Dr Stafford talking to Atlas Laboratory Staff
1 September 1975
Ray Rolfe
Geoff Manning
Bob Hopgood
Jim Hailstone
Jed Brown
Barbara Stokoe
Eric Thomas
Paul Nelson
Ray Rolfe
Bob Hopgood
Paul Bryant
Jim Hailstone
Jed Brown
Ken Robinson
Barbara Stokoe
Dr Stafford
1. Share of 360/195?
Questions: Click on a pink button to listen
2. Where do I work?
Paul Nelson
3. Atlas Representation?
Eric Thomas
4. Somebody from Atlas in charge?
5. Movement of staff?
Paul Bryant
6. Separate complement?
Anton Walter
7. Operators to transfer?
Ray Rolfe
8. Turnover of 195 staff?
Andy Jackson
9. New Engineering activities?
Ken Robinson
10. Redundancies?
Andy Jackson
11. New computers?
12. Who runs Atlas?
Jed Brown
13. Future ACC? Non-science work?
Bob Hopgood
14. Who will go to Daresbury?
Barbara Stokoe
15. Links to Daresbury and NCC?
16. Site Director?
Jim Hailstone
17. Justify two computer centres?
Peter Kent
18. Plans for 1906A?
Barbara Stokoe
19. Contract cleaners?
Kitty Northover
20. Atlas in the middle?
Anton Walter
21. Movement of staff?
Anton Walter
22. Crystalise roles?
23. Will C&A run ICF?
24. Federated Campus?
Anton Walter
25. Position of students?
Tim Gover
26. Organisation at Daresbury?
27. Contingency plans?
Peter Nichols
Good afternoon.
As you know, following the retirement of Dr Howlett last Friday, the responsibility for the
Atlas Laboratory fell to me, and so I thought it very important to come over here as soon
as possible, I also got back from the United States the same day as Jack did, and introduce myself,
and Dr Manning to you, and then, perhaps, the prime reason for the meeting is to enable all of you to ask
questions of myself or Dr Manning and finally I will try and say what I can about the future plans for the
Atlas Laboratory. Unfortunately they are not very clearly understood at the present time so I may not,
in fact, be able to answer all the questions that you may wish to put to me.
The basis for the plans is contained in the SRC General Notice, which you must all be familiar with,
the 21/75, and in there, there are 3 essential parts which concern the Atlas Laboratory.
The first part is in Section 3, where it is said that: a substantial part of the computing now carried
out at the Atlas Computer Laboratory to advance a wide range of sciences, will be transferred to the
Daresbury Laboratory where it will support a growing variety of work outside the field of High Energy
Physics, and it then says that the transfer will begin in 1976 and will take several years to complete.
Then it goes on to mention the Interactive Computing Facility which is being recommended by the
Engineering Board and that will come to Chilton and it says the part of the Atlas Computer Laboratory
remaining here will form the base of the new facility. Then, finally, there is comment about the Council
considering, with the Department of Industry, the establishment at Chilton of a national computing campus
to which the interactive facility would be the initial SRC contribution. Another part of that national
computing campus is in fact the Computer Aided Design Centre in Cambridge, which is the concern of the
Department of Industry,and so I imagine that, when this is worked out, that the SRC could be responsible
for the operation of the facilities here. Part of those facilities will be something of direct concern
to the SRC, namely the Engineering Board interactive computing, and then there will also be a unit,
resident in the buildings, who are responsible, professionally, to the Department of Industry.
I have already had some discussions with the people concerned about this in the Department of Industry
but there's a lot to be worked out before we really understand how this is going to be brought about.
As was envisaged in the General Notice, it's clear these changes are going to take a very long term to work out
and the general structure that will emerge is clearly not worked out in any detail yet. And it was
therefore essential for me to safeguard the current Atlas scientific programme from today onwards,
and for that reason, I decided that I should ask Dr Manning, who is my Deputy Director, to take personal
responsibility for me and he will report directly to me for the work of the Atlas Laboratory.
And it is in this way that I hope all your interests will be as fully safeguarded as I can possibly
make them.
