail of woe:
My sympathies on your plight. It sounds similar to our situation here:
bio-computing support not being funded from the central administation.
For a long time, we've been funded via an AIDS grant from NIH. However,
this grant has been cut and for the last year we've had to start
charging individuals for use of the computers. While, up to the
present, we've been able to cover costs, the use of the computers has
plummeted as people have started to explore other bio-computing
resources aside from our VAX workstations. I fear as the AIDS grant
dries up completely and as people use the computer less, we will
consequently have to charge so [much] money per CPU hour that even
more people [will] drop our services until we are no longer a viable
operation.
And in "Brother, Can You Spare a Million Dollars?" and subsequent postings,
Mark Reboul, Sr. User Services Consultant, at Columbia University's
Cancer Center Computing Facility, bitterly expounds at length:
< ... saying things that we've all felt, but been reluctant to act upon. SMT >
... let me respond with sympathy in general, disgust at the lousy way
the bigshots are running things these days at almost every American
university, and hardly any useful suggestion whatsoever!
In the 80's the young MBA's brought their soulless, short-term-
oriented, bottom-line mentality to the corporate world and made
working for a living a whole lot more miserable than it already was
for that seething mass of unwashed, namely, us workers. Now it seems
the universities don't want to miss out on any of the action, so
they've taken to hiring these uninspired non-scientists to manage
their academic affairs for them ....
These new career administrators -- or should I say professional
bureaucrats -- were not drawn from the academic ranks, and they
appear to have no commitment to the institution which employs them
or to its traditional missions: education and research. They don't
care about these missions; they don't even understand them. They
seem to view the professors and researchers as THEIR subservients,
whose "job" is to get grants -- out of which they can suck their
overhead cut to finance their luxurious quarters and six-digit
salaries.
And they are now fully in control of the system, which they prove
weekly by building yet more and more sub-bureaucracies, hiring more
high-paid administrators, and at the same time crying poverty to the
people trying to get REAL work done. I am convinced that the claims
of poverty exist in inverse proportion to the true situation.
Essentially, they are changing the culture of the academy. It's no
longer a place for discussion, speculation, and study -- certainly
not as a first priority. The new main priority is fund-raising.
Anybody not of the entrepeneurial mindset is looked down upon, as
somehow not a fully developed professional.
I think I'm qualified to observe and comment on this change because,
in fact, I am NOT an academic. I'm neither student nor professor,
just a technical worker. I worked at two jobs in industry as an
engineer before I bailed out of that and came to Columbia, where I
am now in my second job. Since 1985, I've seen the atmosphere
change. Now it's all meetings and politics, talk about money, etc.
To my eyes, the whole thing is a sham. I would say "a house of
cards", one destined to come tumbling down, sooner or later, when
someone on the outside finds out how much of what the taxpayers pay
for here is nothing more than hot air -- only I don't think that's
the way it works in the modern welfare state ....
We have recently seen a situation here where one very good man,
contributing highly to our local enterprise's research, was laid off
"due to lack of funds", while at the same time nearby administrators
enjoyed an office renovation costing well over $100,000. And then
they spent money again on a party to celebrate their new offices!
In [the] appeal memo you say that [VADMS is] currently supported by an
"informal association between ..." There's your problem right there.
In the modern epoch of Central Planning, nothing informal is
allowed. Bureaucracy does not tolerate informal agreements.
< ... great example of bureauratic blundering deleted to save space. SMT >
All I can say is, the jerks in power now must be some really unhappy
people, judging by their inhumane, thoughtless, un-forward-looking
decisions. Not that I wish anybody to suffer, but I keep reminding
myself the bigshots are way worse off than I am. For one thing, they
take themselves seriously! How hard must it be for them to maintain
that charade inside their own heads?
They continue to sacrifice our once-venerable institutions just so
they can maintain their high-paying jobs and pseudo-careers, to keep
their minds from focusing on their own lack of self-esteem, and this
country's educational system continues to decline, with respect to
those of other countries. Within the U.S.A., all schools seem to be
hit with more or less the same cancer at the same time, so their
relative rankings remain intact, and hardly anybody notices the
difference.
About charging for resources and services ...
Our Facility has charged the users' research grants for years, and
we charge BIG ... You know what? We've got more users than ever before,
paying more than ever before ... Although most users continually
complain about having to pay such high rates, when charges doubled
there was no mass exodus off our system, not even a complaint ....
The logic which seems to prevail has to do with a sense of the
elite, or a worship of luxury. That our resource costs so much to
use means it MUST be valuable -- or something like that. That the
management would even think to suddenly raise our rates by 100% MUST
mean were good. (Sometimes, one has hot air thrust upon him.)
If it comes to a pass where you must seriously consider starting to
charge your users, I cannot recommend our complicated rate
structure, which is broken down into all these different types of
use. It's just too much trouble for the system manager (my boss). At
the NYU medical campus, as of last year, they charged each user one
flat rate, I think it was $450 a year. (Under our system, one of our
user labs alone has paid up to $14,000 in one year.) For that
reasonable fee, the NYU user has complete freedom to compete for
unlimited CPU time, and a limited disk quota, something like 3000
blocks.
The flat rate approach sounds good to me: easy to implement. It does
get a bit fuzzy when you talk about multiple users from a single lab.
Is a 20-user lab to pay 20 times as much as a 1-user lab? If so,
then what prevents the 20-user lab from signing up for only 1
account?
