UGANET, the organization officially recognized by the Campus Information Technology Forum (CITF) as the council of individuals on our campus whose task it is to develop and support our information infrastructure, held a special meeting on June 30, 1999 to again condense and clarify long-standing IT concerns. This meeting was only the most recent step in over two years of direct efforts by UGANET, working with the CITF, to identify our most pressing technical issues and support administrative efforts to address said problems. Our specific reason for calling the June 30th meeting was to identify our most pressing problems, and discuss some possible solutions to address these.
We feel that we offer a unique perspective regarding many of these issues, and would like to offer our common perspectives and experiences, shared by our more than 200 members. The membership of UGANET is made up of the technical professionals working directly with the faculty, staff, and students on this campus to support and maintain the technology infrastructure and services. We are also the direct source for much the information included in the assessments and goals documents with which the IT community at UGA has been working. Our goal is to obtain and maintain technological excellence on this campus, not just survival, and it is in this spirit that we offer the fruits of our discussion.
The issues discussed by UGANET as being core to our technological success tended to fall within four areas. These areas (along with the relevant items addressed in each) are:
Potential solutions to problems that were apparent to members present in our special meeting have been included in the following synopses.
UGANET is concerned that the development of the network that our campus will rely on over the next two decades is not being adequately funded. The core of this issue is not only the lack of direct funding for our fiber-optic campus backbone (the VENUS project), but also the lack of unified planning and support for developing and maintaining networks within the buildings themselves. This has meant that the campus network, unlike other campus-wide infrastructure resources, has been built piecemeal with the limited available funds within departments. Instead of the campus developing a consistent network resource, this has had the effect of leaving many building facilities on campus in the technological dark ages, while some have been able to develop utlizing modern technologies.
The lack of funding for the completion, maintenance, and support of the campus network can too easily remain the most critical concern facing UGA in the coming years. Indeed, while the VENUS project creeps along slowly due to its lack of central funding, there has been no discussion or funding priority placed at the University level on updating the networks in the approximately 55 buildings on campus which are too antiquated to be connected to VENUS in a beneficial manner. There is a separation between the campus network and our building networks as far as funding and responsibilities are concerned. This means that even with adequate funding for VENUS, our new network will be effectively unusable by a third of our campus. The funding for renovations to building network infrastructures has been drawn from existing operational budgets, instead of being planned for and supported by the University as a whole.
These funding procedures appear to be caused by a lack of administrative understanding as to how utterly critical our data networks truly are. Buildings can and do continue to function when telephones or water are lost for hours at a time, but losing network and system access is akin to losing power for many units on campus, in that all work stops. We rely on our campus network for everything from basic communications between students and professors to purchasing and administrative record keeping. Yet there is no central funding for its support and maintenance. Indeed, networking is classified fiscally as an expense, not a resource. It is no wonder that there exists a desire to minimize this expense as much as possible, instead of providing the best resource possible.
Campus networking truly deserves to be funded in a structured manner. This would mean that instead of paying for networking piecemeal, renovations and expansions could be planned in advance and paid for by discrete budgeting, instead of when networking needs become critical. There is at this time no discrete funding within the University for campus units to renovate their networks.
On a related note, the funding issues surrounding networking on campus can also be seen in the procurement of individual systems and software. There are not distinct budgeting items for technology on this campus, which means that computing resources compete for funding with other supply and expense items and are frequently paid by monies earmarked for these items. There is also no carryover for budgets either, meaning that campus units frequently make purchases for their computing and software needs in a whirlwind of buying at the end of a fiscal year using collected funds, instead of distributing the costs of computing upgrades evenly over time. In other words, a server-class machine which may be purchased once every three years, can only be supported by the budget in one of those years, not evenly distributed.
One other infrastructure issue that needs to be mentioned is the University's access to the commodity Internet through the state Peachnet network. This access has been periodically unreliable, and the amount of bandwidth available has often been unacceptable. In fact, many people have, at times, experienced better bandwidth to the Internet through dialup connections than through Peachnet. It is our understanding that the Regents have created a Peachnet advisory structure, both administrative and technical, to guide the development and improvement of this network. We are cautiously optimistic that this mechanism for providing input to the Regents will help improve the network services offered through Peachnet.
