The Crisis in Information Technology Support:
Has Our Current Model Reached Its Limit?
McClure, Polley, A., Smith, John W., and Sitko,
Toby, D., CAUSE Professional Paper Series, #16
 | problem:
 | increasing numbers of computer users in
universities |
| fundamental changes in higher education |
|
The Crisis Triad and How
We Got Here
Central organizations are overwhelmed by demand
| demands for:
| dial-in lines |
| web page accounts |
| faculty training |
| better classroom facilities |
| support of student computers |
|
More customers need more services
| up from 20% of faculty, staff & students to almost
everyone today |
Per capita demand for services has
increased
| possible to generate massive amounts of data now (MB vs.
KB a few years ago) ... leads to the need for faster
networks |
| contemporary users are not as knowledgeable and
fault-tolerant as were early users of technology
| demand for "complete products" |
|
Multivendor, distributed technology
requires high-level support
| lots of different technologies strung together ... leads
to need for much more maintenance and support |
Funding models are inappropriate
| most institutions attempting to offer computing for free
... difficult to match costs and benefits |
Support quality
deteriorates
| can't provide the support to match the demand |
Centrally provided primary customer
support does not scale
| primary support = dealing directly with the customer |
| secondary support = provision of information, resources,
tools, etc. to front line support people
|
| problem:
| now too many users to deal with - can't answer
all the questions |
| now too many technologies to deal with - can't be
an expert on everything |
|
Assignment of support responsibility is ambiguous
| departments go ahead and buy without clear understanding
of what they really need
| responsibility for maintenance is passed back to
the computer services dep't, but it has not had
input into the original purchase (all of the
blame with none of the control)
|
|
| some departments create their own support mechanisms
| but can't keep up with technology because they
have other tasks |
| aren't worried about how their own world connects
up to the rest of the university
|
|
| everybody gives up |
Distributed systems need special
support
| not possible to service local area networks from a
distance
| need for house calls means that some people get
service, others are left out |
| everyone is frustrated |
|
Every machine in the institution is
different
| since everyone has been making their own decisions, there
is a hodgepodge of installed equipment
| too many different kinds = unsupportable |
|
Central units are merging
| current trend toward merging of academic, administrative,
telecommunications, media services and libraries
| good point - overall understanding of the system |
| bad point - time taken to set it up, little
concern for any individual unit |
|
Central IT organizations are the scapegoat
| everyone has incorporated technology into their work and
they won't/can't go back to the old ways
| problem: they don't know the true costs |
|
| central IT organization is the "fall guy" ...
no way of avoiding it since most of the big money has
been going to it as opposed to individual departments |
Computing has become truly distributed
| the distributed aspect of the computing environment is
too big for any one department or central organization to
manage |
Technology and content are more
integrated
| with the hypermedia concept, the technology is both the
container and the goods in the container
| content creators (faculty) will not let the
carrier (IT deparment) control technological
decisions |
|
New users want authority, but lack
expertise to make decisions
| "point and click" users assume that they know
more than they really do about how to make a complicated
distributed network function smoothly |
| the large amounts of funding/growth in the 1980's is
expected by everyone to be continued in the 1990's |
User involvement in IT decisions is
insufficient
| average departments have not been involved in decisions
re technology deployment ... therefore they feel no
ownership and are more critical |
Elements of an Ideal
Technology Support Environment
A whole-product focus
The new "average" user
| must assume that the new average user does not want to
understand "why" or "how" technology
works - just "that" it works |
Integrated service, not more
technology
| present strategy:
| offer more capacity |
| training in use of specific techologies |
| answer any questions which arise
|
|
| proposed strategy:
| provide integrative assistance ... email example
|
|
| should provide "whole-product" environments:
| correspondence |
| teaching |
| grant management |
|
Consistent and reliable whole-product
environments
| very important when faculty depend on real-time web
access |
Problems with some current funding
models
| everyone expects that capability and capacity are free
... therefore a dramatic increase in consumption
|
| it is possible to use technology with very
little understanding of how it works, but it is not
possible to support it without knowing what you
are doing
