Skip to main content

The History of the Department of Computer Science

The department was 25 years old in 2009, making it a good time to record the department's history. If you can add or correct details, please contact Ian or Rob or Nigel. As well as this history, we are compiling a complete list of staff, with their dates:

Staff List


Pre-history, 1970 - 1983

In 1968/69, a Computer Science group was created as a sub-department of the Mathematics Department. The first professor of Computer Science was Mike Rogers, pictured in a more recent photo on the right, who had been associated with the Computer Centre. Three lecturers who were already in the Mathematics department became part of the group, and these were joined by Eric Lewis in 1970, and the group was formally recognised as a separate entity by the University in 1971. These early pioneers were:

In those days, the academic Computer Science group had difficulty differentiating itself in other people's minds from the Computer Centre, with which it had close ties, but which which was purely a service department. The Computer Centre was renamed as the Computing Service some time later, eventually becoming part of Information Services, while the name Computer Centre now refers to a building.

In 1975 Mike Rogers became Director of the Computer Centre. Some time after that and before 1983, he became head of the department of Mathematics, in addition to continuing as head of the Computer Science group.

During these years, Computer Science became a department in all but name. It was a subject area within the School of Mathematics. The University didn't formally recognise the school structure, so the School of Mathematics was officially a department, and Computer Science was officially just a subject area, alongside pure maths, applied maths, and statistics.

However, the subject area was recognised by the University, e.g. staff were appointed into Computer Science and not just into Mathematics as a whole. Computer Science had a great deal of autonomy, and built up its numbers. The list of lecturing staff before the department separated from Mathematics were:

John Pryce, however, stayed with Mathematics after the split. Some of the people also around at the time, as research assistants or technical support staff were:

Mike Wilkins later transferred to the Electrical Engineering department and then tragically died. Paul Buttner eventually moved to the Computing Service where he now heads up the PC Systems Support Team. Rob Thomas has continued with the department to this day, ensuring that the department retains its smooth and reliable hardware and operating system infrastructure. Andy Jones became a lecturer before leaving in 1987.

Equipment

Computing itself was very crude in those days, but we were very forward looking. For example, even though there was no Internet, no Ethernet, and only crude networking, we did set up email within the department and accepted the first electronic submission of course work by email in the early 1980s. We also had a Cambridge Ring network and a couple of colour graphics monitors with an astonishing resolution of 4096 by 4096, though we had no software to do anything impressive with them.

The department's original shared resource was a PDP-11/44 running Unix V7 and driving dumb terminals. Some went via the University's Terminal Switching Service (the PACX) and others were directly connected. The PDP had a 'massive' 1MB main memory and roughly 250 Mb Disk. The machine was the first in the department to acquire external networking in the form of X25 with Coloured Book protocols.

One so-called quirk of the Coloured Book protocols was that domain names were 'backwards', e.g. uk.ac.bristol.cs.www. Our department supported the battle to keep this ordering in the newly emerging Internet, but the battle was eventually lost to the American conventions. It is sobering to think that if the battle had been won, a URL today would be a single, logical, unified, flexible whole (e.g. uk/ac/bristol/cs/history/index.html) instead of being split in two parts going upwards and then downwards from the middle and mentioning a specific computer (e.g. www.cs.bristol.ac.uk/cs/history/index.html). The situation is strongly reminiscent of America's infamous illogical date convention which starts with the month, which is fortunately still being strongly resisted in the rest of the world.


The Creation of the Department, 1984 - 1985

In 1984 Computer Science split from the Mathematics department, becoming a separate department.

There were various reasons put forward for the split. In one meeting, Judy Holyer (an applied mathematics lecturer) and Ian Holyer (a computer science lecturer and Judy's husband) were sitting together, and Judy stood up and said "it is clear that mathematicians and computer scientists can't live together". But the most compelling reason was the feeling that as part of Mathematics we were treated financially as a pencil-and-paper subject, whereas if we were a separate department within Engineering, we might be treated as a laboratory-based subject and attract a bit more money.

At the time of the split, Mike Rogers was both head of the Computer Science group, and head of the whole School of Mathematics. He naturally became the head of the newly formed Department of Computer Science.