Our concerns regarding the use of central data stores on campus (or lack thereof) are twofold. The first issue at hand is that there does not exist a single, central resource for data warehousing (a term used to define the storage, organization, and use of large quantities of operational data). This means that not only are all of the individual units needing to work with data stored outside of their units required to acquire and convert data obtained from different resources and stored in different formats, but that often it is not clear what the definitive source for operational data is. UGANET understands that the University is working with outside consultants to help clarify and resolve these issues, and we applaud this effort.
Secondly, there is the issue of lack of unity in application and resource development which stems from disparate data maintenance. Student information is spread over four or five separate databases across campus, everything from the UGACard system to the campus health services system, to the UGA online directory. This redundancy exists not as a precaution, but because there is no central data source in existence to maintain the individual data needed by these differing systems, while unifying the redundant information maintained by all. This sort of replication exists in almost all facets of the University's data operations.
For example, in order to use the University's budget development system, departments must set up a local database made up of budget and account information downloaded from University systems. The department may then work with their database and print out the changes they make to individual records. These changes are then sent up to their college level, where they are entered into a similar system at the College level, before being once again printed out and sent to the University level to be entered into the original database. There are then two, sometimes three different copies of the same data being worked on, and the same information is retyped multiple times. At the end of each budget year, each group using a local copy of the master data is therefore required to download the University's copy of the data again and reconcile the changes they have recorded against the ones which were made at the University level.
Similarly, the lack of central data stores and centralized development of applications to work with this data has led to different campus units developing different interfaces to their systems. Thus our software tools are comprised of radically different underlying technologies, interfaces, and capabilities from tool to tool. Certainly, centralizing all of our critical stores of data and the resources used to support and develop for them could help alleviate this problem.
Staffing issues may well be the most pressing concern for the University. No matter how modern our networks or how unified our datasets are, without the technical expertise needed to maintain these systems or train and support the users of them, their power remains untapped. It is regrettable but true that the University cannot compete directly with outside salaries, but we are unable to offer salaries that can keep information technology personnel on campus for any reasonable length of time. Albeit technical staffing issues and personnel shortages come to bear on commercial organizations as well, the University suffers from additional burdens that continue to weaken our technical human resources.
The first finger pointed usually directs all eyes to salaries as the key to our staffing issues. While this in fact is true and must be remedied, the problem is indeed far deeper than simply being an issue of salaries. In combination with a salary scale in the range of half to two-thirds of what commercial organizations can offer, the University also lacks unified support for training and certification as offered by their commercial counterparts. Such resources are critical to those in technical fields in an industry where continuing training is not an added benefit, but a necessity. Knowledge and training in the computing and networking fields is not useful for decades or often even years, but instead suffers a useful lifetime measured in terms of months. Such training is not necessary to expand the horizons of technical staff, but more frequently to keep them competent in the evolving responsibilities with which they are charged. A working group under UGANET is currently investigating certification and training issues, and we ask that their findings and suggestions be considered at such a time that their research is complete.
The University has had to compensate for the lower salaries it offers technical personnel by lowering its expectations in hiring. While this has allowed many technical personnel to be hired over the past several years, it also became apparent that the salaries of these untrained individuals would not increase over time as their expertise had. Such individuals found that careers begun here at the University would then stall should they remain on campus, and the tradition of UGA keeping technical staff just long enough for them to complete a period of basic training was born. It is no secret that the members of our technical community only stay on campus if they have their own personal reasons for staying. It is not their employment or the environment which keeps them in the service of the University, but more frequently family or educational obligations which keep them in Athens. This tradition does, indeed, need to end as soon as possible. With a shortage in the U.S of over 350,000 technical workers, it is most definitely an employee's job market at this time, and the University cannot afford to lose the already dwindling numbers of experienced and capable individuals remaining in its technical ranks.
In order to offset the low salaries offered for technical positions (which must be addressed), inflated job descriptions are more frequently used to set reasonable salary ranges for technical personnel, instead of accurately reflecting the needs and descriptions of the positions to which they are tied. Because of this, the classification system is more of an obstacle to hiring than a boon. In addition, salaries are tied to positions, not to individuals; that is, the same person, with the same level of expertise, can generally only improve his or her salary or negotiate funding for training by changing jobs. This has led to turnover rates in technical positions to be many times beyond those in more traditional positions. It is also a frequent occurrence that three positions may be changed into two positions, not because a third staff member is not needed, but because the funds available for hiring technical staff may be better used to pay two positions salaries high enough to retain technically competent employees, instead of offering three positions for which no competent applicants may be found due to the salary offerings. In addition, departments and divisions are, in ever-increasing numbers, attempting to rely upon part-time, short-term student workers to fulfill their technology support needs, with varying levels of success.
