Transforming Technology Support Services: A Proposal
Tue, 10/14/2008 - 1:57pm — memcinto
Name of proposed new unit, OR Title of submission:
Transforming Technology Support Services: A Proposal
131 Transforming IT Support Services.pdf
Name of contact person for this proposal: Michele Norin
Contact person title: Chief Information Officer
Contact Address:
P.O. Box 210073
Tucson, AZ 85747
Contact Phone: 520-621-5972
Responses from President and Provost
Response #1> Back to List of White Papers
If you do not wish to comment publicly on this site, you may post your comments at uatransf@email.arizona.edu
Comments
The African American
The African American Advisory Council would want to insure that Transformations processes including consolidations and workforce reductions would not erode The University of Arizona's efforts to recruit and retain diverse faculty, students, and staff.
The following memo was sent
The following memo was sent to all SBS faculty and staff as well as other decision makers on the campus. We welcome comments.
Mark Aldenderfer
_______________________________________________________________________
The College of Social and Behavioral Sciences Information Technology Advisory Committee has reviewed the white paper "Transforming Technology Support Services: A Proposal," by Michelle Norin, Chief Information Officer (CIO). Our consensus is that the plan will not result in a demonstrable improvement in the delivery of information technology (IT) services to the campus. We have two primary concerns: the assertion that centralization at the level of the campus is the best model for the delivery of IT support services, and the failure to provide an acceptable level of consultation, governance, and participation with the academic community.
Centralization and its presumed benefits
In every enterprise, be it academic or corporate, there is a constant tension between the advocates of centralized vs. decentralized models of IT service support. This is a false dichotomy. The real question that must be asked in any discussion of a reorganization of IT support services is what the appropriate level of centralization for specific IT services should be.
The white paper describes the benefits of centralization within the context of large business enterprises, not academic institutions. There is no question that the centralization and improvement of IT services for many of the administrative areas of the campus could be beneficial. But it does not follow that academic activities--the research, teaching, and outreach missions of the university--would likewise benefit.
Academic units have distinctive and complex IT support needs that are best served at the appropriate level of control and oversight. This may be within the department, program, or college. For instance, the College of Social and Behavioral Sciences has recently centralized most of its IT support at the college level. This model has resulted in budgetary savings and improved end user support. Further, since resources, equipment, and staff are appropriately situated, this model offers real flexibility to deans, department heads, and individual faculty members to respond to funding opportunities and innovations in pedagogy. Lines of responsibility of the IT staff are clear and well defined. Our experience in this process demonstrates the merits, but especially the limits, of centralization. We've seen firsthand that individual and unit-level needs can only be met with reasonably proximate administration of support.
In contrast, the plan offered by the CIO offers none of these benefits. It creates an inefficient, inflexible, management structure that has little capacity to respond in a timely manner to the complex and ever-changing needs of the faculty and staff. Because IT service providers will report to a distant central entity, not the local academic unit, end users will have little influence or control over how resources are deployed. Lines of responsibility and oversight of IT staffers will be blurred and confused. It is difficult to believe that such a bureaucracy will deliver better, more efficient service to its stakeholders than do existing, local support groups.
We believe that the majority of local IT support groups across the campus are successful and serve the academic mission of the university in an exemplary manner because they are responsive to the local needs of their users. They already use their limited resources efficiently. What is the benefit, then, of creating a large bureaucracy with monopoly control over IT decision making when it is acknowledged in the plan itself that aggregate resources may not be adequate? As we have argued above, the presumed efficiencies for the academic units are unlikely to materialize and that the quality of service will actually diminish from current levels.
Consultation and governance
As originally presented, it is obvious that no academic unit at any level was consulted in the development of this plan. We have recently learned that the CIO is now constituting a committee (based upon a recommendation by SPBAC) composed of a variety of end users, including representatives from the faculty, the research community, students, and existing IT support units. The charge of this committee will be to study the proposal and determine, which, if any, IT support services could be usefully centralized. We applaud the formation of this committee, but wish to emphasize that any reasonable plan for IT reorganization has to take stakeholders seriously and find out what it is they do and how they do it. Existing IT deployments must be studied and clear arguments offered as to why reorganization would be of benefit to those stakeholders. Again, our own recent experience with the SBS IT reorganization has demonstrated that the widest possible consultation is essential. As our own experience has shown, this is a time-consuming process that not only looks at lists of services but also at structures of governance and oversight. Here, the active involvement of the faculty is essential. Given the complexity of the academic mission of the university, it is unreasonable to give a campus-level non-academic unit full responsibility for the deployment of IT services across the university with no faculty oversight whatsoever.