Is a 20-user lab to pay the SAME fee as a 1-user lab? If so, then
isn't that unfair to a young professor starting out, who hasn't had
the chance yet to become a bigshot with a big lab? I'm not sure how
those questions are resolved.
< ... suggested fee structure breakdowns deleted to save space. SMT >
... I started out with cynicism, moved to pessimism, then into
realism. I'd like to end with optimism ....
If worst comes to worst and you have to charge, try to look on the
bright side. Because of the perverse incentives in today's society,
you may end up garnering MORE respect for your service in the long
run. And if you work it right, you may end up with MORE control over
your internal finances. With luck, in a couple years you might find
yourself BETTER off than before, able to hire more people, able to
buy more advanced equipment. I am not talking about empire-building.
I'm talking about somehow getting into a situation where the people
in control are the same people who care most about the operation,
namely you and your people ....
Mark adds in a subsequent post:
Pardon my previous rant, which focused more on a sidelight of your
current plight than on the chief issue: Who needs a facility like
yours (or ours)?
Most people, certainly including the people above you with the tight
belts, know nothing about computers. They don't know how they work,
they don't know what they do, they have no awareness that some
programmers are responsible for how their Lotus thing looks and acts,
they are unaware of 99% of the functionality available from the machine.
Basically, if they use computers at all, it is in some very simplistic
application, and they are convinced that computers are "easy" to use.
Thank you to the computer industry for having hyped that message to the
max.
As a result of this false perception, the people in charge of your money
probably think that your user support function is, strictly speaking,
not necessary. At certain times, they may grudgingly concede some
limited value to your services, but when the money's tight ....
On the other hand, I'll bet your users are VERY grateful for your
services. It's embarassing to beg for praise, but if you can get enough
of your users to write letters to the bigshots, it might influence those
misguided attitudes.
Since all the administrators care about is money, it would probably be
useful if your users' letters could make the connection between
sophisticated data analysis done in conjunction with your facility and
its experts AND the winning of grant funds.
For example, there have no doubt been grants awarded to molecular
biologists in which sequence analysis results played some part of the
work described in the proposal. Presumably, the proposal would have
been less credible without that supporting sequence analysis.
Presumably, that sequence analysis would not have been possible, or
at least not done in the best way, without your help.
If it doesn't seem too distasteful, perhaps you could suggest that
line of argument to any of your sympathetic users who might go to bat
for you.
Probably this is obvious, but I figured it couldn't hurt to say it.
Aside from your particular problem, I don't know how we're going to
get out of this mess nationwide. "Cancer" is an appropriate word for
it, because it seems like it will continue to spread until there is
nothing left to feast on. It's just too bad. I thought I had escaped
[this] nonsense by bailing out of defense contracting, but no such luck.
And now the universities are big 'gummint' contractors themselves.
They're already taking money from the DOE (formerly the AEC) for
biomedical research. Pretty soon they won't be able to resist applying
to DOD for grants -- and DOD is already offering grants in these
gene-related areas. (What's the plan, to breed radiation-hardened
humans?)
The universities have become arms of the federal government, carrying
out research using LOW-PAID labor like captive graduate students,
almost-as-captive post-docs, and miscellaneous technicians ....
And in "A couple more thoughts overnight" Mark further states:
In your appeal memo you provide e-mail addresses for two of the
bigshots. I think it may be unproductive for your supporters to
contact bigshots by e-mail.
First, unless your bigshots are significantly different from our
bigshots, what makes you think they read e-mail? (What makes you think
they know what the word "login" means?) Sure, they all get an e-mail
address to be current with technology, but do they use it? Even if they
do use it, do they use it for anything more than arranging the next
3-hour lunch? And, if they only login to read their e-mail like once
every two weeks, the impact of a given support message may be dulled.
One thing's for sure, they don't use e-mail the way we use e-mail,
which is to say, constantly, internationally, verbosely, and for
exchange of complex information relevant to our jobs. They may not be
comfortable with the idea of writing or reading long messages on subtle
topics.
But even aside from that .... Career bureaucrats do not take seriously
anything which is expressed without formality. We here have been
instructed by our own administrators (who are supposedly on our side?)
that ALL communications are to be made on the official department
letterhead, that is, in print, on the sacred papers.
< ... frustrating but all too common anecdote about the stupidity of the
official paper-trail deleted to conserve space. SMT >
So, what I'm suggesting is, if you want your supporters' messages to be
taken seriously -- as seriously as possible -- by the bigshots, then
those messages better be written in a style that bureaucrats
comprehend. I don't know whether that means letterhead at your
institution. But my guess is that an e-mail message to an administrator
is like a file copied to the null device! ....
furthermore, Mark writes:
Prior to your recent INFO-GCG posting, I had seen your name on the
network a number of times .... That means that, amongst the hundreds
or thousands of technical workers and scientists around the world who
are engaged in this sort of work, your name is known. You're famous.
So here's my next question....
Do you think your bigshots have any awareness that every day you are
making WSU look good, by your intelligent messages which are seen by
thousands of respected scientists around the world? If they don't
already know that, could it possibly help your situation if they
could be made to know that? (I'm sure they have no idea what actions
you carry out to do your job!)
My guess is, No, it probably wouldn't make any difference. Since they
have no appreciation for the value of this ongoing worldwide technical
discussion, the fact that you're a part of it would not impress them
much. It might even convince them that your job is LESS necessary, if
it allows you so much time to "schmooze" globally. (Remember, most of
their communications -- in meetings, phonecalls, letters -- are so
airy and imprecise that they're not much different from schmoozing.