Much needs to be done to change the face of our technical support hiring methods, and these changes will indeed take time, but there are at least discrete steps that can be taken that improve the situation dramatically. The University has demonstrated its intent to work towards staffing solutions by its recent change of policy to allow for mid-year competing offers to be made to individuals with job offers from outside of the University system. It remains to be seen how effective this strategy is, but it is most certainly a positive step towards more flexible pay.
Much discussion was offered by the members of UGANET addressing the staffing issues on campus. In these were suggestions of increasing the numbers of technical staff, increasing base and relative salaries to at least match what the state of Georgia offers elsewhere for similar positions, "broad-banding" to allow for flexible job classifications to be associated with and salaries to be offered for positions, and establishing ongoing training and certification support for technical personnel within campus (either through the provision of funding for technical staff to obtain training and certification off-campus, or more preferably to offer such training on campus).
One of these suggestions that may need further explanation would be broad-banding. Although other definitions of broad-banding may exist, our definition involves the creation of a small number of job classifications with broad descriptions of duties. Each classification would have a broad salary range, and supervisors would have the flexibility to offer any salary with that range to new employees and would be able to offer flexible salary increases to incumbents in a position. Thus, less-experienced personnel may be hired into a position with a given classification, and as their skills develop and their responsibilities expand, their salary can be gradually increased without having to change the position description itself. This not only provides incentives for individuals to grow within a single position instead of changing positions solely for salary purposes, but prevents job descriptions from being distorted for the purposes of providing salaries high enough to obtain any applicants at all. This staffing method has been used with some significant success in organizations as large as the State University System of California.
There is frequently no hierarchy in place for technical positions, which means that there is usually only one knowledgeable person to support any particular department's services or systems. When such a person leaves, there is usually no one who can provide fill-in support as needed until other staff can be hired and trained. Instead, the users of such system are short-changed. Even in departments where there are multiple technical positions, the vast amount of work facing technical staff persons tends to rule out any redundancies in training and expertise.
Discussions of technical support improvements need to include existing staff shortages. These exist primarily because of both an explosive boom in the use of technology on campus and the lack of similar growth in technical positions. Indeed, in a time frame where the users of IT systems have multiplied tenfold (from 4,000 email accounts on campus to 40,000), the professional technical staff of the University has only grown marginally. With the nature of these jobs becoming more and more technical and demanding each month, the workloads of the individuals in full-time technical support positions on campus are growing exponentially, as do the workday demands upon them, while their numbers grow fractionally, if at all. In a sadly ironic twist, some individuals are leaving campus now, by their own assessment, to accept lower-stress positions in commercial organizations.
At a recent meeting with the Deans, one of the speakers noted his belief that there are no problems to be found in our situation, only opportunities. While it is true that our current situation, as detailed above, does offer many opportunities for growth and progress, these problems have become so ingrained into our campus culture that change may not come easily or readily.
The members of UGANET are thankful for the creation of a Chief Information Officer for campus, and do hope that this individual will help to focus efforts on the problems we face. We are also glad that the apparent goal of this newly created office is not to gloss over the problems which exist, but to dig down to their roots and reshape the way our campus functions in the technological arena. We also thank the objective attitudes of senior administrators towards the problems facing technology and the staff which support it on campus.
The primary goal of UGANET, as the official voice of technical staff on campus, is to act as a sounding board and resource for technical staff in order to help resolve technical matters. Collectively, we feel that by resolving these problems which face us all in the most expedient yet thorough manner possible, we can make the University the highly desirable place of employment for technical staff that it was in times past, and this will in turn provide the most benefit for our remaining technical staff. We are equally aware that the greatest danger lies in doing nothing, as we are fast approaching critical junctures for many of the issues presented here.
We are equally aware that having a concise description of a problem does not necessarily reveal its solution. We do embrace the University's current efforts, offer our thanks for their undertaking, and wholeheartedly wish to offer our support and expertise in this ongoing process of repair and rebuilding.