Conclusion
An appropriate, well considered approach to centralization may well have potential for improving the delivery of some IT support services for the university. This plan, unfortunately, does not have that potential. It creates a massive bureaucracy with no academic governance or oversight. It will not improve the delivery of IT services for the vast majority of end users on this campus, and thus, it should not be implemented.
We welcome a campus-wide discussion that avoids the false dichotomy of advocacy for either centralized or decentralized IT support services. We believe in layered models of IT support wherein a blend of local and non-local management, resource ownership and deployment, and control are realized. We recommend the SBS reorganization of IT services as a model for initiating this discussion and we are happy to share our experiences in this regard in this time of crisis.
Respectfully submitted,
Mark Aldenderfer, Chair, SBSIT Advisory Committee and Professor, Anthropology
John J. B. Allen, Distinguished Professor, Psychology
Joseph Bonito, Associate Professor, Communication
William Dixon, Professor and Head, Political Science
Martin Frické, Associate Professor, SIRLS
Michael Hammond, Professor and Head, Linguistics
Veronica Peralta, Business Manager, Mexican-American Studies
Sally Stevens, Executive Director, SIROW
The UA President’s Hispanic
The UA President’s Hispanic Advisory Council (HAC) is pleased to provide its evaluation of this White Paper/Proposal, with the specific goal of relating it to the joint UA/HAC goals concerning diversity and inclusion, recruitment, retention and graduation of Hispanic students and faculty, and on meeting the UA goal of becoming a Hispanic Serving Institution by 2012. Additional criteria used when reviewing this proposal include the following: UA’s land grant institution status and thus its location in the southwest and the changing demographics of this area; innovative program design, including instruction methodologies; and whether the proposal realistically addresses the UA’s business needs. HAC’s mission it to strengthen relationships between the UA and the diverse communities within the State of Arizona by serving as a communications conduit and developing mutually beneficial partnerships.
HAC has reviewed this proposal and offers these comments:
Revamp - there is little to no mention of the impact on students and instructional methodologies in this proposal (related to any students, diverse or others.) It is unclear whether this proposal realistically addresses the UA's business needs, including its need to fulfill its land grant mission or other critical missions for this area. Additional work is needed to address these and other criteria of importance to the Hispanic Advisory Council.
This white paper should
This white paper should clearly state the ultimate goals of a university, that is, to better support research and teaching. With this common understanding, we can discuss "centralized" or not.
There are needs to centralize certain IT operations, but "Rome was not built in a day". I would suggest discussions/actions should be taken: what IT services can be centralized first (agreed)? What will be the next? How do we define success? How do we measure users' satisfaction?
Here is my personal list for centralization, but more discussions are needed before taking actions.
1. email / calendaring:
I think that email servers/calendaring/web conferencing (in majority of cases) need to be centralized. I have 2 email accounts at UofA. Sometimes this can be confusing for me and my colleagues. I have to email/call several times to schedule meetings with faculty and staff in other departments.
2. telephone/data:
Anyone against?
From what I read/hear, some of negative comments are from unhappy (if not bad) experience from CCIT. Personally as a faculty member I find UITS's customers' services are not satisfactory. Recently I called 10 times and left several messages, but nobody replied! Without getting hold the real person on the phone, it seems to me that I even won't get an update! UITS needs to build trust from faculty and staff. UITS must improve its services to at least reach current standard that faculty and staff have.
My concerns have to do with
My concerns have to do with actual cost savings vs. shuffling costs between departments and a central unit in a way that actually increases the overall bill. It all comes down to how the IT function is paid for. Would it, for example, be payed from central funds, or billed to the departments like we pay for the phones? This might be a problem. When the phone services were reorganized. We were all told repeatedly that "we would love the new level of services we would be provided." However, my department had already determined it couldn't afford the full range of services and had cut many of them, whether we loved them or not. When the service was reorganized, we paid whether we could afford them or not. This removed the department's ability to prioritize what little budget we had. I would not like to see this extended into the computer realm, since we are currently paying locally (in personnel time) for only the functions that are essential to our unit.
The flip side of this might occur if we have IT processes (and we do) that no or few other departments have. Would we have to pay an additional amount to cover these uncommon functions (e.g., HIPAA compliance, tracking student certification requirements, clinical operations software)? Maintaining these functions are not up to us, so someone has to cover the cost. If the centralized IT doesn't serve these more specialized functions (as suggested by the document), it is more cost effective for us to just pay the people we pay now, who handle both the more general IT jobs as well as these specialized functions.