| therefore a hidden support economy has been
created (e.g. the word-processing expert, the
hardware wizard, the web wizard, etc.) |
|
The true cost of computer support
| nobody knows the true cost of supporting computers in a
university
|
| many people do "computer" work without having
the work outlined in their job descriptions
| development of "seat-of-the-pants"
expertise is not very efficient |
|
Achieving a functional economic model
| need to do three things:
| understand the true costs |
| link the costs to whomever gets the benefits |
| distribute the funding responsibility between the
central IT department, individual deparments and
individual faculty members |
|
Measure costs and benefits to the institution
| cannot measure benefits just in terms of the cost of the
original equipment
| need to consider institutional goals |
| need to consider longer term issues like
maintenance, upgrades and replacement |
| need to consider cost of wasted time (e.g. in
repairing hardware, in upgrading software) |
|
Map the responsibility for costs to the
location of the benefits
| correspondence, library access, written reports, grant
applications, etc. are functions which need to be done by
everyone in the university
|
| specific features (e.g. Spanish language keyboard) should
be paid for by the department(s) and individual(s) which
will benefit from the feature |
Designate appropriate funding
responsibilities
| must move from supporting technology to supporting
institutional functions
|
| subsidize the following:
| technology which benefits the common good |
| technology with a large fixed cost |
| strategic services |
| technology in which consumption is encouraged |
| technology which allows for no user control |
| technology which is in plentiful supply |
| technology which is a renewable resource |
|
| distribute the costs of the following:
| technology in which the benefits are received
locally |
| technology which has a variable cost |
| technology which is an established service |
| technology in which constrained consumption is
encouraged |
| technology which is within the control of the
user |
| technology which is in limited supply |
| technology which is non-renewable |
A support mechanism focused
on customer needs
|
More complex support needs
| current support model is based on the notion that the
central IT organization provides the means and the user
implements
|
| problem comes when the user does not have the necessary
skills to implement what the central IT organization has
provided (e.g. the user has not studied Powerpoint or
does not know how to hook up a computer to the internet) |
The help desk next door
| give distributed support:
| support is as close as possible to the user ...
next door |
| department-based support providers know the
questions which are important to the personnel in
the discipline |
| they know the users personally and can provide
customized help |
| departments are in the best position to know what
they need |
| departments know which/when technology should be
introduced
|
|
| in order to succeed, a deparment-based support system
needs:
| people who understand the department and its
unique needs |
| people who are trained in accessing resources and
tools which can help them achieve their task |
| tools which are available to the support
personnel (e.g. PC diagnostic software) |
| a flexible approach to job definition (e.g. let
the department define the job) |
| responsible deployment of the support personnel
(i.e. their time is not monopolized by one or two
faculty members so that the rest of the faculty
members must turn to the central IT organization) |
| a very high quality expert service provided by
the central IT department (i.e. where an expert
level of service is needed, it is provided) |
|
Implications for central IT
organizations
| central IT organization should take responsibility for
training, consulting, tools and provision of the proper
baseline environment |
Getting smarter about training
| almost impossible to give full training in all software
applications ... a "just in time" approach has
to be used
|
| central IT department responsibilities:
| beginning of semester classes |
| instructional packages available for
self-learning |
|
| departmental support person responsibilities:
| be "up to speed" on most applications |
| determine which courses/modules to offer to their
own department |
|
Tailoring consulting services
| departmental support person responsibilities:
| first contact for all questions |
| decisions as to when a question is frequent
enough to warrant a training course instead of
consulting
|
|
| central IT department responsibilities:
| backup consulting for when the department person
is unavailable |
|
Developing technology tools for
distributed support
| central IT department responsibilities:
| make sure that it is EASY for users to access
things (e.g. getting into the library holdings) |
| develop and promote a macro library of
institutional information (for use by departments
in creating customized information recruiting
packages) |
|
Reallocating support
responsibilities and staff
| departmental staff are the primary support system ...
