Talk:Institute of Computer Science

This Talk page created by George Coulouris to encourage discussion aimed at clarifying and expanding the proposed Institute of Computer Science page. All are welcome to contribute notes and thoughts below.

Question: How much depth of narrative do we want in this article? I'm sure we all have a few affectionate recollections that we could contribute if that seems appropriate.

Some random notes from George Coulouris:

Date of establishment: is 1960 correct? Mike Bernal would know.

MSc. course: I believe it was first taught in 1962 (the year I joined the institute), can anyone confirm? Can we track down any syllabus or teaching material. I taught a course on Theory of Computation in 1962 & 1963 based on Martin Davis' book.

My own tenure: I was a Research Assistant at the Institute only from 1962-65, then leaving to join the Department of Computing and Control at Imperial College and then moving to QMC in 1971. So it isn't correct to say that I moved to QMC on the closure of the institute in 1974.

People: I have added Benedict Nixon and Alan Fairbourn

There should probably be a section for non-academic staff such as Joan Spencer, Jan Vincent, ...

Some notes from Jean Dollimore

People I will add David Barton and Mike Clarke (statisticians) to the real page.

Comments Should note that the University of London Computer Service was in the basement - at first there was a Ferranti Mercury and then an Atlas. Then there are the operators, Bernadette, Diane and Margaret(?)

My own tenure: I was a Research Assistant from 1963 until 1974, —Preceding unsigned comment added by Jeandollimore (talk • contribs) 18:06, 3 July 2010 (UTC)[reply]

I wrote the initial draft of the Wikipedia article having had input from a few ex-colleagues from when I was at ULCC. This was a LinkedIn discussion under the "ULCC Alumni" group. I have a raw copy of this discussion if required but here is my (slightly edited) version that discussion. If you get bored then skip to the end tag corresponding to the one above! -

LinkedIn Discussion: University of London Institute of Computer Science

JID: JID=John Davies, freelance / CTO, Movirtu

JID: Over the years since I left ULCC (and even when I was there, and before) I have occasionally heard of the (now long defunct) University of London Institute of Computer Science. I recall someone telling me that it was founded with the grand intention of being an "LSE for Computing" and was killed off by withdrawal of funding. A lot of bright people seem to have been involved (e.g. two who I have known slightly - Sidney Michaelson, George Coulouris). But I can find no account of the Institute anywhere on the web. If there are any ex-members of the Institute known to this group then lets encourage them to write something. But tell me if I'm mistaken and someone has written all this down already!

PF: PF=Peter Furniss, Technical Director at HealthyConsulting Ltd

PF: I believe it was absorbed into Queen Mary College. When I was hacking into the CDC 6600 from Silwood Park, Pute Jackson and others (including Bob Pockney, IIRC) were at ICS, investigating the capabilities of the ULCC machines as well.

JS: JS=John Seymour, Head of Network Operations for JANET(UK)

JS: When I joined the London Network Team (LNT) early in 1981 I remember putting them on a list of all London sites I was asked to compile only to be told they didn't exist. I believe I got the information from a UoL year book (I guess 1980). The list was to plot who had RJE station and plans for moving to X.25 and Red Book, and who needed Camtec PADs). I also remember George Coulouris but, sorry George, I can't remeber from where.

MK: MK=Mick Kahn, Head of Applications Services at University of London Computer Centre

MK: Have a look at http://www.dcs.bbk.ac.uk/50years/50yearsofcomputing.pdf , page 13 which give some information. Also mentioned in http://www.cs.ucl.ac.uk/history.htm and http://www.cs.ucl.ac.uk/staff/j.washbrook . If you wantthe full story there are 25 boxes of archives in the Senate House Library - see http://www.aim25.ac.uk/cgi-bin/vcdf/detail?coll_id=7740&inst_id=148&nv1=browse&nv2=repos

JID: I'm beginning to compile information about ICS including all your responses and Mick's links. George Coulouris was Prof at QMC and later at Cambridge (he's still around but emeritus). He offered me a place on his M.Sc. course in 72 but I turned it down in favour of Manchester (I'm still not sure if I made the right decision!) I have looked online for the UoL yearbook but haven't found it. Any ideas? Does anyone know a way to contact Pute Jackson or Bob Pockney? Or anyone else who was at ICS? Once I have put something significant together I will ship it to all concerned for amusement and criticism If nobody else does I'll make a Wikipedia entry for ICS and put the stuff into it.