This left the rest of the Mathematics department in a quandary, because there was no senior member of staff who was both willing to become head and who was able to command enough overall support. There was a ground-swell of opinion that a junior member of staff should be appointed, against the usual convention, and it was regarded as a victory for democracy when lecturer David Evans was appointed head of Mathematics, and quickly promoted.

In 1985, Computer Science formally left the Science Faculty and joined the Engineering Faculty. The long-planned move also took place. The department pulled up its roots and moved from the Mathematics building into the Queen's building, taking up one of the wings on the ground floor. The space we took up in the Queen's Building had been vacated by Geology.

Judy and Ian Holyer were away on sabbatical in Australia when the actual move took place. When they got back, they discovered that they had lost their parking space, and would have to go onto the three-year waiting list to get it back again.

The department was still having trouble differentiating itself from the Computing Service, partly because of its still-famous COMS abbreviation. A request was sent to the University authorities to change the abbreviation to COSC, but the change was made only in one or two central places, and never filtered down into unit codes and so on. The COSC abbreviation is still confusingly used in some central University database tables today.

At exactly the same time as the University forming the Computer Science department, Hewlett-Packard established their European Research Labs in Bristol. Despite the HP Labs web site claiming the Labs were founded in 1983, the staff at Hewlett-Packard inform us that it was indeed 1984. Over the years there has been numerous interactions and projects between Hewlett-Packard Labs and our department. Indeed, there has also been an occasional exchange of staff. Two current Professors in the department (Dave Cliff and Nigel Smart) are ex-employees of HP Labs.

Equipment

At roughly this time the PDP11 was pensioned off and a SYSTIME 8750 (a clone of a DEC VAX 750) was acquired for Computer Science and Maths via the Computing Service. The machine ran 4.1 BSD Unix and was again networked with X25/Coloured Book. Computer Science also acquired its own AT&T 3B2 Super Micro multi user system with Unix System V (source licensed). This was never fully networked, being remotely accessed via a PAD (X25 terminal concentrator) running in 'reverse'.

Research

1985 also saw the first of the UK's exercises in assessing research, later called the Research Assessment Exercise. In those days it was called the University Grants Committee (UGC) Research Enquiry, and the submission from Computer Science was only three typed pages.

Even at this point in time the document talks about the natural fit of Computer Science in Engineering, pointing out the linkages between Computer Science and the Artificial Intelligence group in the Engineering Mathematics department and Microelectronics work in Electrical Engineering. It should be born in mind that Bristol is perhaps unique in having two Mathematics departments, one in the Science faculty and one in the Engineering Faculty, and Computer Science has always had strong links with both of these groups.

The major research areas of the department at this time are listed as:

Teaching

Teaching was quite mathematical in those days. To begin with, the only degree available was the BSc in "Computer Science with Mathematics" later to become the joint degree "Computer Science and Mathematics", which is still in operation today as "Mathematics and Computer Science".

The structure of the degree was strongly influenced by Science Faculty conventions. Students in Science would study their own subject and two others in their first year, then specialise after that. As a result, first year computer science students typically spent one third of their time studying computer science, one third studying maths, and one third studying some other science such as physics or chemistry.

The teaching gradually starting to gain its own distinctive computer science flavour. For example, here are some lecture notes on Complexity Theory from this time for a short MSc course in a fully mathematical style, and lecture notes on Analysis of Algorithms for a third year undergraduate course which had a slightly more direct and practical computer science style.


The Queen's Building Years, 1985 - 1994

Once upon a time, the department went a quarter of a million pounds into debt. Ian Holyer was (reluctantly and temporarily) looking after the finances at the time. Out of the blue, the University's finance department contacted us and claimed that we were very nearly 250,000 pounds in debt. While it was common for us to be in debt at the time, this was a huge shock, and we rushed around like mad things trying to find out what the problem was, hoping it was a mistake. We pored over the figures, and finally found the problem. The department had earlier asked for a short term loan from the University of 100,000 pounds for buying new equipment, and the time had come to pay it back. But someone had entered the repayment with the wrong sign, so that we appeared to owe 200,000 pounds instead of having cleared the loan.