Another concern has to do with the ability of the department head to set priorities for IT personnel. With local help, I can push jobs up in the queue that are critical to our functioning or revenue generation. In a centralized system, I would have to argue that my department's immediate needs may need to trump someone elses. Who arbitrates that? If we look at other examples of centralized services (ORCA, Sponsored Projects, Human Subjects) the answer is an office staff person who is frequently ill positioned to make decisions that may ultimately affect student services or revenue flow.
A final concern is that, as humans, we tend to form loyalties to small groups of people, not institutions. I wonder if the IT people will feel disenfranchized from the units they serve and start to be treated more like the copy repair person than a valued member of a unit.
All this may seem a bit negative. It is actually just a reflection that if this proposal goes forward, the devil will indeed be in the details. There are many ways to get this wrong, so procede with extreme caution and get a whole lot of input on what departmental IT people are currently doing with their time.
Let me start with a few
Let me start with a few points about the Transformation Process that may have been overlooked:
“All of the White Paper Proposals are to serve as discussion starters. The Transformation Process is about how we can improve our work together, and our success will be shaped by the quality of our deliberations more than by the ideas that are suggested in the first stage of the process.” This quote is from the provost’s site on “UA Transformation Information and Communication.”
Most of the comments on this white paper are not about improving or transforming the UA into a top 10 public research university; they are about keeping the status quo. President Shelton said, “The status quo will no longer work.”
Of all the white papers, this one has elicited the most comments because it will have a broad impact on the UA. I believe that is what President Shelton was looking for when he said, “The time has come to take bold action that will radically change the way we operate.”
If this white paper is a starting point that will change the status quo and have a broad impact on the UA, then we need to have more discussions and deliberations on the idea of changing IT on campus. The discussions and deliberations should not be on whether to centralize or not but on where to centralize. Some IT services should be centralize at the campus level, some at the college level and others at the department level.
This white paper should go forward in the process.
I have concerns about this
I have concerns about this campus-wide IT proposal. These concerns are as follows.
1. To implement without any review and input is NOT acceptable. In the spirit of shared goverance -- Faculty Senate, Staff Advisory Council, Appointed Personnel Advisory Council, students and other campus consituencies -- review and input must be sought.
2. To implement within two and a half months does not seem to be well thought out especially if there is no review and input.
3. The 'one size fits all' situation may not be the best approach. (Even today, one-size clothing labels read, 'will fit most').
4. The SBS College implemented, less than a year ago, a college-wide IT support services approach. Give the college-wide model a chance to prove itself. A campus-wide model may not be the best fit for our campus.
5. Any campus-wide situation will require adequate funding. Will departmental and college resources be swept again to fund an unproven model?
6. If resources are swept again to fund this campus-wide IT support services model how and when will this be done and who (i.e., current IT personnel) be affected directly, and its affect on the campus community?
My department grave concerns
My department grave concerns about the feasibility of this project for the following reason:
General comments:
· Applicability of this plan is limited to specific institutional processes where there is significant functional overlap and resource redundancy. It simply does not apply in many departments; especially those that are research and outreach intensive.
· Efficiencies to be gained and resources to be consolidated are theoretical at this point.
· The potential disruption to information systems at every level is huge and unwarranted given the lack of supporting analysis.
· The extremely short implementation timeframe makes real analysis and planning impossible.
Invalid assumptions:
· That the current central / distributed model on campus is the wasteful result of non-standard, un-prioritized spending, when it is really the result of decades of coordinated streamlining and honing efficiency to meet ever-shrinking budgets and ever-growing demands;
· That there are many departments with virtually identical information systems needs;
· That IT staff are interchangeable resources;
· That a client/consultant model will work for supporting research, education and outreach computing needs;
· That all departments have state budgets supporting equipment refresh (i.e. budgets that can be centralized and redeployed)
Negative outcomes:
· The loss of control over mission-critical personnel (hiring, termination, prioritization) represents a serious threat to departmental IT efficiency and stability, and has the potential to negatively impact all areas of the unit’s functions; ultimately diminishing effectiveness in support of University goals.
· Centralized IT-support priorities fluctuate with each new University administration. Departmental IT staff minimize the impact of these fluctuations by sustaining effective local systems. This buffer would be minimized or removed under this plan.