MK: Bob Pockney retired from ULCC earlier this year, but I'll try and put him in touch with you

JID: There is now a Wikipedia entry http://en.wikipedia.org/wiki/Institute_of_Computer_Science Its fairly basic and leaves much to be desired in both form and content. Please feel free to hack as you think fit (as always with Wikipedia!) Thanks to Mick, John and Peter for input. All the errors (so far!) are, of course, my own. If you flag this to anyone outside the ULCC Alumni group then I'd appreciate a cc so I can keep track - if only vaguely

AH: AH=Andrew Hinchley, CPL Healthcare/own consultancyLead for HL7 UK University Outreach Programme

AH: Happy to add some detail here as an ex-ICS member. I never saw an initial brief for ICS but certainly believe it was formed in the expectation that computer science would be a specialist post-graduate activity and that ICS would shine as such. By the time it shut computer science departments were opening up making ICS increasingly out of line with developments generally. At the end the staff were split between QMC and UCL. George Coulouris was on of the initial professors at QMC while Peter Kirstein the same at UCL. Peter is still an emeritus professor at UCL. In my year one of the other students was Steve Wilbur who became a professor at UCL. At the end of my year a spin-off software company was formed under David Hendry, a lecturer at ICS. A number of students including myself joined. The company focussed on language compilers/translators for Fortran, Cobol etc which was a highly developing market at the time. The company unfortunately did not last. A year or two later I got a call from Peter Kirstein to join his research group at UCL as an associate research fellow - another link back to my time at ICS!

GC: see This page from Computing at Chilton: 1961-2003 for more on 3 of the langages developed wholly or partly at ULICS - ACL, MVC and CPL. —Preceding unsigned comment added by Georgio8 (talk • contribs) 11:22, 4 July 2010 (UTC)[reply]

JID: Thanks Andrew. I had an idea you had done the course! Feel free to add to the Wikipedia entry. Everybody else - someone with archive research skills needs to go through those boxes at Senate House Library. Any volunteers from the NDAD team? I'd be happy to help interpret using my knowledge of computing long ago!

AH: Having now looked at the wikipedia entry for ICS, it has helped me to be able to fill in some more detail which maye be useful in tracking down more information on the origins of ICS. When ICS closed, the staff actually went to a a larger number of institutions than I previously stated,here is a combination of what I know/what I have gleaned from wikipedia:-

Birkbeck: Professor/Director of ICS: Dick Buckingham (now deceased)

QMC: George Coulouris (now emeritus professor and Mike Bernal (not on the wikipedia staff list)

UCL: Peter Kirstein (now emeritus professor), John Washbrook (now emeritus professor), Peter Higginson

Imperial College: Dave Howarth (later Professor)

Dick Housden - went to UEA as Professor (not on the wikipedia staff list)

Dave Howarth provided the major link with computing at Manchester . he designed the operating system for Atlas, which of course was the major computer resource co-located with ICS in Gordon Square, certainly at the time when I was a student there.

JID: Sounds like an "ICS Alumni" group might be a good idea. Much better qualified than me to update the Wikipedia entry! There are probably some Manchester connections too but Atlas was long gone when I was there (I wrote a macro generator for MU5 Autocode)

AH: ICS info has been pouring in from John Washbrook, an ICS staff member now emeritus professor at UCL and from Paul Samet who was Director of Computing at UCL and had an associate position at ICS. As John has mentioned I don't suppose this list wants to have all that info - I will certainly plan on updating the wiki entry myself or in collaboration. Just to mention one or two points.. Co-location in Gordon Square with the Atlas service for London/elsewhere also brought some of the key brains on Atlas design - Professor Dave Howarth and Alan Fairbourn. They designed a virtual management system years before IBM claimed this. The development of C was preceded by BCPL, A Cambridge development but I believe with ICS involvement. London Internet collaboration and research with Stanford must I think have started at ICS. This resulted in Peter Kirstein and Peter Higginson taking this project to UCL, eventually resulting in Peter K, Peter H and myself being cited by Stanford as Founders of The Internet along with researchers at the key other establishments collaborating with Stanford.