It was in the early days, while we were feeling poor, that the idea of running MSc courses was born. A lot of the early work in setting them up was done by Derek Paddon, now at Bath. We did all the marketing, recruiting, admin and everything else ourselves, which put us in a good bargaining position for keeping the bulk of the money generated. As a result, the MSc courses have become one of the three current cornerstones of our reasonably healthy financial position. Our MSc courses are now the envy of other departments, but when asked by other departments how they can emulate us, our best advice is just "start twenty years ago".

Space in the Queen's building was getting short, and the department temporarily took over some extra space in the Chemistry building, which we called the Chemistry Annexe. There were labs in there, a small machine room, and a couple of unpopular staff offices. The offices overlooked a small building belonging to Chemistry called "the solvent store". Naturally, we enquired about its safety. It had no sensors or alarms of any kind, because putting electricity into the building would itself have caused a hazard. The building was rated by the fire service as the second highest risk in the city (after a similar store in the BRI). We did get one piece or reassurance however: "if it goes up, you won't know anything about it".

In the late eighties and early nineties there was an annual barbecue hosted by one of the professors in rotation. This social side of the department continued in the late nineties and early 21st century with the annual departmental treasure hunt (or pub crawl as others might call it) and the occasional departmental pantomime. To the right you can see one image from such a pantomine, other pictures are here. At some point Alan Chalmers decided that the department should settle on The Scotchman as its local pub, and the tradition of Friday night drinks in The Scotchman is still going strong.

Equipment

At the time of relocation to Queen's the Department acquired it's first Sun Micro-systems machines. There was a Sun 3/160 Server (the first Aloha) and around 8 of the first Sun 3/50 workstations to come off the production line. The 3/160 had a single 380Mb disk and each of the Diskless 3/50s provided the magic 3Ms (1 MIP, 1MB RAM, 1 MegaPixel monochrome display). These were all connected by an internal ethernet with TCP/IP protocols. The workstations formed a Computer lab affectionately known as the "Sun Lounge". This was doubly appropriate as, like all of our subsequent Sun workstations, they were named after Hawaiian locations in honour of the original ALOHAnet.

The Ethernet also carried Coloured Book traffic via 'Pink Book' (X25 over ethernet). We now had external email via Grey Book (Coloured Book email). The network of Suns expanded. Every office had an ethernet connection and we slowly moved to a workstation in every office. Eventually the ethernet became connected to the Internet. The number of workstations and servers evolved and expanded moving to Sparc Machines and X terminals (thin clients).

The early work on digital media, including our vision work, was supported by some early parallel computers. Barry Thomas and Erik Dagless (from EE) had a number of RAs and PhD students building the hardware to designs of Dave Milford. As time progressed more standard hardware was available, and a number of Silicon Graphics workstations were acquired with video post production software and the Alias Wavefront 3D modelling suite, later to become Maya.

In about 1990, a lab full of Macintosh computers was established for first year students. The lab continued to be used until about 1999 when it was replaced by Sun Ultra 5s and 10s, leading to a single set of student labs for all years. Around 1991 we acquired a number of HP 6800 workstations as a donation from HP. This became our first colour graphics lab.

In December 1993 the name www.cs.bris.ac.uk was registered and our first web site was set up by Neil Davies.

Research

In 1989 the RAE changed its name to the UGC Research Selectivity Exercise, and the submission now contained more detailed statistical information. Although the main descriptive document was still only three pages long.

There are seventeen staff, including support staff and research assistants, listed in the document:

Research was now focused in two areas: Logic Programming and Parallel Computation. Mention is made in the RAE of the start of the department's work in Computer Vision under Barry Thomas. This work in vision then led into our work on graphics and robotics in later years.

In 1992 the RAE was now called the RAE and the UGC had changed its name to the University Funding Council (UFC). There were now twenty two academic staff members listed for the entire return period. But those in post on the census date were:

It should be noted that in most RAE returns made by Bristol, staff in Engineering Mathematics who work in Artificial Intelligence are returned in the Computer Science submission.