Now, with these changes of course, or let me put it slightly differently, one reason for these changes
is the unhappy financial situation that the Science Research Council finds itself in,
and the country I suppose finds itself in, and they are aimed at making some substantial savings
in the recurrent expenditure in the Atlas, Rutherford and Daresbury Laboratories over the next five years.
And again, although I haven't had time to give detailed thought to these, nor has Geoff Manning,
nevertheless they are bound to have some impact on the Laboratory, but clearly this is not a consequence
of merging with the Rutherford Laboratory, but really a consequence of the rather dire financial situation
that we are in. But, bearing that in mind, I would like to stress that the two laboratories must,
in the future, be considered as one unit with the Atlas Laboratory as one of the Divisions, the operating
divisions, within the SRC Chilton complex for which I am responsible. This could well imply a number of
administrative changes so that we can have a common practice in the two laboratories; and, for this reason,
I asked Dr Valentine, who is the Laboratory Secretary, to be present this afternoon so that he can hear
any comments you make at first hand. Now, if there are to be changes, then I will make announcements from
time to time as they have been worked out, particularly after Dr Manning has been able to learn, and
myself has been able to learn, more about the operation of the Atlas Laboratory than I know at present.
Now, before I end and throw the meeting open for a general discussion, I would very much like to emphasise
not only the importance of looking at the two laboratories now as one unit in the future, but to emphasise
that it is now my job to do my utmost to safeguard your jobs and to make any changes that are necessary
with the very minimum of inconvenience to all concerned. On the positive side, I believe that by being
part of a larger unit this may well, and should well, provide greater opportunity for advancement than
might have existed in Atlas as a smaller Laboratory. So let me stop at that point and throw the meeting
open for a general discussion. Please feel free to ask any questions that you like, however aggressive
or otherwise, or however uncertain they may be formulated at the beginning; and although this is taped,
it is not my concern that it is being taped, and it will not be used in evidence against you;-)
So who is going to start? Right
Are there any immediate plans ..
Could you give me your name, then I will get to know people ;-)
Are there any immediate plans for the change in the set up whereby we have a 20% share
of the time on the 195?
No, the amount that you use at present is somewhat larger than 20% and there are,
in fact, no plans for making any significant change in the way in which you use the 195.
I might just qualify that slightly, and say that the Director of Daresbury and myself have been
charged by our Chairman to discuss what part of the computing can be transfered to
Daresbury, and so there may be some moving up there which would have a consequential effect
on the use of the 195.
Paul Nelson
I had a form to fill in today which said where do you work and I wasn't sure whether to put
Rutherford Laboratory ...
I think it is worth commenting on this quite a lot. What you should put for the moment is
Atlas Laboratory.
Administratively, this creates rather a nonsense in that one thing that is clear is a common
unit here on the campus, the effective unit, the division of that unit, must be in terms of
Operating Divisions. And there is a Division Heads Committee which manages the affairs of the
SRC's work on this campus.
Now it is also clear that when these discussions on the regrouping took place at the
Science Research Council many people in the universities and elsewhere were rather keen that
the title Atlas Laboratory should be preserved because of the reputation that it had gained
in the world of computing and, as always is the case, people are rightly proud of being
associated with a successful operation and Atlas is well known throughout the world and
therefore there is a problem really in reconciling these two situations.
It is also, in my mind, aggravated by the fact that some of the activity is going to be
moved up to Daresbury and as the notice says that the substantial part of the present
activities are going to be moved up to Daresbury, it could be argued that that part which
goes up to Daresbury should be called the Atlas Laboratory.
So as you can see I hope I have confused the situation really very successfully. In fact it
is really rather confusing and I don't know how best to resolve this at the present time.
One way might well be to have a series of Laboratories on the Chilton campus forming part,
say, of an institute but this is just playing with words really
For the moment, until we have had a chance to think about this, the only thing to do is say
Atlas Laboratory.
Are there any plans
Name please
Thomas
Are there any plans to widen the Division Heads Committee to include Atlas personnel or
are we effectively downgraded,in that what used to be the Group Leaders are reporting
to a Division Head who then reports to you rather than to a Director who was on a par
with you further up if you see what I mean.
I don't accept the way you put that.