Since the administrators so pride themselves on their "political"
skills, they may not even consider schmoozing a disreputable way to
make a living. For them, I mean, not for you. For you to "waste" time
unnecessarily refining certain small details by e-mail ... -- that might
be their attitude.)
I mean, this is the real problem with having administrators not drawn
from the ranks: They don't understand what anybody's working on.
In any event, you might think about this aspect of your work, the
implicit global outreach in your attempts to resolve technical
concerns, and try to decide if it's something worth stressing.
Tim Littlejohn at BCH.UMontreal.CA wrote, suggesting inclusion of two relevant
documents (see Appendix II):
I sympathise with your problems at VADMS and do give a damn. I
would like to point out 2 major movements already afoot which could
only help your cause.
Firstly, I and Duncan Rouch have written 2 articles on the issue of
biocomputing services, one was posted to bionet.software recently.
I have attached this one below. These articles will be printed in
the UK journal "Binary"; they are just at the proof stage now,
however. Also see Duncans other two articles, posted in 1992. If
you can't get access to bionet archives, let me know.
The other (also attached) is the The Biocomputing Environment Database
(BED) organised by Ernie Retzel and Duncan Rouch. It is a database of
all biocomputing resources around the world. You should talk to Ernie
about where this database is, I am sure it would have much useful info
for you.
In response to the strong vote of support from bionetland I wrote ---
"THANKS from the VADMS Laboratory,"
Message-Id: <930420143125.20e0098c at JAGUAR.CSC.WSU.EDU>
It's text follows:
>Dear Bio-NetLanders -
>>WOW, Thank You for the GREAT Outpouring of Support!!!
>>We received several supportive messages from you all concerning our recent
>budgetry crisis. Thank you so very much. Special Thanks go out to ...
>> David Morgan at Brandeis University
> Bruce Roe at the University of Oklahoma
> Rick Westerman at Purdue University
> Mark Reboul at Columbia University
>> ... who wrote particularly strong letters for us!
>>Unless any of the individuals that wrote me object, I would like to pull
>excerpts from everyone's comments together (along with a big fat disclaimor
>"...the opinions expressed...") into a statement of:
>> "Why an interdepartmental molecular computational biology facility
> is a good thing .... the world view .... "
>>to send to the "bigshots." Seems so obvious, but some people ... you just
>gotta slap up the side of the head. 8^)
>>If this type of document would prove interesting to the general reader, let me
>know and I'll post it here. I'll try and keep the world notified as to what
>the ultimate resolve up here in Washington State is.
>> Thanks again, Steve and Susan
>> Susan Johns and Steve Thompson
>VADMS (Visualization, Analysis & Design in the Molecular Sciences) Laboratory
> Washington State University, Pullman, WA 99164-1224, USA
> AT&Tnet: (509) 335-0496 FAX: (509) 335-0540
> INTERnet: PRCADAMS at wsuvms1.csc.wsu.edu>THOMPSON at wsuvms1.csc.wsu.edu
The response from my Thank You note and offer to summarize was overwhelmingly
positive. I had recieved an onslaught of email by the next day; many people
ARE interested in what is going on here. Selections follow:
Ernie Retzel at lenti.med.umn.edu wrote:
It would be very useful to have something like that posted to the
group...
if you decide not to post, I would very much like a copy for my files.
Mark J. Duffield at Parke Davis Biotechnology wl.com wrote:
I would love to get a copy of such a document.
Jeff at ECUVM1 added:
I would like a copy of your final letter. It might provide some
useful insight for me when I try to justify the purchase of a
workstation and server for our GCG.
Jasper Rees at molbiol.ox.ac.uk promised delivery tomorrow:
Sorry I've only just caught up from vacation and am still planning to
write something for you, give me 24 hours....
In "Justification for computational molecular facility"
Mark Courtney at nsf.gov also requests a copy:
In the event that you don't receive enough feedback to justify posting
it generally, I'd like a copy of the synopsis you mentioned in your
recent email message.
And the most vocal advocate in this discussion for our type of facility's worth,
T. Mark Reboul at CUCCFA, kicks in his offer:
... Good Luck with everything!
I don't know if it's necessary to post your document net-wide, but I
wouldn't mind getting a copy of it -- for my future reference in case
our own situation here degrades ...
Jason Seaman, VAX Systems Manager, at MITWCCF Whitaker College affirms:
Yes, Please pull these thoughts together while you still have them.
I didn't get a chance to read your original post in its entirity yet.
But that's because I'm in a similar situation here at MIT (similar in
some ways and different in others), and wanted to think through what
you wrote.
So, it would be great to pull these thoughts together, and especially
the ones that make sense to "bigshots." Of course each situation is
different, but there is probably a lot in common too.
Tom Zucker-Scharff, Director Scientific Computing Facility,
Albert Einstein College of Medicine at aecom.yu.edu wrote:
I have not read the news lately, so have missed the discussion ...
I would be very interested in this type of document. I'm a new
director of a scientific computing facility, this coming grant year
I will need to do budget justifications and the like ....
Don Katz at the University of Wisconsin - Madison, psl.wisc.edu, joins in:
Please post it (and/or send me a copy). Some folks here are trying
to get a similar facility set up (which we may be managing, if it
actualizes), and your document might be useful to them.
Dan S. Prestridge, Director of the Molecular Biology Computing Center
(MBCC), Department of Biochemistry, University of Minnesota,
pogo.cbs.umn.edu shows that faculty support can help:
The College of Biological Sciences here at the U. of Minn. went through
a similar threat about 18 months ago. A faculty committee was formed
to investigate and report on the need to continue the Molecular Biology
Computing Center here. The result is that we have increased funding and
a stable future, at least for a couple years. We received very strong
faculty support.