JID: Sounds like Andrew has, in effect, an ICS Alumni group so I'll step back from the Wikipedia entry now. Thanks Andrew. BCPL (you may recall) was our implementation language in ULCC Comms group (Edinburgh IMP would have been better, IMHO, but Tony Peatfield was the boss!) If further spadework is required, such as the Senate House Library archive, then I'd be happy to help (but we could still use an archive expert from the NDAD team at ULCC!)

AH:There is then one specifically relevant point on the development of BCPL. On the wiki entry for BCPL its development is credited to University of Cambridge and ULCC, however I wonder if in fact there was more of a split between the development of the language involving ICS staff such as Eric Nixon plus then an implementation/roll-out by ULCC after ICS was closed down?

GC = George Coulouris (userID georgio8)

GC:There was very little relation between BCPL and ULICS. BCPL was produced solely by Martin Richards in response to Christopher Strachey's suggestion that he design and implement a lower-level subset of CPL for systems programming. I think it was Martin's PhD topic at Cambridge, and he first implemented BCPL while on a visit to the MIT Project MAC group in 1967 and there were some contributions from members of the Project MAC team. Christopher Strachey subsequently adopted it for an operating system project at the Oxford Computer Lab after he had moved there. The wikipedia BCPL page tells more.--Georgio8 (talk) 11:10, 4 July 2010 (UTC)[reply]

JID: In http://en.wikipedia.org/wiki/BCPL I can't find any reference to ULCC (and we certainly imported our PDP11 compiler). I also wrote the IBM 3270 emulator for the BBC Micro in BCPL with some embedded 6502 assembler. The compiler and run time system was a standard product for the Beeb micro and it even supported coroutines (which I used for multitasking in the emulator). See "Networked IBM 3270 emulation on a microcomputer", John I Davies (Interfaces in Computing Volume 3, Issues 3-4, September-December 1985, Pages 189-198 Elsevier B.V. )

JID:The demons of Wikipedia seem to have reverted my piece to a "userspace draft" although its still at - http://en.wikipedia.org/wiki/Institute_of_Computer_Science In case of further accidents I have appended the Wikipedia source text to an email to conrributors Peter, John, Mick, Andrew - as above. Does anyone know how to make this a "real article" while keeping those demons happy? </Initial entry from John Davies>

Senate House Archive

The reference to those "25 boxes" right now seems to be - http://www.aim25.ac.uk/cgi-bin/vcdf/detail?coll_id=7704&inst_id=14&nv1=search&nv2=basic —Preceding unsigned comment added by Jid (talk • contribs) 10:25, 4 July 2010 (UTC)[reply]

AH: Paul Samet provided me with the following:- Before ICS was wound up the University set up a committee to investigate it, under Sir Harry Melville, principal of QMC. That was late 1972 I think. The report gave a complete history of ICS with a commentary on how well it had functioned, before giving its conclusions and recommendations (ie that it be closed). That report must be available in Senate House papers.

AH: here is a more complete draft list of where staff moved from ICS when it closed down, including information supplied by John Washbrook and Paul Samet.

Birkbeck: Professor/Director of ICS: Dick Buckingham (now deceased)

QMC: George Coulouris (not initially, now emeritus professor and Mike Bernal (not on the main wikipedia staff list for ICS)

UCL: Peter Kirstein (now emeritus professor), John Washbrook (now emeritus professor), Peter Higginson

Imperial College: Dave Howarth (later Professor)

Dick Housden - went to UEA (directly?) as Professor (not on the main wikipedia staff list)

Westfield. Alan Fairbourn, David Hendry and Michael Bell

The University of London Institute of Computer Science – My mmeories

Peter T. Kirstein, July 4 2010

1. Preface

I joined the Institute of Computer Sciencecomparatively late in its history in 1967. However from early 1964 I had been following the development of Computer Science research In the UK as part of my job. Hence I was aware of various broader management background that has not been mentioned in the material so far included in the Wikapaedia. I am writing this as a separate contribution, so that it can be corrected by those who knew the situation better than I. It is always easier to correct other people’s mistakes than to start from scratch.