Dr Manning is the man in charge of the Laboratory, of the Atlas, for the time being and
he will be a member, as he is now, of the Chilton Division Heads Committee and will have
full authority and say there for the affairs of the Atlas Laboratory
He himself will run meetings I would imagine similar to those which were run by Jack Howlett
when he was here. You will get full representation by HIM, or through him, and one could
argue that this would in fact be fuller representation than you had in the past.
The fact that he then reports through me to the Chairman is a difference but that is
something which the Chairman himself has decided and it is for him to decide how he delegates
his responsibilities.
Now I don't think in any way that this means - and I think it's important for us to spend some
time enlarging on this - that this does not in any way mean that the Atlas Laboratory will be
downgraded in its sort of authority or responsibility.
If I could comment a bit further on that, this is a personal view, I believe that in the
past within the SRC because the Atlas Laboratory was a relatively small unit, in the
discussions at the SRC, it was really very vunerable and was not sort of as protected as it
can be within the Rutherford, within a single unit.
I am really very convinced about this that the scientific affairs of the Laboratory and the
affairs of the staff are no wit changed to the detriment by this merger.
Why didn't you choose somebody from the Atlas to represent us?
That is very easy to answer. There are two ways of looking at this.
If I had named somebody in the Atlas Division/Atlas Laboratory as it stands then it would
have been that much more difficult for me to get to know what was going on because the chap
is somebody who is here, and he would in fact have all full knowledge of what was going on
and therefore you are not starting from scratch.
But a more important reason is that, with all these changes taking place, it is not at all
clear yet what the structure should be in the Atlas Laboratory and it is therefore not at
all clear what the qualifications are for a Division Head who will eventually take full
responsibility for Atlas. This arrangement for Dr Manning is an interim arrangement
although it is bound I think to take about a year or so before there can be a change but
it is very much a holding operation. Now as it has to be a holding operation because of the
uncertainties in the future the very best that I could do is bring in the most senior person,
other than myself.
That's the reason. The idea of naming somebody to take Jack's place was, obviously, in my
mind and would have been the simple solution but it would have preserved that division.
Essentially what would have happened would have been that Dr Howlett would have left and
somebody here already would have taken his place.There would have been no essential
change in the situation. It would have been very difficult to operate in that way.
Are there any plans in the immediate future, in the immediate future being let's say Christmas,
to move staff from this Division to work in other Divisions, or conversely from other
Divisions to work in this one.
Divisions in the Rutherford Laboratory?
Yes
The proper answer to that is no, there are no plans. On the other hand, both Rutherford and
Daresbury are, at the moment, above complement and there is a manpower squeeze within the
SRC and so that the first question that will be asked if vacancies occur either in Atlas
or Rutherford, is there somebody that can fill that post. There is nothing fixed.
I was concerning myself there with the scientific work. There is another area which may well
take place before Christmas although again there are no plans and I have undertaken to make
no changes until Dr Manning has had time to speak to the people here, particularly
Jim Hailstone, about the arrangements but one area which I foreshadowed in my introductory
remarks was in the area of Administration. Because there, clearly, one has to have common
administrative services in the two laboratories.
But again I have promised not to make any changes until we have looked at them very carefully.
I would imagine, however, there may be changes being made there before Christmas.
I would like to generalise Paul's question slightly. The way you phrased it seemed to imply
that Atlas would still essentially have a separate complement from Rutherford.
Is that what you meant?
It is a little hard to know for the following reason that when one goes forward to the SRC,
annually, one has to put down there the manpower, or one has to seek a complement level,
for each actvity and this will, in fact, be a common exercise in the future and so that in
the papers that go forward there will be an Atlas activity with so many people associated
with it. That activity will really be divided into two parts, the direct effort and the
overheads - as it is really at the present time. Now if you take those two together then
one will certainly have that activity defined and not all those people will be resident in
this building. This applies at the moment really in the sense that some administrative
services are provided to this Laboratory by the Rutherford, engineering effort, design effort,
I think we pay your salaries.
Will any of the operators be transfered over to the 360?
Not just like that but if, as I answered earlier, if there are vacancies then you will be
given the opportunity to apply against those vacancies
But will we want the opportunity though? Myself, I don't fancy the idea of working a
12-hour weekend shift.