I will drop a copy of that report in the mail for you.
Jim Myrick at gel1.2dlab.cdc.gov in "'Bigshot' report" requests:
Please post the report that you prepared for the "bigshots" or
send me a copy. Thanks
Sandor Pongor, Head, Protein Structure and Function Programme
and Computer Services International Centre for Genetic Engineering
and Biotechnology, Trieste Italy, icgeb.trieste.it joins in:
I would certainly be glad to get a copy -
Thanks in advance and good luck,
And a great summary from David J. Pierotti at sdcc12.UCSD.EDU:
Wow, sounds like things are changing up there by the minute.
Here I was thinking your job was safe and comfortably secure,
and the next thing I know you've been branded parasitic by the
gods of administration. Why is it that administrations never
seem to understand the basic concept of the scientific method?
Do they think all discoveries are marketable inventions?
================================================================================
APPENDIX I:
Copy of E-Mail message distributed to BIOSCI/BIONET boards
Biological-Software and Molecular-Evolution and LISTSERV board INFO-GCG
================================================================================
Date: Fri, 16 Apr 1993 15:57:22 -0700 (PDT)
From: THOMPSON at wsuvms1.csc.wsu.edu (Steve Thompson: VADMS genetics)
Message-Id: <930416155722.20204857 at BOBCAT.CSC.WSU.EDU>
Subject: trouble at Wash State Univ (long)
Friends and fellow supporters of bio-computing netwide:
I apologize to those among you who would rather not bother with the plight of a
small program at a relatively insignificant university in Washington state for
my disrespect of precious bandwidth, my cross-posting and my cluttering of
mailboxes. You might as well delete this message right now. For you
individuals who do give a damn --- The VADMS Center at Washington State
University is getting shafted by the political/economic policies of the
university's administration in the present round of budgetary belt-tightening!
The VADMS Center exists at WSU as an interdisciplinary service to ALL molecular
scientists campus-wide. We have demonstrated ourselves, as verified by a very
positive recent faculty review, to be a vital and much relied upon resource at
WSU. The success of many principal investigators across campus is contingent
upon the full existence of both the research/service and instructional
capacities of the Center. Indeed, it is nearly impossible to separate them,
even though in a constant and futile attempt the administration tries to do
just that --- a student is exposed to us through our instructional efforts,
continues to consult with us after completion of the course, and the work
eventually ends up in the P.I.'s research publications and grant proposals!
However, because we have not been successful in generating our OWN grant
income, we have lost all favor with the central administration. Funny, how in
these days of "excellence in education" and the "molecular initiative" those
programs most closely allied with those concepts are the first to receive the
budgetary executioner's axe.
This is not a plea for monetary assistance, rather it is a philosophical query
for sympathetic supporters. If you feel that the existence of a central and
comprehensive facility, such as VADMS, that all of a university can access,
rather than merely the use of specialized, individual laboratory resources, is
a good and special thing, please read and respond in any way you desire to the
following letter which summarizes our situation to our users. Susan Johns, my
supervisor and the letter's author, has authorized me to release the memo to
the nets. It is a copy of the version which I hand delivered to our most
ardent supporters on campus.
Sincerely, Steve Thompson
Steven M. Thompson
Consultant in Molecular Genetics and Sequence Analysis
VADMS (Visualization, Analysis & Design in the Molecular Sciences) Laboratory
Washington State University, Pullman, WA 99164-1224, USA
AT&Tnet: (509) 335-0533 or 335-3179 FAX: (509) 335-0540
BITnet: THOMPSON at WSUVMS1 or STEVET at WSUVM1
INTERnet: THOMPSON at wsuvms1.csc.wsu.edu
enclosed memo begins here---------------------
To: VADMS Users
From: Susan Johns and Steve Thompson
Date: April 12, 1993
Subj.: Appeal for VADMS Support
This is an appeal to keep VADMS alive. If you feel that your
research efforts, classroom activities or graduate training have benefited
from VADMS, please read this completely, form your own opinions and act
accordingly.
To date, the VADMS budget has recently been reduced by more than 50%
and proposed staff layoffs may close the VADMS lab completely. With the
current financial situation, all of WSU's divisions are scrambling to survive.
A unit such as VADMS that serves numerous colleges and departments but is not
directly controlled by any of them can easily be forgotten. We feel that the
continuation of VADMS is important to the university. It must not fall through
the cracks.
VADMS activities are currently supported by an informal association
between the Graduate School and Systems & Computing. The Graduate School
provides funding for Steve Thompson's position, operating funds for software
and database maintenance, operating funds for hardware maintenance on VADMS'
major pieces of equipment (i.e., the MicroVax and the E&S) and clerical
support. Systems & Computing provides funding for Susan Johns' position,
office space for VADMS' technical staff and its clerical support, space for
the VADMS central laboratory, most of the computing platforms on which VADMS
houses the software and its databases, and the associated disk storage costs,
account fees and printing charges generated by VADMS.
VADMS' success is primarily the result of the efforts of its technical
staff, Susan Johns and Steve Thompson. Each of them is in a different
administrative unit. These units are acting independently in this budget
crisis. By not interacting with one another, one or both of the two VADMS
technical staff positions may be lost. One position elimination has already
been suggested by Systems & Computing. VADMS has grown to the point where it
cannot be handled by one person. The loss of either position would destroy
VADMS' effectiveness.