2. History

In 1961, as Paul Samet wrote, there was an agreed 2nd tranche of computer equipment for the then comparatively few universities. This was prior to the Robbins report and the doubling of the number of universities in the mid ‘60s. There was already a “Buy British” policy, and it was decreed that each university would be provided with £250K to acquire an English Electric KDF-9 computer (or two in the case of the University of London because of its size). Most agreed with this allocation, but there were exceptions.

Manchester U wanted to complete and extend its Atlas I computer which it was developing under Tom Kilburn). Cambridge U, who had already built EDSAC, wanted to use its money do the Atlas job properly; it proposed to design and build Atlas-2 instead. The University of London saw a chance to get a far more powerful facility and make a profit at the same time. It proposed to use its allocation of £500K together with additional funds from the private sector, to acquire an Atlas-1 computer for £1.5M. It would pay for this acquisition by leasing one shft on the Atlas to BP, offer a general computer service on a second shift, and do its University work on a third shift. Since on paper the Atlas was much more powerful than six KDF-9s, this looked a very attractive proposition to the University. The government warned the university that if there were financial problems it would not bail it out, but agreed to the proposal. Thus the Bloomsbury Comuting Centre, which

included the Atlas Computer Service (ACS) all under Alec Robinson were formed in 1962.

In its wisdom, the University decided to form an academic arm to operate alongside the ACS. Hence the University of London Institute of Computer Science (ULICS) was formeds; Professor

Richard Buckingham (Dick) was its only Profess, though there were to be one Reader and several lecturers.

The Atlas was very late, finally working properly only around 1965 or 1966. This had severe financial ramifications for the university, who were not very sympathetic to such financial adventures thereafter, and also caused problems for ULICS. The ACS operated until 1968. At that stage the concept of such a public-private venture was abandoned. It was replaced by the concept of a Regional Computer Centre, and the University of London Computer Centre (ULCC) was established, initially still under Alec Robinson, in Bloomsbury – with no further links with ULICS.

Michael Barrett was appointed as the Reader, and wanted to do research which required the use of the Atlas. When the Atlas was late, he wrote some highly derogatory articles to the press and resigned with considerable, and undesirable, publicity in late 1964. The University reacted by deciding to re-advertise the Readership, with a closing date of June 1965. I saw the advertisedment, discussed it with Dick, and though it was after the deadline applied. Dick suggested I come over again from Switzerland for a further discussion on the position. I was horrified to find a full Board including the Principal of the University and under the Chairmanship of Professor Hermann Bondi. While I had firm ideas of what I wanted to do, it would need substantial outside funds – which ULICS were quite unable to provide. I did not get the position, and Brian Higman was appointed. However, I was informed that another Readership had been magically established, and ULICS would back my proposal to the Science Research Council (SRC) if I was interested.

I was interested. I applied and was awarded the INDRA grant for over £50K for remote graphics. This was a huge grant at the time. However, the grant depended on my being in post, and all posts had open competition. Again there was a Board, but this time the University had the choice of appointing me or losing the grant. I will never know whether I was the best candidate, but I was appointed Reader in 1966, and took up the post in July 1967.

In 1968, Brian left for a Chair elsewhere. It was decided to replace him with a Professorial appointment. This time David Howarth was appointed. As far as I remember, David was appointed to run the teaching programme. Soon thereafter Dick organised the research into two groups: an Information Processing one under Keith wolfenden and a Systems

one under me. Both Keith and I were appointed Professor in 1970.

Around 1970, the University realised that it needed Undergraduate courses in most or all of the big Colleges – but it was not prepared to meet the expense. For this reason, it set up a Committee under Melville (the Principal of Queen Mary College) to decide on the future of ULICS. It fairly quickly decided that the Institute should close, but the staff should be largely free to move to which ever College they wanted. One group, which included Mike Bernal and David Howarth, moved to Imperial. Another, including Alan Fairbourne went to Westfield, a third with David Barton and Mike Clarke moved to QMC and probably the largest (Peter Higginson, Benedict Nixon, Eric Nixon, John Washbrook, Steve

Wilbur, Keith Wolfenden and myself moved to the newly transformed Department of Statistics and Computer Science at UCL. After the official dissolution of ULICS, two years elapsed before the UCL contingent could move to UCL. It was during this period that David Cooper was appointed Professor at UCL.