Well if there are vacancies there and a surplus of effort here then I think, although you may
not fancy it, you are bound by your conditions of service to move to the jobs that you
are asked to do.
Andy Jackson: Won't that come under mobile grades though?
I don't think so, it is one campus.
If through the shift system you are working you get a high turnover of staff, don't you
think this might effect the staff that you move over to the Rutherford Laboratory?
I think we are beginning to get this out of balance.
There is a high turnover of staff both on the 195 and on the 1906A [??] and there is no
deliberate policy to move people from the 1906A on to the 195. But once again, with respect
to the last comment, surely one wants to look at the opportunities that arise within the SRC
on the Chilton campus and make the best of those opportunities, whether its the 195 or the
1906A. There is going to be no deliberate drafting of people. On the other hand, I am sure
that there are going to be opportunities in which people will move. It will probably be
better for them to move
I think it is probably safe to say that our attitude about transfering people from the
Atlas Laboratory to a Division of the Rutherford Laboratory will be the same about our
attitude towards transfering staff from one Division of the Rutherford Laboratory to another.
You will all be staff of one Laboratory with the same conditions of service and we will treat
the one the same as the other.
Is there any chance of the shift workers over here coming under the shift workers at the
Rutherford Laboratory?
We don't know at this present stage. Clearly, it is a possibility, we have no view on it yet.
The inverse can occur as well. It is quite possible that one fills vacancies here from
people in the Rutherford Lab.
If the turnover of the 1906A was the same as the 360 you wouldn't find many people left here.
We are always hearing of people leaving on the 360 which I think personally is something to
do with their shift working. I know it has to be manned 24 hours a day but asking people to
work 12-hour weekend shifts is wrong. If the operators are going to come under the same shift
working agreement there, you'll get more people leaving. You say Rutherford are above
complement but you have adverts for operators on the 360, it is a bit funny.
Re timescales, there is this Engineering Board and this Atlas Computer Complex.
Do you have any idea when these decisions are going to be taken.
Good question
Are we talking about months, years, decades?
Let me try and give as much information as I have and if there is anything that either Jim
or Jeff can add then please do so. An operative date for the transfer of the Computer
Aided Design here is some time in 1977. The reason for that is that there is a lease on
accommodation in Cambridge and that runs out I think in 1977 and so that fixes some deadline
for their move. The transfer of activities to Daresbury was stated in the General Notice to
begin in 1976 and take several years to complete, so that's another variant.
I haven't yet begun to talk to Professor Ashmore about this but certainly their computer
is heavily loaded at the present time and although they are supposed to be shedding some of
the computing on to the 195 clearly it seems to me that it is going to be rather difficult
to see any rapid movement in that area.
Now the engineering computing, well that proposal has just been worked out by a Working
Party and is going to the Council in November so the final decision on whether or not there
will be an interactive graphics will be taken then and, so clearly, until that is taken it
is not very profitable to speculate on how long it will take to be implemented because,
well we can make some estimates. It will have to go presumably to the Department of Education
and Science for final agreement, if there are large sums of money involved.
But, I don't know Geoff, whether in your paper you have estimated any timescale for this?
The assumptions being made are that one is endeavouring to get some computer installed by
something like 18 months to 2 years, so that fixes a rough timescale.
That means you want to see some people, I suppose, recruited before then for this work and
so that might well fit in with a year from now when things start moving I would guess.
End of 1976
Do you see any redundancies out of these moves? Can you see what groups will be remaining
at the Lab?
There are two questions there.
Redundancies: there is no policy of redundancies in the SRC but certainly there is going to
be a very big squeeze on manpower and its therefore going to be very hard to have a sensible
movement of staff, particularly if there is unwillingness to move as was put there. There is
certainly going to be, let me put it more generally, it's not specifically on Atlas, but the
manpower within SRC is envisaged to reduce over the next five years by a few per cent,
more like 10% in 5 years is the sort of reduction. I am talking of a total manpower of 2800
and it is coming down by 200 or so is the forecast figures but there is nothing specific in
that to the Atlas Laboratory. I might put it slightly differently. There is a lot of
computing being planned both here, at Daresbury, and the Rutherford and there is some part
of this reduction in the total manpower associated with computing, but again that is expected
to be within the bounds of the normal wastage that takes place.