Centralizing management and
Centralizing management and forcing a “one size fits all” IT service for the entire university is not a solution for transforming academic and administrative support functions. In Cooperative Extension, we are disseminated across the state in fifteen counties, on tribal lands, and agricultural centers. The current structure that manages our IT services in Maricopa County Cooperative Extension ensures compliance with ABOR and university policies and federal and state regulations. The quality support that we receive in our office and through the College of Agriculture and Life Sciences enables the faculty to conduct and deliver its extension programs to the community and clientele. Creating and establishing service level agreements will lead to greater authority of an IT service unit influencing and driving the research and outreach education of Extension or activities of other academic units of the university. Creating greater bureaucracies to create inefficiencies will diminish our ability to perform our jobs in remote locations from campus. There should be more thought given to any new structure with authorities, responsibilities, and accountabilities clearly defined within the context of being a first class academic institution.
Congrats to Michelle for the
Congrats to Michelle for the paper with the most comments. Wow. I take a bit of exception to the previous comment on engineering and I believe that there are more strengths in this proposal than people are seeing. The College of Engineering actually had 7 independent IT groups plus a college group that handled Admin and one department. Each group has its own organization based on the uniqueness of its faculty/staff/student clients. Desktops were windows based and infrastructure was both unix and windows based. Each group had its own storage arrays, own back-up, own authentication, own email, own software and hardware purchases, own student PC lab, and own research computing infrastructure. Each program has IT staff doing desktop support and infrastructure management (largely identical and some much better than others). Finally, we have a group of faculty that insist that they know better and manage their own machines with the help of their grad students. One would have to be blind not to see opportunities for cost savings with all of this duplicate work. Through a moderate pain process, we merged 4 of the groups (4 others still remain) and asked UITS for help (after about a year of more pain). After a series of weekly meetings with UITS staff, a plan to reduce infrastructure and use UITS services was developed and is now under implementation. I think that our largest gain was from using UITS windows authentication (catnet) and this alone eliminated something like 20+ servers from our infrastructure. We still do not do research computing but we are trying to move in this direction. We are undermanned in support (as is almost everyone else based on normed workloads for support), but we are now using more automation and working more with the OSCR 24/7 group to be our first line support and to increase our knowledge base to solve problems faster. Are we perfect - no and far from it. Have we had some bad bumps - yup. Are we on a better path to success - clearly. As we gain more knowledge we will be able to gain more flexibility (and move to other platforms as the previous emailer suggests we need) but right now, we can barely manage Windows boxes well. For example, we now have software that will enable users to load software without admin access (the major issue in locking down machines and the need to use admin privledge) and we are installing when faculty want this feature. So, why do I like parts of this white paper (this is not yet a proposal as there are no details of implementation and we are not yet in that part of the process). - I think that UITS is the strongest group in computing on campus. this is their business and not my business. I would think that they will always be better than me at this and hence I would rather us my hours doing things that I am good at. This is not my core competency and I do not believe that I can hire this any cheaper than UITS is doing it already. - I like what some centralization can do for students. I love the idea of students going to any PC lab on campus and being able to log in to a central storage area and getting their standard set of software. We have done this for 2 labs in ENGR and are looking at a 3rd. UITS/OSCR should be managing all PC labs on campus using a standard set of procedures. At some point, we will move away from hardware based labs and everyone will have a mobile device. The goal then will be to provide connection and software. We already do this with some of our engineering packages. - I do not want to secure and monitor my servers (data). It is too much risk if something goes wrong. I do not have staff or an automated system to do monitoring so if there is a breakin, I do not discover fast enough. I do not want to be responsible for backup - I would rather have a reliable central back-up system that is affordable. The people that do not worry about security are the ones that are not responsible when something goes wrong. - I find it hard to manage IT professionals as I know a small amount, but not enough to schedule/prioritize tasks for employees. I can barely evaluate employees as I cannot evaluate good designs and good alternatives. I am not up on state of the art and this was made painfully obvious during the past two years. Again it is all part of "let the best people do the job" for me. I do not need UITS to teach engineering design and recruit students - I will work on those. There is more that I like in centralization - software purchasing power for one. But, there are clearly other areas where you have to keep things local. There is nothing in this white paper that forbids this. In Engineering we have found a great partner in UITS an we are working to get better. We still have 4 local groups and even within the new group, we still take responsibility for 2nd level desktop management and some server management. Closing - much of what I am talking about is what I call "operations." We are doing so much operations, we can barely even think of anything strategic. At some point, IT must be a strategic advantage for the campus. It is a potential strategic advantage for recruiting faculty, recruiting students, and obtaining grants. I see little hope of the UA getting to this level when IT strategy has to be done all over campus. Many of the notes in this thread talk of differences, but my guess is that we have a lot more in common than most think. By the way, this proposal will really help the "have-nots." Providing a minimum level of IT support for everyone is a reasonable goal for this University.
