1. General Overview
To provide basic IT infrastructure and other IT services to the whole University and College community in a cost-effective and efficient manner. OUCS focuses on those IT services that can only be or are best provided by a single provider within the University.
It believes that this has been accomplished in a most satisfactory fashion. It has maintained existing infrastructure at a very high level of availability and performance, it has responded to ever growing demand by judicious expansion and growth of facilities and services, without incurring substantial additional expenses (indeed, while keeping costs static); it has introduced new services as the need has arisen or it has been requested to do so; it has ensured that fee-for-service offerings do in fact cover their costs (overall); and it has closed down, redeployed or transformed services where need has changed. It has done this through careful attention to technological change, to changing demand, and to the needs of the University community as expressed through the large range of consultative and advisory groups which guide it.
Overall turnover for the year was £6.25 million, to which the basic General Board grant contributed £3.16 million. The balance was made up of income from the sale of services (nearly 1.5 million, with an overall surplus of just £5,000), external grants (£0.5 million), and special General Board grants for special purposes (eg the Microsoft Campus Agreement), totalling just over £1 million. Overall, OUCS had a small deficit for the year, which was covered from reserves. This was only kept modest by careful pruning of the equipment spending, and cost cutting in other areas wherever possible. Great care was taken not to defer equipment upgrades which would only create congestion or poor performance, resulting ultimately in false economies (as well as irate customers!).
The year to come, 2000/2001 will be quite difficult, as the general cuts in General Board grants will be hard to absorb, and it will almost certainly not be possible further to delay some equipment expenditures already deferred from 1999/2000.
Much discussion took place during the year, principally in the IT Committee and its General Purposes Committee, as to the funding arrangements to be put into place under Divisionalisation, from 2001/2002. In many ways, OUCS is well placed, since its finances have been scrutinised carefully several times over the past few years. In particular, services have been categorised as `core' (ie infrastructure, vital to the whole University, and so a necessary top-slice on Divisions' funds), secondary (some income is derived, but not fully self-financing — these services have now been reduced to just two), discretionary (eg trading, which are expected to be fully self-financing), and overheads (administration and other unavoidable costs, so again a necessary `tax' on each Division).
OUCS is also proud of the fact that the main general grant it receives, as a fraction of the overall turnover of the University, is the smallest of all the main UK research universities. This indicates that OUCS is the most efficient central IT unit of all these 19 universities. The University and the Divisions can take comfort in this fact, when agreeing to the OUCS grants for 2001/2002 and beyond.
As indicated above, there continues to be much growth in IT activity employing OUCS services and facilities. In some cases (eg Janet traffic), this continues at the staggering pace of doubling every year. All systems have satisfactorily withstood this assault, and there has even been a small improvement in availability and reliability of many services. This is partly a testimony to the improving performance of IT equipment, but is also attributable to the sterling efforts of many OUCS staff who keep them running in the face of faults or other problems (eg those caused by expanding workloads!), and often also to the judicious choice of equipment and software that gives long-term greater return on investment than merely the cheapest in capital terms.
The above growth in activity is unfortunately not restricted to legitimate academic activity within the University. The high profile of Oxford University, coupled with its well-developed networks, ensures that it is often the target of hacker attacks. In conjunction with many systems staff across the University and Colleges, OUCS has built up a wide repertoire of defences from such things as `denial of service' attacks, where computers outside Oxford (and usually, outside the UK) launch hordes of network queries on particular Oxford systems, which quickly overload the system's ability to cope, effectively denying it the ability to service genuine requests.
OxCERT (Oxford's Computer Emergency Response Team) is the lynchpin of these defences. These staff offer concerted responses to actual or suspected attacks, often working at the most unsociable hours, and have developed considerable skill in detecting and dealing with incidents. They also collaborate fully with similar teams elsewhere in the UK and internationally, as well (as needed) with police authorities. Their skill and diligence have kept Oxford free from the worst damages.
What is of great concern in such a diverse and distributed university as this (geographically and administratively) is the frequency with which computer systems are poorly configured or poorly administered when they are introduced to the network. It is not really possible to monitor or check for this, but the rest of the University's network (let alone the information assets of the systems concerned) is jeopardised as a consequence. The Firewall and some policies introduced with the backing of the IT Committee will help, but significant exposure remains. Those who employ such OUCS services as the Network Systems Management Service (NSMS) can rest assured that their systems are configured and maintained to the highest standards; but it is regrettable that shortages of IT staff resources, or shortcomings in training programmes, may be exposing the University unnecessarily. OUCS will continue to address this by: raising `public' awareness of the risks; promoting use of its professional services (eg NSMS); advocating and running appropriate training courses for IT staff; and bringing its concerns to the political arena, to ensure that the University as a whole is aware of the dangers.
This has been the first year of the scheme, introduced following the 1996/7 Review of OUCS, whereby those departments in receipt of RAWP funding could opt not to use the OUCS Help Desk and receive a portion of OUCS' grant. Several departments opted out, and OUCS funds were reduced by over £90,000 in total.