3. Research

I do not remember all the funded research activities at ULICS – and certainly none before I came in 1967. The only ones I remember are some on databases under Keith Wolfenden, on One systems under Alan Fairbourne and several on networks under me. David Hendry’s led to a company being set up. This was not successful financially, and did not last long. Alan’s was going strong at the time of the move to Westfeild, and moved with him. Mine had the most people – inducing the Arpanet node mentioned in some of the references. There were quite a few research and systems staff involved, and some three PDP-9 computers and an Arpanet IMP. It was a considerable shock to the Department of Statistics that it was expected to house this lot – hence our group staying in Gordon Square a further two years while the relevant premises were readied on the main campus.

</contribution from Peter Kirstein via John Davies>

mention of Barnett[edit]

If the insertion of my name and title in the list of staff and my book (which discusses the Institute in a couple of sections) is COI delete it. Thanks. Michael P. Barnett (talk) 19:24, 4 February 2011 (UTC)[reply]

further additions[edit]

1. Richard made a MAJOR contribution to theoretical chemistry. He published beautiful papers long after his Buckingham potential paper. I did a minimal cleanup on the Buckingham potential -- it was rather weak, but if anyone else is interested, please run a literature search on Richard -- and find a way to put in the references.

2. I do not know how you got away with the editorializing "... still alive". I agree. But I think it should be moved to discussion before the wiki legalists home in.

3. I would like my DSIR grant to be mentioned -- it was subject of an earlier Times article -- but I do not want to push my luck.

4. I have copies of the Ph.D. theses of Gerard and Sambles and will provide references to justify mention before wiki legalists attack.

5. Can anyone find the notorious John Davey article?

6. Can anyone find my letter to The Times, just before I left, about the Atlas error that made exp[2]*exp[2] NOT exp[4]? Ferranti people told me that it upset Sir Basil.

7. I changed Mike to Michael. This should be formal. Also, I think several of the staff had middle initials.

8. I can't help my name coming first alphabetically.

Michael P. Barnett (talk) 22:10, 4 February 2011 (UTC)[reply]

Date of founding[edit]

Just found this on what I think is obits of presidents of IFIPS [1] re Buckingham: "He was the founder Director of the University of London Computer Unit in 1957, which later became the Atlas Computing Service and the University of London Computing Center." 128.112.70.30 (talk) 02:21, 21 February 2011 (UTC)[reply]

Notes from Peter Higginson[edit]

A couple of additional names:

Ian Gould (who was my MSc Supervisor) Adrian Stokes (who wrote the RFC 588 referenced)

If you want more references, Ian Gould and I wrote three papers on the PDP-9 work for DECUS 1971/72 and Peter Kirstein and I wrote "On the Computation of Cyclic Redundancy Checks by Program" for the Computer Journal in 1972.

PeterLHigginson (talk) 14:29, 15 January 2012 (UTC)[reply]

Note from Bruce Bovill ULACS 1965-1968 and ULCC 1968-1991[edit]

Elsewhere it was noted "The Atlas was very late, finally working properly only around 1965 or 1966"

Not specifically ICS-related, but I joined the Atlas Computing Service in March 1965 and it was fully operational then and, as far as I understand, had been since at least the start of that year. I had been interviewed there as an operator in January/February 1965 by Ken Mackenzie, when it certainly appeared to be in production. — Preceding unsigned comment added by 90.217.129.242 (talk) 12:37, 13 December 2012 (UTC)[reply]

Odd-looking list of RFCs[edit]

Does anyone remember why this particular list was chosen for the article? The rationale isn't clear to a non-expert like me, but hopefully obvious to someone else.

North, J.B. (August 1971). "RFC211 ARPA Network Mailing Lists".
Anon (October 1972). "RFC402 ARPA Network Mailing Lists".
Anon (January 1972). "RFC300 ARPA Network Mailing Lists".
Anon (May 1971). "RFC168 ARPA Network Mailing Lists".
Anon (May 1972). "RFC329 ARPA Network Mailing Lists".
Anon (August 1972). "RFC363 ARPA Network Mailing Lists".

--Northernhenge (talk) 19:27, 13 July 2015 (UTC)[reply]

External links modified[edit]

Hello fellow Wikipedians,

I have just modified 3 external links on Institute of Computer Science. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 09:51, 14 November 2017 (UTC)[reply]