Academia inherently invokes a
Academia inherently invokes a hierarchical computing infrastructure that includes laptops, desktops, workstations, research and administrative servers, high performance computers, and teraflop super computers. Moreover, education and research are enhanced by the very essence of the diversity in hardware and software (e.g. operating systems) platforms that are available throughout the world today. To believe that a corporate computing and IT model should be adopted in academia is myopic at best. The current proposal other than being rather incomplete, fails to embrace the computing diversity that is needed, especially if the University of Arizona is to fulfill its mission of accomplishing world class research and education.
A more thorough proposal must engage the faculty, staff, and students to more clearly determine where the real savings are, not just in dollar amounts, but in providing the resources and support that lead to developing the best educational and research programs on this campus. The diversity of the IT support staff throughout the campus should not be compromised as they provide essential services that are often focused on the special needs of research and education. Any attempts to reduce the available platforms (and operating systems) will lead to reducing the creativity and productivity that is essential in growing the university to new heights. The worst example of control of IT services is exemplified in the College of Engineering where support for anything other than Windows platforms is nonexistent and completely ignored. Moreover, the paranoia of IT security is leading to less productivity instead of developing better means of controlling unwanted traffic. [It reminds me of clogging the toilet to stop flushing unwanted sewage to the sewage station, meanwhile, you can’t dispose of the regular sewage.] If the rest of the university adopts a direction of reducing resource diversity and reducing access to computers, our computing infrastructure will fail in achieving its real mission.
A final reminder to all is that IT should not and must not dictate how education and research is carried out on this and on any other campus. The principal educators and researchers are the faculty, and IT should play a strong role in such endeavors while not handing out shackles and placing blame for security issues. As long as there are hackers there will be security compromises, otherwise, life would be boring and there would be no IT security challenges.
I am a faculty member in
I am a faculty member in Family Studies & Human Developement in the College of Agriculture. I agree with those commenting above that this proposal should be vetted through Faculty Senate just like the rest. I also agree that this proposal seems very likely to make things more expensive and less effective, not the opposite. My research depends heavily upon the decentralized IT support that I receive in my department. I run a computationally intense psychophysiology lab and often need immediate problem solving of both software and hardware. A centralized service could not provide the level of responsiveness that exists now and could significantly impede my research productivity.
I am writing to express my
I am writing to express my concern regarding the above sited proposal in its current form. This very broad reaching proposal may have some merit, but by and large it will be extremely detrimental to the activities of BIO5. I am genuinely concerned that the design, motive and rationale listed in the proposal lacks understanding of how our research-intensive units function and the essential role for IT in our research missions. This is exactly the type of proposal that will do major harm to our university’s ability to continue to pursue big science.
I have no doubt that there is duplication in IT personal and function across campus and that broadly speaking this is an area where some consolidation could make a great deal of sense and provide cost savings. It is likely that aspects of this proposal will have merit for departments with limited IT needs and little to no IT infrastructure. However, I am convinced that making the majority of IT positions and budgets report to and be managed by UITS will have a severe detrimental effect on current research activities within BIO5 and with our ability to continue to compete for large projects requiring significant IT infrastructure. The current research projects that will likely be negatively impacted within the Keating building alone include over $10M per year in grant funds. These projects represent major genomics and cyberinfrastructure programs with very tight benchmarks that need to be met, and which depend entirely on the extensive IT infrastructure and personnel for success.
I recognize that there are aspects of this proposal that may allow BIO5 to fall into the category of "Units with complex needs” that may “require dedicated, locally managed resources". However, my reading of this proposal is that it is ill conceived and could not have included in its formation research-intensive units with very complex IT needs. I urge SPBAC and the Provost to reject this proposal in its current form and charge the CIO to develop a more inclusive and well articulated proposal that includes the needs of units with large IT needs. This will require extensive consultation and buy-in with and from those leading and implementing large IT infrastructures. By working together a plan and vision can be put forward that will meet the goals of saving money, while also enabling our researchers to pursue ‘big science’ and take the university to the next level. To not do so runs the risk of destroying any trust the UITS leadership and staff may have with our leading research units.
In today’s research environment where interdisciplinary team science and virtual organizations are ubiquitous, it is paramount that faculty and departments be agile and effective in marshaling both institutional and departmental resources to effectively compete for funding opportunities. By centralizing the IT talent pool and resources we run the risk of destroying the ability of self forming faculty groups to pursue funding opportunities that necessitate creative and innovative use of IT at multiple levels to address computationally driven research questions.