The main descriptive document now mentioned three research areas: logic programming, machine intelligence and parallel computing. Logic programming was focused on work in declarative languages and meta-programming support. The work in machine intelligence mentions robotics (which was to become in future years a major research area with our work with the Bristol Robotics Lab), computer vision, speech and data fusion. The parallel computing group looked at working with transputers; a form of small micro-processor which allowed cheap multi-processor machines to be built. Transputers were made by the local company called Inmos, and were invented by a David May who was to become quite important in the department in the next period.

Mention is first made of the Advanced Computing Research Centre (ACRC), which was then a centre linking the computing work of the departments of Computer Science, Engineering Mathematics and Electrical Engineering. The ACRC was itself born out of the Information Technology Research Centre. In more modern times the ACRC has been reborn to comprise the entire University's work in High Performance Computing, see http://www.acrc.bris.ac.uk/ for details of the current ACRC.

Teaching

In 1984-5, the department's one degree was given a complete overhaul, and a straight Computer Science degree was added, as well as the Computer Systems Engineering degree (CSE). The introduction of the CSE degree led to a number of new staff being appointed.

There were no units or credits in those days. Instead, everything was measured in "standard courses" (a standard course being 40 credits in today's terminology). Each standard course lasted all year, and was typically split into part 1 and part 2 (i.e. one 20 credit unit in each teaching block, as we would say today).

In the first year of the new structure, a student would take the standard course CS1, a mathematics standard (Maths 1M or 1A or 1C), and another Science standard (typically physics, chemistry, economics and accounting, psychology or philosophy). For students taking other subjects, the subsidiary course CM1 was provided.

In the second year, a student would take standard courses CS2A, CS2B, and CS2C. The two parts of each of these produced 6 marks, but the worst two (often the two parts of CS2C, which was regarded as too mathematical) were averaged into a single mark (i.e. half-weighted) to form the year mark.

The third year of the degree contained options, with an option being worth a fuzzy amount, somewhere around 15 of today's credits. A typical third year, according to the documentation from that time, would consist of "7 to 8 options, or 1 standard and 5 to 6 options, or 2 standards and 3 options".

It was compulsory to do an individual project, which could count as either one or two options-worth, depending on size. The other options around that time were VLSI, Compiler Construction, Databases, Software Engineering, Analysis of Algorithms, Advanced Architectures, Programming Methodology, Computer Graphics. Also available were Theory of Computation from the maths department (given by Prof John Shepherdson, a logician with a strong interest in theoretical computer science) and Artificial Intelligence from the Engineering Mathematics department.

Each department's examination results need to be approved by an appropriate faculty committee. The Computer Science department's examination results were first approved by the Engineering Faculty in 1993, hence even up to 1992 we still (for examination purposes) were part of the Science Faculty.


New Directions and Growth of the Department, 1995 - 1996

With Mike Rogers about to retire, the search started for a new head of department. There were disagreements about the suitability of some candidates, making the situation difficult. In the end, David May was "head hunted" as someone suitable with both academic and industrial experience. He became the new head in September 1995. It was to be a few years later, in 2003, that we were saddened by the news of the death of Mike Rogers, who had effectively been the founding father of the department.

David May set to work to build up the department, particularly in terms of numbers of staff and research activity. He also made us re-evaluate and refresh everything that we did, including an almost complete re-design of our undergraduate courses.

Around this time, a plan was devised for a new building to help house Computer Science and the department of Electrical and Electronic Engineering. Half was designed to be let out commercially for a few years, to make the building self-financing.

We also began our current style of department handbook, and a complete set of copies of this helps in placing activities and dates from this point on. Staff who arrived just before or just after the move to the new building include:

Equipment

The most important event related to the department's computing infrastructure in 1995 was the registration of the name mail.cs.bris.ac.uk in the DNS. This marked the transition from Grey Book email to Internet (SMTP) mail and the introduction of IMAP.


The Early MVB Years, 1996-2004

The department moved into the new Merchant Venturers Building (MVB) in 1996. The commercial sound of the name contrasts with the academic sounding name "University Gate" used for the commercial half of the building. There is still some doubt to this day whether this was a clever way to mark the integration of the University with the city, or whether it was the result of an administrative mistake over which half of the building was which.