If there is going to be a new computer, what type of computing do you have in mind,
a PDP8 or a STAR-100?
The sum of money which was included in the original paper which went to the Science Research
Council describing broadly what they were interested in, which was accepted in principle,
was a sum of money of the order of £5M, so it is not trivial, that is not all
for computers but is certainly more than a PDP8.
That was just the engineering?
That was the engineering computing over a period and included more than just a computer,
it included terminals etc.
There's also a discussion at the present time taking place on the more immediate updating
of the computing power within the SRC and at the moment that's heading towards, say, an
updating of either the Daresbury computer or the 195. It is heading towards an IBM machine.
But that's to do with the present programme and is in addition to the £5M that Geoff mentioned.
At the present time the scientific affairs of the Lab have been a direct concern of the
Science Board and the Science Division. Do you see a gradual change on the campus to direct
concern by the Engineering Board and the Engineering Division and, during the changeover,
do you see some joint responsibility and what part does the Nuclear Physics Board and the
Nuclear Physics Division play in this?
I'd like to correct what you said if I might. The direct concern of the Atlas Laboratory
is mine scientifically, and I will present the activities of the Atlas Laboratory, which
will be my responsibility, to present these to the Science Board with the Science Division
in London acting as the vehicle for doing this but it is my concern, and Geoff's concern
through me. Now, similarly, when it comes to the Engineering Board's interests in the
Laboratory, it will again be my personal responsibility to present this to the Engineering
Board and ensure that the requirements and the interests of the Laboratory are presented
directly to the Engineering Board.
Now, if you then say well it looks as though I am going to be spending a lot of my time
presenting these things to the various Boards, clearly you are right and some changes in
what I do will be necessary, and this brings me to the Nuclear Physics Board because the
Nuclear Physics Boards are increasingly only a small part of my responsibilities and so
that, I see the possibility - this has not been agreed by the Chairman - but it seems to me
that one possible way of meeting this problem is to have, say, a number of people delegated
by myself to be responsible for particular activities at the various Board levels and this
could be someone responsible for Nuclear Physics, somebody else responsible for Science Board
activities, say, and somebody else responsible for Engineering Board activities, with the
general coordinating role being my own with the senior people.
I am just thinking aloud here because none of this has been discussed outside the Lab,
nor agreed, but its obvious with this changing structure some changes are necessary.
It is also important in this respect for the Nuclear Physics activity not to be seen to have
some preferential treatment in the Laboratory which might be the case if I, trained as a
Nuclear Physicist, had direct interest in one and not the others.
It is a general problem incidentally because already, you see, we have the Neutron Beam
Research Unit in the Laboratory, there is likely to be a Laser Centre as part of the activity,
and then there is the Engineering Board activity and the additional computing.
It has been reported that many Atlas Staff have been sensitive about the impact of the large
Nuclear Physics Laboratory on them. I should say that really it is most unwise,
perhaps that's the wrong word, you are not right in being nervous about this.
For a long time operated the Neutron Beam Research Unit in the Rutherford Laboratory,
I believe satisfactorily from the user's point of view and the man, who is in charge of it,
has been able to represent the interests of the users at the Division Heads Committee,
even though he was one of a number of people there and he was certainly not dominated by
Nuclear Physics and he has had much advantage in being able to call upon specialist activities
in the Laboratory when he has been required to do so; and I was asking some time ago really
what people's reactions were about this and what Dr Hobbis's reaction were and he was quite
clear that he benefitted by being part of this larger unit rather than being an independent
operating unit.
So you don't think there is any possible question of conflict of interest on your part
when it comes to Engineering and Science?
No, there is certainly not on my part but the problem will arise in a slightly different way,
which will have to be resolved by Council, and that is that when you add up every year the
bids and the requirements of the Engineering Board, Science Board, and the Nuclear Physics
Board, they are going to exceed the resources that we can put in then clearly it is the
Council to decide how those resources are allocated, not me but the Council, and once the
Council has decided, with whatever advice can be given how those resources are allocated,
then we have to make sure that we carry out the programme accordingly.