I don't understand the reason
I don't understand the reason for the proposal when the chart says that IT expenditures have been decreasing since 2006. It seems that if you wanted to encourage this downward trend you could simply point out a few successes/reasons for the downward trend and look for simple ways to expand on this. The proposal as it is written is poorly concieved.
Careful consideration must be
Careful consideration must be given to removing specialized IT support from departments. At least in the college of science there are a great number of unique task that department IT specialist have been asked to perform. Centralizing and removing this very unique and diverse expertise could create additional cost burdens on research being conducted here at the UofA. Please take into consideration that most of the IT specialists within the college of science do a lot more than what UITS is equipped to do.
I have read this proposal
I have read this proposal with care and think that it will, if implemented, make matters worse rather than better. As far as I can tell it was conceived and written by the Chief Information Officer without consultation, and thus reflects a view of campus IT needs that is actually quite removed from the situation on the ground.
Several other respondents have made the point that the IT needs of the various units vary greatly, and are thus better served by a decentralized structure of IT service. In my own Department (Anthropology) we are well served by the present system, in which an SBS IT technician is stationed in the building and responds to requests for assistance booked through the Ticketdog system on the SBS web page. This system is fast, flexible and efficient, allowing one person to service some 35 FTE of faculty, plus the Departmental staff and about 160 graduate students. I frankly do not believe that centralization can maintain the same levels of service. Reading between the lines of the proposal, it is clear that the author believes that levels of service can be maintained with far fewer people and servers - this seems to be her benchmark for efficiency. My own benchmark would be the amount of downtime that faculty must endure while waiting for someone to come and fix an IT problem. It is my strong suspicion that if this proposal were to be implemented, any measurable gains in "efficiency" by this metric would be more than offset by inefficiencies elsewhere - the downtime while faculty and staff wait for someone to show up. Furthermore, the proposed system would replace "in-house" IT personnel who know the needs of particular units with people from a centralized office who are not familiar with those needs - thereby introducing additional inefficiencies.
This proposal is an illustration of the central problem with this whole transformation initiative. By calling for everyone to submit proposals simultaneously, Shelton and Hay have failed to recognize the interdependence of units. Not surprisingly, they will get a whole lot of proposals like this, which are written to make the unit submitting it look as lean and mean as possible, while ignoring the negative impacts that the proposed actions would have on other units. This approach can only create more problems than it solves. It would be far better for the Deans to mediate between the units and the upper levels of the Administration, talking among themselves and going back and forward to the units in an iterative fashion.
As a Department Head, I can
As a Department Head, I can recall two instances where centralizing/universalizing IT services has cost my department money. The first involved a university agreement with Gateway computers. During that era, my research showed that I was paying prices that were at or above retail for the purchase of computers. Second, when CCIT/UITS standardized and bundled the delivery of phone and network services, my annual bill increased by several thousand dollars, and it appears that we will be paying that added cost in perpetuity. At the bottom of these two remarkable failures at attempted or promised cost savings is a simple fact: If left to my own devices, I could secure computing equipment or telephone services at a far better rate than what the university was able to collectively negotiate and deliver on behalf of all of its units. For this reason alone, one should not put any faith in the wherewithal of UITS to realize actual cost savings at the department level. If this centralization plan is implemented, what would stop UITS from changing any figure that they see fit for technology support services, one or two years down the road? If one does not like their fee structure or cannot afford it, what recourse does she or he have? Forcing the entire campus to do business with a single centralized university-sponsored monopoly is not good business for the end user. In my view, the most likely effect of this plan would be an increase in revenue flowing into central administration, absent any clear plan for how that would translate into cost savings at the college, department, and end user level.
This service should not be
This service should not be mandated by the university but instead it should only be offered as an option. If superior service can be given, at the same or lower costs as current departmental IT service, then the departments will embrace the plan as an efficient solution.
One further thought (see my
One further thought (see my longer comment below).
The main justification for this proposal is cost savings. Yet the proposal suggests implementation 10 weeks from now without having actually analyzed or demonstrated cost savings.
Further it is hard to see where the cost savings come from. It appears that about 1/2 of the IT budget is outside UITS, but lets think about where this comes from.
1) A good chunk comes from departmental budgets - yet departments have already been cut to the bone and most have released staff in the last few budget cuts. Any remaining IT staff in departments are presumably considered among the top priorities for the department and already operating at top efficiency (I know the sys admin in my department keeps more balls in the air than it is reasonable to expect). There is no savings here.
2) Another good chunk comes from external research grants. Aside from the fact that there is no standing to centralize this, and that PIs are notoriously stingy and efficient at maximizing output from their grant dollars, centralizing IT will actually decrease net revenue to the university.