Part of the new Merchant Venturers Building is made up of the facade of an old cinema which existed on the corner of Woodland Road and Park Row. Opposite the cinema was the Prince's Theatre. This building is famous due to it being the home of the dog which was used as the model in the famous "His Master's Voice" painting, which formed the logo of the HMV company. The story of the dog, called Nipper, can be found here. Nipper's owner Mark Barraud was a scenic designer at the Prince's.

The cinema was built in 1910 and over the years the cinema building was used as roller skating rink, an exhibition hall, and during world war one it was an aircraft factory. In November 1940 the theatre and the cinema were destroyed during the Bristol Blitz, but a view of the cinema from the early 20th century can be found on the right. The front of the old Coliseum cinema, which forms part of the facade to the MVB building, now has a blue plaque and stature to honour Nipper.

After the Blitz the site of the building was used as the Vets School for the Univeristy, on the Park Row side, whilst the Woodland Road side still had a few houses on it. Eventually these houses were bought by the University and so the whole site could be used for development. The result was the creation of the MVB/University Gate building.

MVB is a great building, but a few corners were cut and there were teething problems. There were inevitably some problems with leaks in the roof, as you as you might expect from the unusual shape. The wrong locks were installed on the office doors, so they couldn't be opened from the inside, and people could too easily get trapped inside. The railings in the stairwell were inadequate, so that young children could not be allowed in the building. Only one lift was installed on the academic side, and there is a still a spare empty lift shaft.

Neil Davies checked the architects' plans very carefully, and strongly recommended suspending a one-piece plastic sheet under the ceiling of the machine room to protect the equipment from water leaks, with a very clear prediction that otherwise there would be disaster, especially as there were toilets directly above the machine room. It was decided to leave it out, on the basis that there was two feet of concrete between the machine room and the floor above that would do the job. Of course, before long, holes were drilled through the concrete for various reasons, and soon enough there was a water leak on the floor above and the machine room was flooded. The refrain was one of "We told you so", and the toilets above the machine room were decommissioned.

Ian Holyer, Henk Muller and Neil Davies visited the building just before completion in hard hats, took one look at the huge room with two balconies and panoramic views, and decided to put in a bid to share it as an open plan office. They were worried about noise, but got used to it instantly, while other members of staff were complaining about the thin walls of the single offices.

The department grew significantly in these years, both in terms of staff, students, and in its areas of interest. We expanded into Quantum Computing with the appointment of Richard Jozsa, Cryptography with Nigel Smart, the work on Machine Learning was strengthened with the appointment of Peter Flach, and the Vision Group's work expanded with the appointment of Majid Mirmehdi.

However, the close of 2004 saw the premature death of Barry Thomas. Barry was always considered the life and soul of the department, both by staff and students, and his death cast a long shadow over the following years.

Equipment

With the renewal of the department and the move to MVB we expanded and began to attract outside interest. To mark the move into MVB, Sun provided an E4000 server, our fist Enterprise Class machine. This was unique in that although designed as a server it had three 3D graphics cards. It was used as a graphics engine driving three ceiling mounted projectors to provide immersive panoramic environments. Other notable large machines acquired on research initiatives included an HP N Class server with 1Tb of disk and an SGI Onyx which took over from the E4000, which was reallocated as a powerful multi user machine. In August 2001 the Department installed it's first Linux Systems. These replaced the SGIs as a platform for Maya. Linux usage subsequently increased, until it became the main stay of our desktop environment.

We also saw the delivery of our Motion Pod simulator. This is still in use in our undergraduate third year projects. There is a suspicion that students try and make the markers as sick as possible when being examined by the staff. Neill Campbell swears he enjoys the day of marking these projects, but always looks a bit green by the end.

Research

There were two RAE's in this period, the first in 1996 and the second in 2001. The 1996 submission was hurriedly created by David May on first entering the department. He had only been in post for a few weeks, when he was told he needed to write a submission. Quickly David reorganised the research groups around three themes of Computer Architecture, Declarative Systems and Multimedia, along with a fourth group on Artificial Intelligence (representing the staff from Engineering Mathematics).