Steve Thompson provides sequence analysis support. He provides
consulting services and keeps current on developments in this area. He also
collaborates with individuals working in the field. He is responsible for the
operation of the VADMS Fulmer lab and maintains equipment in VADMS workstation
sites. Steve's role in VADMS is very visible since he visits workstation sites
and the Fulmer lab regularly. He also attends relevant seminars in the field
across campus.
Susan Johns primarily provides resource support with some molecular
modeling consulting. She keeps the software and databases updated, puts new
programs into service and keeps VADMS functioning in general. She provides
software development services for those whose needs are not met by existing
VADMS software. She is responsible for the VADMS central lab operation and its
equipment. Susan's role is less visible, but vital to VADMS' existence. This is
particularly noticeable if you have molecular modeling, visualization, or
software development needs or if something doesn't work.
Susan and Steve jointly conduct the laboratory portion of BC/BP GenCB
578. Susan developed the core materials for the lab during its first three
offerings and both she and Steve have been embellishing these in their
respective areas during the last two offerings. This course requires both VADMS
staff positions, not only because of the number of students involved, but also
to keep the materials relevant to current field developments.
The Graduate School is moving Steve's FTE plus a small amount of the
operating budget over to the Division of Sciences, probably to be
administered by Biochemistry. The success of VADMS' instructional activities
and its lack of extramural funding has prompted this move. While there may be
advantages, it also generates some questions. The primary one is, can VADMS
continue to serve the entire WSU campus or will it be restricted to only the
Division of Sciences? The amount initially proposed in operating funds ($8,000)
is also too small to allow VADMS to maintain its software and hardware. It must
be increased to a minimum of $20,000.
Systems & Computing's plans are uncertain at this time. They will
probably generate three 5% cut proposals, allowing the central administration
to choose among them to produce the required level of cuts. We believe that
the reduced level of support for VADMS by the Graduate School over the last
few years coupled with its further reductions insures that Systems & Computing
will place its support for VADMS in one of these cut proposals.
If Susan Johns' position is lost, VADMS would have to radically change
in order to provide any level of effective service to the WSU campus. Since
Steve Thompson's strengths are in sequence analysis, VADMS would become
solely a sequence analysis resource. BC/BP GenCB 578 would only teach sequence
analysis techniques. Only sequence databases and easily maintained sequence
software packages would be updated. As demand for disk storage space by the
sequence databases grew, non-sequence analysis software and databases would
have to be removed from the system. VADMS users of modelling, visualization
and calculational tools would be on their own and their software and databases
would be unsupported.
Even if Susan Johns' position is retained, the shortfall in the
operating funds being supplied by the Graduate School will force VADMS to go
to a user subscription fee system starting July 1, 1993. The details of this
subscription plan are currently being developed. User input is sought in
determining a fee structure. Contact either of VADMS technical staff with your
ideas (Susan at 5-0424, prcadams at jaguar.csc.wsu.edu or Steve at 5-0533,
thompson at jaguar.csc.wsu.edu).
If you support VADMS' continuation and want it to continue serving the
entire WSU molecular sciences community, please contact the individuals listed
below through memos, e-mail or by phone and express your views. NOW! When you
call on the phone, if you can't reach these people directly, please tell
whoever you talk to why you are calling and what your views are. The message
must get through to:
Dean Robert Smith, Graduate School 335-3535, FAX 335-1949;
Tom Mueller (mueller at wsuvm1.csc.wsu.edu), Systems & Computing 335-8616; and
Provost Tom George (george at wsuvm1.csc.wsu.edu) 335-5581, FAX 335-0103
Thank you.
================================================================================
APPENDIX II:
Copies of documents described by Tim Littlejohn:
What Makes a Successful Biocomputing Service &
The Biocomputing Environment Database (BED)
================================================================================
What Makes a Successful Biocomputing Service.
The best use of computer-aided techniques by biologists naturally depends
on the quality of centralized biocomputing facilities. Previous discussions
have looked specifically at optimizing front-line user-support [1, 2].
Here we look in more general terms at what underlies good service from a
computing facility, and the implications of this for a global biocomputing
strategy.
Duncan Rouch (University of Birmingham, UK: D.A.Rouch at Bham.ac.uk)
Tim Littlejohn (Universite de Montreal, Canada: Little at ere.umontreal.ca)
Contents:
1 Introduction: What Success Means in a Biocomputing Service.
2 What Makes Success in a Biocomputing Facility?
2.1 Training of Service Staff to Deal With New Technology and Software.
2.2 Training Staff in Care of User Scientists (Customer Care).
2.3 Training Customer Scientists to use the Services Offered.
2.4 Obtaining and Acting on Feedback From Customers.
2.4.1 Working With Customers in Designing a New Aspect of the Service.
2.4.2 Socializing With Customers.
2.5 Planning for Future Needs.
3 Making and Maintaining Successful Biocomputing Services.
3.1 Promoting Success in a Biocomputing Facility.
3.1.1 Staff Morale.
3.2 Implications for a Global Computing Strategy.
3.2.1 Success for Different Types of Facilities.
3.2.2 Site-wide and National Facilities Complement Each Other.
3.2.3 Site-wide and Specialized Facilities.
4 Conclusion: Keeping the Best Arrangement of Computing Facilities.
5 Acknowledgements.
6 References.
1 Introduction: What Success Means in a Biocomputing Service.
Biocomputing services collectively support four classes of computer-aided
methods for biologists; (i) functional- and (ii) evolutionary-sequence
analysis, (iii) molecular modelling and (iv) analysis of related information
databases, such as literature.