For example, if the Laser Lab comes to the Rutherford Lab, it will have to be done with no
increase in manpower. The consequence of that is that the resources available to
Nuclear Physics will reduce, but this is understood.
Where does the Atlas Computer Committee sit in the hierarchy? Is it above you or below you?
I am not sure how long the Atlas Computer Committee will remain in existence, and the reason
for that is that there is a whole programme of change proposed by the Chairman to the Council
on its Committe structure, and one proposal is that there should be what's known as
Facilities Committees as distinct from subject committees which are more closely associated
with Boards and I just don't know how that is going to effect the Atlas Computer Committee.
Second Question
The original announcement was that a substantial part of the Science Board work of the Atlas
Laboratory was going to Daresbury. Somewhere along the line the Science Board bit seems to
have gone missing. Is there any definite plans about what happens to, say, the Space Research
work that was currently being done at the Atlas Lab and other things like that.
Just let me check Bob that you have got your words right It says here: a substantial part of
the computing now carried out at the Atlas Computer Laboratory.
That wasn't the original statement, that's a new one that has got dropped in half way through.
I see, so just repeat your question again.
What about all the non-Science Board work which at one stage the assumption was, as far as
we were concerned, that this would stay at the Laboratory. Now the query is where it is going?
I don't think it's determined
As I understood it, it was Science Board computing which was primarily the target for
move to Daresbury.
OK is that just an omission in the paper that came out or is there a change in policy is
really what am I asking?
I don't know, there is no change in policy that I know of.
There were in fact 12 drafts to my knowledge of that particular paper, 12 drafts and a final
version, and I am not at all surprised that one or two words got modified in the process.
But there is a substantial amount of work on site that is non-Science Board
It seems to me the only way in which this question will be answered is take the programme
that exists here at the present time and then ask the question:
a) can this work be done on the 165?
b) is it appropriate for it be done on the 165?
And there are again two aspects to this, the first aspect is on what computer is it being
done, that is less importance I think than where are the people resident who support that
activity.
Now the argument that was put when these proposals were made: there is some common link
between much of the working that is going on at Daresbury particularly when the SRS comes
into operation, solid state type physics, and some of the work that goes on here and,
therefore, if you require scientific manpower to support this work then it should all be
lumped together to have the largest attack possible, a common attack on the general
problem and not set up two units
That I think is the criteria that one will ask is: how does this link with the work that is
necessarily resident at Daresbury and that's what will be the target for moving;
that's obviously relevant because in a few years time whatever computer we will have got
will be closley linked together.
What sort of people might move to Daresbury? Will it be the scientists or the support
staff or what?
I honestly have no ideas whatsoever I really just don't know.
Can I turn that round. One of the things that I consider I need to do very urgently,
first of all find out what is going on, and, secondly, get the views of the staff what they
themselves are interested in. It may be some people want to transfer and maybe other people
don't want to transfer; and I would like very much to hear the views of the individuals
concerned and, before we come to any conclusions, I would like to go through that process.
It is perhaps fair to say when the original estimates were made, the figure that was quoted
for the number of people moving to Daresbury was what Geoff?
The figures that were put in, which had very little in the way of quantitative basis, ..
They were put in by the Chairman as his personal estimate, he made that clear
It was 55 for Engineering Computing and the number added to the Daresbury number was 45 if
I remember correctly combined with their numbers already to be the computing activity there,
and those two everybody can determine does not add up to the total of the Atlas staff.
I assume they are complement places, presumably?
Correct and not necessarily bodies that will move.
Certainly if you took those figures then they will certainly need more operators up there
if they are going to do more jobs but they will also need the appropriate supporting and
scientific manpower.
I don't think that Jack and Alec ever got anywhere it making those more quantitative,
did they Jim?
Indeed!
A couple of things that I am not clear about following up from that, is that people who
do move to Daresbury, what if any will be there links with the people left here, and if no
links, how will they be represented in terms of the discussion we've had earlier in deciding
on what they traditionally do?
I am not quite sure what the position of the National Computing Campus is, which I believe,
if I heard you right early on, said that the Engineering Board thing might be a contribution
to this which seems to suggest that the NCC has some life outside the SRC?