Take as a further example, an idea that probably does make sense - have a centralized email and calendar program like Microsoft Outlook/Exchange or similar. This is going to cost money up front. Where will the savings come from? You can't take 3% off a departmental sys admin who probably is only spending 3% of his/her time on email. The hypothesized savings of centralization often prove elusive. And when they occur, they occur only if staff are cut and it is extremely rare to cut resources without degrading service.
I would want to see a very hardnosed analysis of where the savings are before taking any radical, overly hasty action that could very negatively affect research, teaching, and administrative function.
I strongly recommend against
I strongly recommend against centralizing IT support for the entire university. Specialized research programs requires specialized IT support and consolidating the already knowledgeable and capable IT support personnel within individual departments with a general "pool" of support technicians would significantly reduce the efficiency of both IT support and the research that depends on it. Would you combine the police department, fire department, emergency medical response department and road maintenance departments simply because they all wear uniforms and drive vehicles with flashing lights? Furthermore, I am suspicious that such a consolidation of IT support would offer any financial benefit: Our research group would still require a full-time IT support person whether they be formally employed in the department (Geosciences) or UITS. Where is the cost savings? Finally, and somewhat incidentally, I dread the day I have to submit my IT support request online and wait some unknown time in the queue for some unknown person on the other end to (hopefully) look at it.
As a faculty member in
As a faculty member in Geosciences and a user and small-scale developer of iT resources, I think this proposal is ill-conceived and inappropriate. Intelligent centralization of resources to save money sounds attractive, but it has to be done right (that is, carefully) to avoid severely damaging the research mission of the University. If we fail at that, any hope of achieving or maintaining top-ten status in any field is gone. I've served on high-level University committees for 16 years, and this is the worst proposal I've seen.
I thought the transformation
I thought the transformation of the U of A was about positive change. Centralizing IT is certainly NOT positive! Most of the Geosciences department depends on specialized IT personnel that were self-trained to fit each group's needs and to respond promptly to a problem. Not having access to specialized staff would require the researchers to spend more time waiting for IT issues to get solved and would most certainly reduce productivity time. I see centralizing IT as a step back in this transformation.
"The potential for overall
"The potential for overall savings is difficult to demonstrate at this point in
the process."
Perhaps the authors of this proposal should have elucidated on the potential cost-benefit before submitting it to public scrutiny.
I agree that the centralization of email and web servers and other lower-level IT functions is a logical and necessary response to the budget cuts. However, it's completely reckless to consider consolidation of all department and research-group specific IT services. The level of specialization and effectiveness of these independent units will be difficult to reproduce from a centralized source, when computing resources are managed and modified on a nearly continuous basis by research groups in the College of Science.
Given that this proposal has
Given that this proposal has elicited more comments than all the other white papers combined, it would appear that there is great interest by many in the campus community to engage in a thoughtful and rigorous analysis of the centralization option. Unfortunately, such a process is precluded by the ill-advised implementation schedule suggested.
I have great concern that
I have great concern that some centrally-imposed withdrawal of perceived decentralized IT funds (many units do not have an "IT" budget, but may be paying IT staff) may simply result in added cost to departments that need dedicated IT support in cases where this newly empowered central unit deems they do not.
If UITS makes a good case for efficiency and service - then there is no need for a policy change to direct this. Departments who can utilize their services will do so as it would be meet their needs while being cost-effective and efficient.
If the UA needs to compel use of UITS and force budget redistribution - this would indicate an insufficiency in quality and service. In other words, offer relevant, high quality and cost effective services and departments will choose to use them so long as it is optimal to do so. Let supply and demand work (i.e. "if you build it, they will come").
This is a very bad idea for
This is a very bad idea for many reasons but I will mention only two here. First, it is simply wrong to suggest that this proposal is purely administrative, does not affect what faculty do, and is therefore not subject to faculty governance. In today's world virtually all teaching and research activities involve information technology resources. This proposal definitely needs to be heard by SPBAC and the Senate. Second, centralization will hinder those teaching and research activities we want to encourage. My college, SBS, took a full year to plan and then implement IT centralization. I acknowledge there is not unanimous opinion on the outcome, but from my perspective the result is a dramatic drop in service and satisfaction and the rise of stealth IT support throughout the college. By stealth I mean that people are relying on tech-savvy faculty and students for advice rather than go through the cumbersome formal procedures. Of course the result is lower productivity for those students and faculty. Full campus wide centralization would be a genuine disaster.
I came to UA from another
I came to UA from another (not so prestigious) university where something very similar was tried. I, and everyone else, fought it. But three years later, I had to admit I'd been wrong. It worked, saved the university lots of money, and we ended up with far superior service. As a result, that university is much ahead of UA with regard to IT.