The more successful a service that a biocomputing facility gives the
better it helps biologists to effectively use the four classes of computer-
aided methods in their work: but what makes a good service? Previous
discussions have looked specifically at optimizing front-line user-support
[1, 2]. Here we discuss factors that lay the foundation of a successful
biocomputing service, and what this means for a global biocomputing strategy.
To give a more practical definition, success in a biocomputing service
means helping biologists make the most of the available resources.
If adequate resources are not present then an elaboration of what
underlies a successful service is needed to argue for these resources.
We look at the implications of this definition for the future of academic
services here and in a followup discussion [3].
An edited version of this discussion will appear in the UK journal BINARY
(1993, 5: 51-2).
2 What Makes Success in a Biocomputing Facility?
Naturally the quality of a computing service is partly dependent on such
factors as available hardware and software, but other factors, including
communication with the clients, are at least as important. We can group
the things a biocomputing service does to be successful and, importantly,
to stay successful into 5 categories. This list is based on analysis of
successful services in both the academic and commercial arenas that offer
high-technology assistance to end users.
In summary, the activities that a successful service includes:
(1) training staff to deal with new technology and software;
(2) training staff in care of the user scientist (customer care);
(3) training customers to use the services offered;
(4) obtaining and acting on the feedback from customers;
(5) planning for future needs.
It is essential for optimum performance in the 5 categories that a computing
service has a mission statement that details its objectives. The mission
statement acts as a set of guiding principals that will shape the service's
activities. No good enterprise would be without one.
Furthermore, details of specific services offered by a computing facility
can be documented in "Service Definition Statements". These enhance
efficiency through describing each service in a way that both lets staff
know exactly what they need to do to support each service, and also
lets customer users know what to expect from a service. It helps to
communicate the information in these statements if they are rendered
in jargon-free plain language, that non-experts, such as user
scientists, can easily understand.
Now we look at the 5 categories of things a successful service does in
more detail to explain their function. To illustrate these explanations
we have choosen a notional site-wide type service that supplies computing
facilities across a complete institution. This is because only a site-wide
service can reasonably be expected to perform in all the required
categories, since it is most able to perform the various tasks that
require face-to-face contact with user scientists. Lab-, department-based
and national services fail to meet service specifications in one or more
areas. However, all types of service have a place in a cooperative global
system.
2.1 Training of Service Staff to Deal With New Technology and Software.
All successful service facilities should be continually improved as more
efficient or better ways to perform it's functions evolve. For this reason
staff should be trained to deal with the new facilities in an organized
way, not an unplanned (trial and error) way which would normally be
inefficient. People gifted with "the knack" to learn quickly can efficiently
teach themselves to deal with complex technology, but most people are not
like this. To guarantee that all staff learn what they need to know they
should go through training courses. These can be run in-house or by experts
brought in from outside. New staff can be trained by apprenticeship.
2.2 Training Staff in Care of User Scientists (Customer Care).
A facility should have at least an advice/information desk for drop-in
help, a telephone advice line and help via e-mail for front-line user-
support. These avenues for contact should encourage the use of what may be
seen as intimidating new technology, so support staff have to be
particularly sensitive to the needs of present and potential users.
Obviously then, staff should be trained to deal with customers in the most
productive manner. This mostly amounts to having a friendly attitude
combined with a good approach to problem solving. In a successful service
front-line support people need to know both how to help a client if the
problem is simple, and also where get help from if the client's problem is
complex. An in-house expert may be called in to help solve a complex
problem, so they should be the one to talk to the client. Thus many
service staff, not just front-line user-support staff, need to be able to
deal with customer users.
It is useful to rosta the in-house experts of a facility at the advice
desk, to keep them in touch with user scientists. Bringing in in-house
experts to solve user problems face-to-face is valuable in three respects:
(1) it is the most error-free method (compared to leaving the front
desk to pass on messages second hand);
(2) unrelated but useful information can be passed on at the same time;
(3) it helps clients to see the larger picture, how much is there to
help them, which helps the image of the service.
2.3 Training Customers to use the Services Offered.
The services offered by a computing facility are usually complicated to
use. To encourage biologists (especially beginners at the computing game)
to use these services the facility may need to give them a lot of help,
simply to teach them how to use the facilities. Further education of
biologists is necessary for them to come to grips with the science involved
in deciding when to use what program and how to interpret the results.
Teaching the science behind the computer applications is, however, the job
of the teaching resources of clients' home institutions, not one for
the computing facility [2].
Training methods that are used include:
(a) running training courses;
(b) demonstrations of new services, such as newly supported software;
(c) promoting the formation of computing user groups amongst the
biologists. User groups promote information transfer amongst a
user community.
Each computing user group can be informed by the service of new
facilities and information that are relevant to its subject area. This
helps to keep the signal to noise ratio high, which is an essential
feature of a good user group.
So, supporting software is an important role of the biocomputing support
facility. However, the corollary of the notion of supported software, which
includes operating systems, is that some programs are not supported, and
clients are made aware of the situation. This maximizes the support that
can be given to all clients by a facility with limited resources, since
the computer service doesn't have to attempt the impossible by trying to
support everything under the sun. The strategy beomes more efficient
if biologists subsequently respond by reducing the heterogeneity amongst
the software and hardware they use in their labs.