There were a number of cross disciplinary research centres. As well as the ACRC, there was PACT a centre set up with SGS-Thomson Microelectronics (formerly Inmos and now known as ST-Microelectronics), in addition there was the Safety Systems centre. The Safety Systems centre is still in operation, but it is now located in the Civil Engineering department were it conducts work on systems safety in the nuclear industry amongst other things.

A major difference in the 1996 submission was the strong linkage that there was now between the department and the local industry. Close ties had been forged with HP Laboratories in Bristol, with the local micro-electronics industry (such as ST Microelectronics), with the local media industry (Bristol has a major concentration of media technologies), and other local industries such as BAe, Rolls Royce and the local health care providers.

The 2001 submission showed our increased research portfolio. Work was divided into seven research areas: Quantum Computing, Cryptography, Machine Learning, System Design and Verification, Digital Media, Languages and Architecture and Mobile Computing. All of these research areas still form a major part of the departments research efforts.

The setting up of the Quantum Computing group demonstrated the pivotal role the department plays in the linkage between the Science and Engineering faculties, since we have been in both faculties. Bristol's quantum work now spans the departments of Mathematics, Physics, Electrical Engineering as well as Computer Science.

Another example of our focus on the boundaries of Computer Science was the setting up of the Centre for IT and Law, which was established June 2002 as a pioneering cross-disciplinary venture, building on existing strengths in the School of Law and Department of Computer Science. Initial sponsors of the Centre were Vodafone Group Services Limited, Barclaycard, Herbert Smith, Hewlett Packard Laboratories and the Law Society Charitable Trust. The Centre's main areas of research were and remain in the fields of privacy, digital rights management, cybercrime and e-commerce.

Teaching

During this period, the teaching programme was overhauled, and our philosophy of teaching was changed. Rather than teaching a wide range of subjects where 90% of the work was assessed by exam, we designed a top down course that focussed on a few subjects, where up to 50% of the assessment was for practical achievements. This meant we could easily deal with larger groups of students, the quality of teaching increased because staff had more time to teach, and we attracted not just a few good students, but many good students year on year. The change in teaching philosophy also heralded a steady rise of the department in the THES league tables.

Major changes that happened to the programme during this time were the introduction of a substantial group project on the 4-year course; asking all students to come up with their own project ideas; and the introduction of business plan units. The latter has helped many students in setting up their own companies. In order to streamline teaching and the operation of the department, we also created one simple set of rules that covered progression, marks, and expectations, collected in the "Department Handbook" that was first published in 1995.

The admissions procedure was also changed drastically, and we were amongst the first to abolish formal interviews on admissions days, as they were considered an unfair and unreliable instrument for admissions. They were replaced with a procedure where we enabled applicants to make an informed choice.

Automation

In about 1996, the department began its long and successful history of automation, culminating in a web based administration system which is now the envy of many departments in the University. We are able to monitor student progress, record important facts and analyse data in ways which make the smooth running of the department an achievable goal, as opposed to simply a wish.

The static web site, which had been created in 1993, started to be used in more dynamic ways, using programs written by Henk Muller and others.

The department became one of the first departments to maintain an on-line database of downloadable research publications (initially in compressed postscript, later upgraded to PDF). After a pilot run with publications from three authors it didn't take long for the rest of the department to catch on and realise that downloadable publications would be the way of the future.

Around this time, Ian Holyer set up the first version of the department database to handle teaching admin. By 1998, the department was able to show off web pages containing lists of students and staff with photos (some coming from the newly established University image bank, and some from an engineering faculty image bank). There were automatically generated email lists, including one for each taught unit, and student progress pages showing coursework marks.

The database was a personal one, there was a program written in C and run each night to take data from the database and insert it into web pages, using templates which used our own custom tags. There was also a custom Java program which staff could use to enter coursework marks, which were dumped into files and incorporated into web pages by the overnight process.