The scheme has had mixed success. Those departments which opted out have received their funds, and made their own arrangements for dealing with the increase in user queries. Several have indicated they find the arrangement satisfactory. Inevitably, the resource has rarely been sufficient to deal with all the queries, but that is as much due to the general increase in use of IT, the increase in complexity of IT systems, and perhaps the increasing adventurousness of users, as to the passing of workload from OUCS.
From the OUCS perspective, the scheme has not really been of any benefit. Although its funds have been reduced, it has not been able to reduce its costs (partly because so many of them are fixed, and partly because the costs the Review Committee used in the first place were quite crude estimates), and indeed OUCS incurred additional expense in administering the scheme. Morale has also been affected, among OUCS staff and among disgruntled users who have to be turned away. The only success seems to be the ease with which the undergraduate enquiry load has been able to be absorbed by the Colleges. There has been a decline in use of the Help Desk, but that only continues a trend already established — and this obscures the fact that the degree of difficulty of the queries that are brought has increased substantially.
A fuller discussion of the factors and issues involved in this scheme can be found in the body of this Report, but it does seem clear that a radical overhaul or abandonment of the scheme will be necessary. Indeed, the new funding arrangements under Divisionalisation will require such a change.
OUCS continues to be committed to the provision of IT training widely within the University. Ultimately, this is the only proper way to address the difficulties with IT systems which users invariably encounter (and which are addressed by the above Help Desk arrangements, and their equivalent within departments and colleges). Unlike the Help Desk, training addresses these problems at their root.
In addition, this training provides the means for students to acquire transferable skills, now a national (Dearing) as well as local (North Commission) priority. And it helps fulfil the University's responsibility as an employer towards appropriate training for its employees.
It is to be regretted that some courses had to be curtailed somewhat during this year, due to staff shortages and redirection. This was partly due to the urgent need to establish a training regime for IT staff throughout the University, which has proved very popular and highly effective. It is to be hoped that the University will soon agree properly to fund this activity, rather than have it impact on more general IT training (but equally important in the long run).
Overall, all OUCS staff contribute in a myriad of vital ways to the mission of the University of Oxford, and tribute must be paid to their consistently valuable efforts. OUCS is also grateful for the high degree of cooperation which others within the University and Colleges who also provide IT services accord us. We also do believe that we must and do operate in partnership with academic and other colleagues, rather than `merely' as servants (though we do also believe in a service-oriented approach to our work).
Some members of staff have been recognised for their achievements outside Oxford, which underscores the quality and value of the staff OUCS employs. An example of this is one member of staff who was the recipient of a `Special Recognition Award' from the TeX Users' Group; he was awarded a specially commissioned, nicely framed cartoon by the Californian artist Duane Bibby displaying his TeX-related achievements over the last 15 years.
One staff member was seconded to the Registrar's Office during much of 1999 to coordinate University-wide efforts to minimise the impact of Year 2000 problems. Of course, OUCS itself undertook careful analysis of its systems to ensure there would be continuous operation over the year-end, prepared and published advice and instructions on Y2K issues generally (especially regarding personal computers), and made arrangements for staff to be on stand-by over the year-end itself in case of unforseen difficulties. In the event, the year change proceeded almost without incident, no doubt in part due to the efforts of everyone concerned to deal with potential problems. In retrospect, the potential impact seems to have been over-hyped, but anything other than a fully professional approach before the event would have been irresponsible.
There continues to be a significant turnover in the OUCS staff complement. To some extent, this is only to be expected in the IT industry, and OUCS rates are not noticeably higher than the average, though salaries generally are quite a bit lower than industry norms. This loyalty must be due to the fact that the University environment provides an interesting and challenging place to work, and also to OUCS' commitment to training, career development, recognition of merit (limited though that has to be), and general staff support.
It must, however, again be said that OUCS is in some areas critically dependent on the skills and goodwill of a few key individuals to maintain some of its most important core services. Although OUCS undertakes risk assessment exercises from time to time, it remains true that the risk of not being able to back up or replace such staff remains one of the greatest risks.
Overall, there continues to be a small growth in staff numbers from year to year. This is caused by the growth in externally funded or self-financing systems, or in a few cases by OUCS being asked by the University to `host' some staff (eg those shared between several humanities faculties). This continues to put pressure on the accommodation at 7-19 Banbury Road We still have some basements that are capable of being made habitable, and there is room for expansion on the roof of the computer room, but such increases in accommodation will require much advance planning and will, we assume, be costly (although they do add to the fabric of the University). So consideration must be given now to how this need is to be addressed.
We said farewell to Sue Brooks as Assistant Director (Distributed Computing) at the end of the year (she moved to Sussex University as Director of Electronic Services); however, we were able to welcome John Jenkins (formerly IT Manager at the Said Business School) as her replacement not long afterwards.
Although it fell outside the reporting period for this Report, it should be noted that the Director, Alex Reid, announced his resignation, effective 24-Nov-00, in order to take up a post at the University of Western Australia (where, in Perth, he will be able to rejoin the bulk of his family). It is to be hoped that the University is able to recruit a suitable replacement before long, as technological, structural and financial developments will not stand still!
Alex has many regrets in leaving Oxford, having really enjoyed his seven years here. He is very grateful to all who have assisted, supported and encouraged him during this seven years, especially all the staff of OUCS. He wishes OUCS and Oxford University every success in the future.