Supporting one package when there are similar ones that do much the same
thing can easily suit users. However, the issue is more complex when
there are packages that overlap in function but for which each has
significant unique features, and biologists have a need for these
different features. In such cases a careful assessment of the merits of
the packages may be necessary before making a decision about which
one to support.
2.4 Obtaining and Acting on Feedback From Customers.
To tell how well a service is performing it is essential to get adequate
feedback. If biologists are left confused by the e-mail training course
or the laser printing guide then the computing service needs to know.
It is good practice to ask clients to fill in feedback, or "Quality
Control", sheets at the end of a course.
The usage of services should be monitored to see if they need to be
altered. It should be kept in mind that a service may be underutilized
because potential customer biologists are not aware of the service. For
this reason advertizing of services is important, possibly taking the
form of a directory of services (layed-out in a communicative style)
sent at regular intervals to all registered users. Information presented
in hardcopy form can be significantly more effective than that in online
form, especially for beginners.
2.4.1 Working With Customers in Designing a New Aspect of the Service.
When updating a current facility or adding a new one it is useful to
obtain the thoughts of potential users at the planning stage. Their
assistance in testing the system when it's being set up, will also be
of value in identifying problems early.
2.4.2 Socializing With Customers.
What is that? Getting to know clients in a non-work setting helps to
advance rapport between members of the computing service and the client
biologists, which helps get feedback. The importance of socializing
with customers is recognized by many businesses, usually where a high-value
service is also involved.
With academic circles being more informal than business ones, socializing
with customers is not so important for an academic service. While
taking clients out to lunch is an important part of success for many
commercial services, an academic service has less to lose by not
socializing.
However, there still could be a place for socializing by staff of an
academic service in regard to the two possible types of 'client', (i) user
scientists, and (ii) the decision-makers who provide political support.
(i) The Average Scientist who Makes use of the Service.
Socializing with user scientists might help gain extra feedback
from them, such as in an informal wine-and-cheese discussion
after a demonstration. Beginner users in particular may be a
little reticent to discuss their experience and an informal
setting might help break the ice. Also, advanced users may
discover better ways of doing things that could help the service
help other user scientists: a more social environment might
foster transfer of this information.
(ii) The Senior Decision-Makers at the Institution.
That is, the people who provide political support for the
service (but may not use the service themselves), such as
department heads. Since political support translates into
funds to run the service it might be worth senior service staff
and senior decision-makers getting together. For example,
a formal presentation or discussion could be held about the
service's facilities, with emphasis on how the service can help
to enhance the teaching and research activities of the institution.
This could be followed by a social function where the
administrators could be further persuaded, gently, about the value
of the service (the "soft sell").
The head manager of the service must have a status reasonably
close to that of the decision makers in order to be taken
seriously when talking to them. A lowly rated head manager
will find it difficult to be heard by the decision makers
and so will tend to be ineffective in gaining the necessary
resources for the service.
Naturally, the kinds of social activities and forms of
persuasion that are practical will depend on the social etiquette
of the particular country and institution concerned. For example,
what is appropriate in one place may be too formal or informal in
another.
2.5 Planning for Future Needs.
Anticipating what is coming next can:
(1) reduce the need for ad hoc problem solving resources, such as
the advice desk. On-the-spot problem solving may not always be
the most efficient way to do things since many user problems
may be eliminated by organized education;
(2) help clients get to grips with new facilities faster.
For example, preparatory work may be valuable in releasing new terminal
emulation software. Potential problems users might have can be ascertained
prior to release so that the installation notes can be written to help
biologists avoid these problems. For a highly complex installation it may
be more efficient for computing staff to carry out the installation for the
biologists. That is, fast efficient work by an expert can save valuable
research time and frustration for many biologists.
To help anticipate what is coming up a computing service should have a
strategy document for the future. This should show future goals and
how these will be reached. The lack of a practical plan can lead
to poor direction, including inadequate preparation for forthcoming
facilities, and poor staff morale.
Innovation may be advanced if appropriate staff are given a percentage
of their time to investigate new 'blue sky' ideas that are not certain
to be useful, but might prove to be of interest after some preliminary
work is done. These ideas may relate to the enhancement of an existing
service or the creation of a new one. Naturally this 'blue sky' work
should also have a budget allocated to it, as with any other activity.
3 Making and Maintaining Successful Biocomputing Services.
3.1 Promoting Success in a Biocomputing Facility.
Some funding arrangements can help support user responsiveness in an
academic computing service. For example for a site-wide service, the budgets
of the research and education departments of the host institution may be
called on to contribute to funding of the computing service. In this case
the departments should have some say in what kind of service they should get.
3.1.1 Staff Morale.
A facility will not run at it's best if staff morale is low. When changes
are being made to a service to increase it's performance staff morale may
suffer: staff may feel threatened by the changes. They may think that they
have not been doing their jobs as well as they should have, and also believe
that user scientists think poorly of them too. However low morale should
be avoided, and good management is the key to this, by fostering
communication:
(1) from users to managers/staff, what user requirements
actually are (rather than what staff think they are);
(2) from managers to users, what facilities can realistically be
provided (e.g., via "Service Definition Statements", section 2).
(3) from managers to staff, realistic expectations of staff
performance. This includes (i) practical job descriptions, and
(ii) an allowance for staff to make mistakes, a normal human
activity. Intrinsic to good work practices are procedures to
minimize, identify and correct errors.
With harmony in user-management-staff relations staff may have high
morale since, in part, with improved standards they are now doing a
better job than before.