I believe that although
I believe that although centralization of some tasks has merit, the UITS proposal does not address the specific needs of technology-heavy departments. In my department(s), the departmental computers serve many tasks which will unlikely be well-served by a centralized, one-size-fits-all, service focussed on web and email. I think this proposal seems to have been written in a hurry (no doubt, most white papers were). UITS is already proposing to central all business tasks with "Mosaic" and I imagine this task in itself is sufficient to overload UITS. The added tasks will just make it worse. Where's the cost saving here anyway?
I am concerned by the
I am concerned by the potential for financial waste inherent in this proposal. A current example of this waste is the fees paid by my department for optical scanning of tests. The fees for scanning and the cost for paper are far too much for us to pay. We can achieve the same results, doing it ourselves, at a fraction of the cost charged us by another UA program.
I believe that no one in UA administration is regulating these internal fees. Fees paid for administration of accounts seems to be another example. Perhaps no one in UA administration is aware that these fees exist?
I fear that the IT proposal would do the same, and this would constitute another unfunded mandate from central administration. I fear that we would be charged much more for IT than we currently pay, but we would not be given additional means to meet the increased cost.
I’ve had a chance to stew
I’ve had a chance to stew over the proposal for “Transforming Technology Support Services” for a week now since Michele shared it with us, the Deans’ Information Technology Council (DITC). I was also part of the DITC meeting last Tuesday where Michele tried to present some of the thinking and data behind the proposal, a presentation that she did not get a chance to complete. Here are some of my ramblings..
The proposal’s goal is to “transition a majority of technology personnel, equipment and operational budgets currently managed and performed within colleges, departments, and support units to the responsibility of the CIO, ...”, which has led to the IT personnel understandably being upset (about 100 messages in the NetDiscuss listserv) with the prospect of having to report to a new boss or being laid off in the process, along with the affected deans and department heads that will have their IT budgets swept up and their IT services provisioned via a service level agreement (SLA), which they’ll have to “negotiate”.
In trying to take a big picture view of all this, I suppose if I was someone coming in from the outside looking to make an organization run more efficiently, I would look at the cost centers, like IT, HR and business/administrative services. Perhaps from the big picture viewpoint IT can be categorized as a cost “center” but when you look a little closer here at the U of A, IT is pervasive and distributed to the nth degree, from data collection for a research project to communicating with students about an upcoming assignment. It’s even being played out on YouTube and iTunes. All these activities are being supported across our campus in central and distributed ways depending on the IT service. The premise of the proposal is that central IT can provide better service and at a savings due to the efficiencies that can be brought to bear through standardization and volume purchasing power.
What I haven’t heard a lot of in the past week is how this proposed change might improve the performance of our faculty and students making this institution better. From a researcher’s perspective I would surmise that having to work with an IT person who works for UITS based on a SLA negotiated by the dean or department head might be a bit frustrating. Suppose the research takes one down the path not covered by the SLA? How would students that require special software and support being provisioned locally be supported – through the 24/7 help line? I can see improvements for those departments that are unable or unwilling to properly fund IT support and services. What this proposal will do is to bring up the level of IT service for those “have-nots” at the cost of bringing down the service levels of those units that have invested in IT and view it is as a strategic differentiator.
There are a lot of undercurrents to this proposal based on the fact that it was not properly vetted with the affected parties: from Provost’s website re Transformation Plans – “All proposals are to include input from faculty, students, and staff.” The fact that this proposal bypassed the above premise has lead to speculation that this proposal is 1) a power grab 2) an attempt by the provost to rein in resources centrally and 3) we’re finally succumbing to what ABOR has wanted us to do for the last few years – to centralize IT. I’m sure there are other themes that have been bantered about.
One of the goals of the Transformation effort is to better the research infrastructure. I believe that our research enterprise is where it’s at in part because of the distributed nature of IT. Pulling it in centrally and divvying it back based on an SLA will stymie progress and innovation.
There are some things in the proposal and specifics that Michele presented at the DITC meeting that can be turned into tangible projects that can save the institution dollars. I believe that a formal project to replace the current email and calendaring systems with one that will satisfy the functional requirements of the large number of departmental email systems users on campus with an accompanying policy stipulating that units must use the UA system unless a formal review and exception is granted has merit. Additionally, a process for volume software license purchasing where the institution spends millions of dollars annually should be formalized. Working together on projects such as these can also lead to new possibilities for collaborative efforts or additional cost savings and efficiencies.
Post new comment