During 1998, Ian and Henk put forward a vision for the future of the department's automation. They advocated moving from a personal database to a shared (Oracle) database which everyone could access. They suggested fully automated web page generation using a specialist data insertion language. (They were thinking of implementing this themselves, but it turned out later that Tomcat and JSP would do the job.) They suggested automatically taking data from central University sources in electronic format. They painted a picture where all members of the department would be able to enter and edit data via web interfaces, from work or home, regardless of platform. Finally and most importantly, they proposed employing a full time information officer to look after the whole enterprise, to take the growing load off the shoulders of academic staff.

The department set out along this path, with the first information officer, Georgina Crawshaw, starting in 2001. The second information officer, Lucy Smith, took over in 2004, and our current information officer, Martin Baker, took over in 2005.


The Later MVB Years, 2005-2009

The last few years have shown a steady expansion of the department. Our work on Mobile and Wearable Computing has branched out into areas of robotics, cheap devices and issues related to HCI. Our cryptography group has expanded, as has our work in more theoretical aspects of the subject.

Since 1996 a major plank of the departmental ethos has been the entrepreneurial activity. We have embedded this in the curriculum, and a number of companies have resulted from undergraduate projects in the department. The most successful of these has been the spin-out in 2006 of Xmos.

In 2006 David May stood down as head of department, so he could help in the setting up of Xmos. Nigel Smart pulled the short straw and was given the role of taking on the department, which had now grown both in staff and standing to be one of the premier teaching and research institutes in Computer Science.

Also in 2006 the MVB building had a major refurbishment. The disused toilets (i.e. the ones above the machine room) were finally converted into office space to house our increasing number of RAs and PhD students. In addition the atrium area was made into a new meeting area, and social space, for use by students, staff and visitors.

Social traditions are also still going strong. There is a long standing tradition for people (staff and PhD students) to bring in donuts and/or other similar goodies when they have something to celebrate, such as birthdays, weddings, joining/leaving the department, etc etc. Traditionally a nice selection of donuts from the jam, plain ring, iced and chocolate families have been selected. Other variations provided in the past have been cookies, flapjacks, muffins etc. The time of year can influence the choice also, choc-ices in Summer, mince pies around Christmas, Easter biscuits near Easter,... A special mention must be made of home made items; brownies, cake, flapjack, cheesecake have all been homebaked before now and have been received extremely well.

Equipment

Today, the department has a computing infrastructure based around Sun servers for file, mail and web services. There are also a mix of Sun and Linux interactive machines. Our teaching labs are Linux based and staff/research desktops are primarily Linux or Windows based PCs. Storage is provided by 4 main RAID arrays offering over 10Tb of file store. We also have access to one of the most powerful computers in the country, BlueCrystal.

Research

The 2008 RAE contained the largest return for Computer Science from Bristol. The department now groups its research into the following themes, each one consisting of a set of more specific topics:

It is interesting to look back on the growth of these areas as the department has progressed over the last 25 years.

Teaching

Much of the structure from the early MVB period in teaching has carried forward to the present day. We now run a successful set of MSc programmes. The newest one being a joint programme with Electrical Engineering in Microelectronics. This programme has been created due to the demand we were experiencing from local industry for skilled workers in this area.

Over time the curriculum has become more mathematical, however the core of providing students with practical real world skills and various transferable skills continues. The focus on entrepreneurship and on making sure the subject, and our graduates, make a real contribution to society is still at the core of our philosophy.


The Next 25 Years

Its been a long journey and who knows where it will lead? What we do know is that the department in twenty five years time will look as different to us, as the current department would look to someone in 1984. That is simply a function of the pace of technological change.

In the immediate future we are looking to expand our linkages with the local micro-electronics industry and grow our work on devices (both building and their applications). Our theoretical foundations are likely to be expanded, as so-called theory is playing a greater and greater role in applications (such as cryptography and data-mining). Finally, as evidenced by the growth of Google, we see as vital to continue our work on efficient data mining and learning algorithms; especially when applied to the vast data sets now made available on the Internet.


Compiled by Ian Holyer, Henk Muller, Nigel Smart and Rob Thomas, with contributions and memories from others including Eric Lewis and John Gallagher.