Other factors that promote, and oppose, success are examined in a
followup discussion [3].
3.2 Implications for a Global Computing Strategy.
3.2.1 Success in Different Types of Facilities.
In general site-wide entities are in a position to perform
well in the 5 activity areas required of a successful computing
facility (section 2). In contrast, the other types of facilities
tend to be less successful; these are national (due to geographical
separation from, and of, users), department- and laboratory-based
facilities (due to the lack of resources of small-scale operations).
For the non site-wide services, as long as a facility is large enough to have
proper management it is practical to set and meet objectives for a limited
performance in the 5 activity areas, appropriate to the level of resources
available. If it small enough not to have professional management then it
will not be possible to determine if users are obtaining the best deal.
3.2.2 Site-wide and National Facilities Complement Each Other.
Whilst computing facilities can be used remotely, optimum support
for biologist users requires local input. It is difficult for a
remote service to supply the necessary front-line user-support, for
example, on-site training courses and drop-in advice service. The
essential reason for this is that human communication occurs optimally when
face-to-face. Therefore, some type of site-wide facility is essential.
To put this another way, a remote facility, such as a national one, should
not be relied on as a sole supply of computing needs. However, national
facilities are an essential component of computing support. They can
provide a wider variety and more up to date libraries of software and
databases through their international connections than can site-wide
facilities. Moreover, the accumulated expertise at national sites can
help solve complex user problems that are outside the experience of local
user-support staff [2].
National or regional facilities acrue an advantage through being able to
specialize in biocomputing, due to the large number of potential
biologist users. In contrast, site-wide facilities generally support
different types of computing areas, to accomodate the different types
of users at a local site.
In the absence of a site-wide service, such as at small colleges, high
schools, and universities with inadequate computer services, local user
biologists or science educators may depend on national biocomputing
facilities. In this case users could form a local or regional user group
to swap information about experience with the national facilities to
help each other.
3.2.3 Site-Wide and Specialized Facilities.
At a local site facilities can occur at different levels, supplying the
whole site, a department, or a laboratory. A site-wide service generally
provides a particular level of support for all scientists. However,
specialized facilities are required for some research projects, such as
those involving NMR methods and X-ray crystallography. Specialized
facilities would normally be provided as entities at the laboratory or
department levels, and so separate from a site-wide computing service.
Site-wide services offer the greatest economies of scale and accumulation
of expertise at the local level. Optimally, these will offer valuable
support to local system administrators in departments and laboratories.
Furthermore, the site-wide service may be responsible for the communication
network across the site that links department and laboratory facilities
to remote facilities.
4 Conclusion: Keeping the Best Arrangement of Computing Facilities.
The global biocomputing strategy outlined here may simply
seem to describe the architecture of the current system, more or less.
However, if we cannot argue coherently for the optimal setup, in terms
of what gives the best efficiency, then we cannot expect facilities to
be maintained when threatened by budgetry restraints. Site-wide facilities
in particular may be targetted for closure, as has been virtually
the case at the University of Bath, UK. Moreover, if we do not have a clear
idea of exactly how computing facilities can optimally support science then
they may not necessarily be providing the best support possible with the
given resources.
5 Acknowledgements.
This document is a product of the UK Science and Engineering Research
Council's collaborative computational project in Biosequence and
Structure Analysis (CCP11), and is funded by a Medical Research Council
(UK) grant to Professor Nigel Brown (D.R.) and CCP11.
6 References.
[1] Duncan Rouch, Frank Wright & Alan Bleasby (Bio-Soft, October 1991).
Future of Computational Molecular Biology.
[2] Duncan Rouch, Frank Wright & Alan Bleasby (1992).
Front-line User-Support: Proposals for Greater Efficiency in
Computational Molecular Biology. Binary 4: 129-132,
& Bio-Soft, February 1992.
[3] Duncan Rouch & Tim Littlejohn, (1993)
Future of Academic Biocomputing Facilities. Binary, in press,
& Bio-Soft, April 1993.
===============================================================================
From: ernest at lenti.med.umn.edu (Ernest Retzel (1535 49118))
We wish to start a database for useful information about research
biocomputing environments in order to help people learn from each other,
so that people don't have to re-invent the wheel every time they have
to get somewhere. We've named it:
The Biocomputing Environment Database (BED)
WHAT IS IT?
The database will collect info about biocomputing at all levels, specifically:
(1) hardware and operating system;
(2) application software;
(3) other site information related to providing (managing) a
successful biocomputing environment.
Initially, we are basically after info on (1) and (2) to get started.
HOW WILL IT WORK?
Basically, we will collect information that can be initially perused,
and, later, queried in more sophisticated ways. You might be able to
find another site that is either using a package that you are interested
in, or might have a similar problem that you are experiencing. Or you could
simply use the information to see what is, in fact, most used.
The database can't carry all possible info, but it can signpost the way.
ACTION:
If you'd like to join the database, simply send the following
information to us at bed at lenti.med.umn.edu
(1) A contact name, postal mailing address, phone, fax and e-mail
address;
(2) What main hardware and OS your site utilizes;
(3) What application software you run, biocomputing and otherwise;
* (4) What the range of your interactive service is: does it serve
* interactive users in a lab, department, institute, or nation?
* (5) Any suggestions you have for anything else to go
* in the database.
Options 4 and 5 are optional; information can be provided by a list,
a narrative, or whatever suits you. All levels of groups are invited to
participate.
Hoping to hear from you!
Ernie Retzel, University of Minnesota (ernest at lenti.med.umn.edu)