central IT staff create a reliable and robust environment
which can be used by departmental staff |
A reliable baseline
information infrastructure
| assuming no dramatic increase in availability of funds,
there is a need for a good information infrastructure
which everyone can use:
| defined by user needs |
| intuitively easy to use |
| consistent across different hardware and software
platforms
|
|
| should include the following:
| office suite |
| robust email package |
| web browser
|
|
| would be nice (but probably not possible) to support only
one hadware platform and one software suite |
| if possible, it is best to determine beforehand, the
major information needs, databases and information
processes which are required by most people in the
institution
| keeps focus on the important problems |
|
IT staff must manage environments, not fight fires
| unreliability is to be avoided if at all possible
| emphasis on fireproofing as opposed to
firefighting |
|
The infrastrucutre must be hierarchical
| from the base up:
| baseline = hardware platform, operating system,
networking, office suite, email, web browser
 | new technologies can be tried out in
special rooms (e.g. video conferencing)
and moved into the baseline structure
when necessary |
|
| departmental environment
| built upon the baseline environment |
|
| individual's environment
| those wishing additional capability can
add it at their own expense |
|
|
The new architecture must be
implemented with support from users
| if hardware cannot be repaired quickly by the
departmental consultant, it should be able to be replaced
at a central swap-out pool |
| environment standarization will reduce the problems
associated with turnover |
| should be possible for faculty to be able to use standard
electronic technology in any classroom |
"Seat" capacity
| the more students know about the standard equipment used
in the university, the will be better able to purchase
the right type of personal equipment |
| the more people using a particular type of standard
equipment, the easier it will be for all users to obtain
help from one another |
We Can Get There From Here
| faculty and students need to be educated on the true
costs of their demands (e.g. dial-in lines, high speed
access from home) |
Engage users in decisions
that affect them
| big decisions should be made by all constituents ... if
they don't understand the problem(s), educate them |
Redefine roles toward a federal
model
| there should be a dialog about the complementary roles of
the central IT organization and individual departments
|
| best model to follow is a federal model:
| central IT organization:
| encourages and supports a quasi-open
market economy in information services
(it does not make all the decisions by
itself) |
| uses policies and standards to influence
individual departments |
| manages key functions critical to all
departments |
| offers the basic technical infrastructure |
|
| individual departments:
| have freedom to exploit technology in the
best way(s) in order to maximize their
effectiveness |
|
|
Create effective distributed support
models
| models which might be useful:
| special group in central IT organization ...
clients = departmental support personnel |
| discipline-focused, physically dispersed
sub-centers |
| creation of an autonomous central department ...
clients = departmental support personnel |
| setup of expert systems and online training for
use by departmental support personnel |
| equipment/software/certification incentives for
departmental support providers |
| staff transfer from the central IT organization
to departments
|
|
| key point is that the individual departmental personnel
take responsibility for managing their own unit's IT
environment |
Mentor individual staff
| there is a need to retrain staff who have grown up under
the old system
| high technical functions concentrate on designing
systems for maximum robustness and manageability |
| less technical functions concentrate on designing
institutional processes, structures, tools and
incentives |
|
Recruit and replace, if necessary
| if mentoring cannot be done soon enough, need to replace
existing staff and recruit new staff |
Seeing Our Future in an Evolutionary
Context
Where we have been
| The Iron Age:
| mainframes, time-sharing |
| do work at centralized computer centers |
| populated by scientists and engineers |
| emphasis on technology, brilliance, individual
user's projects
|
|
| The Industrial Age:
| personal computers (and networks) |
| emphasis on customers |
| group orientation, small center support |
| help desk |
|
Where we are going
| The Information Age:
| develop skills to manage that which does not
belong to us |
| need to become focused on institutional
processes, instead of personal processes |
| have to partner faculty (content expertise) with
administrators (functional expertise) |
| have to realize that we are attempting to
"manage" things which are outside of
our control |
| support needs to be distributed, but not anarchic |
| have a whole system approach |
| think globally |
|
Back to Computer Needs Main Page
Back to Dr. D's Home Page
Last Edited: August 23, 2000
|