This seems to raise another series of managerial things and I wonder what kind of Transkeii
positionable, separate but equal development, we will have.
Is this yet again another of things that hasn't been worked out yet?
Probably you are too young to have known Sir John Cockcroft very well, but he was a great
chap and, in a case like this, he would say Yes. Yes meant: yes I heard what you say, not
yes I agree. He never said yes, I agree.
Let me try and answer as far as I can. If any of you move to Daresbury then you are part
of the Daresbury Laboratory. It's as simple as that and so the representation is through
the Director of the Daresbury Laboratory.
Talking about the National Computing Centre, I mentioned this earlier but let me just try and
repeat and amplify what I said. There will be several components to this computing centre.
Administratively, clearly if you have got buildings here, they have got to be owned by
somebody, and then they have got to be serviced and managed by somebody. Now, I believe, the
only logical way of doing this, let me emphasise that this is not agreed, would be for the
SRC to own the real estate and to provide the appropriate services: transport, accomodation,
cleaning and so on. As regard the scientific activities, they fall there into the
following categories: what's left of the existing activities - let's call those for sake of
argument the Science Board activities - the new Engineering Board interactive computing
activities. All the activities will be the responsibility of a Division Head, who will be
answerable to me for all the Science Research Council activities that go on in this
National Computing Centre. And he will have the same seniority as the old Dr Howlett role
and will be able to speak in the various bodies within the SRC for those activities.
Now in addition to that which is clearly SRC then there will be some other Groups here;
one that is clear, the Computer Aided Design Centre, they will come here; they will have will
have accommodation; they will be charged for that accommodation, and the services that are
provided by us at the appropriate level, we get that revenue. The reason they come here is
that much of their work is scientifically common, or to some extent common, with that which
the SRC wants to do, and, therefore, you put them here so that they can interact
scientifically and technically with the SRC staff, but, of course,they will be answerable
for what they do to their own department. Now I believe also that the SRC Division Head who has
to be appointed would have to have some responsibily overall for the running of the Laboratory,
not scientific but he clearly would have a coordinating role to play, and so its a big job.
I don't think it has been determined whether it will be a Federal Laboratory or a Federation
of Laboratories. Clearly there is a significant difference between those. That's the degree
of certainty at this stage.
Except they would be using a common computer and living in a common building.
If you asked me to suggest the logical procedure then I would agree with what you have said
but I don't think its formally agreed
That's the way I see it but Geoff's right, nothing has been agreed. Although, in talking
to them, they seem to be fairly flexible at the current time.
This question of the site director, because it is the key item to this whole business of
changeover management, and one of the things that has been given almost no attention: I made
the same remark to Professor Edwards and he has nothing more to add except that, yes it is
obviously necessary to have a site director, and this is where the difficulties come in,
in the domination by High Energy Physics argument that the Division Heads organisation is,
at the moment, inappropriate to the site direction. Do you have any comments on this
Yes: I disagree. Let me go through the Division Heads Committee
There is one man who is responsible for High Energy Physics
There is one man who is responsible for Administration
There is one man who is responsible for Engineering Services
There is one man who is responsible for the 195 Computing
There is one man who is responsible for Applied Physics generally and not specificall related
to High Energy Physics
There is one man who is responsible for Theory, which is admittedly High Energy Theory but
the number is 10 people so it's a small unit but nevertheless it is very valuable to have a
theorist on the Division Heads Committee.
There is one man who is responsible for Neutron Beam Research Unit, which is a Division of
the Rutherford Laboratory,
So there is certainly no domination there by High Energy Physics, and, in fact, the reverse
could well be argued: that with the scientific activity being dominantly High Energy Physics,
its representation at the Division Heads was not proportionate.
Now it follows, therefore, if you now have somebody representing Computing other than the
195, then it is certainly going to have a fair say, and you must remember that we discuss
management affairs there globally, and we also discuss such things as the presentation and
preparation of papers that go outside the Laboratory to the various Boards, and these have
been diverse in the past, obviously from Neutron Beam work to High Energy Physics, and again
this has proved extraordinarily valuable, in the sense that you get a group of people
seeing papers which have been prepared by specialists, and so they look at them somewhat
on the lines of people not directly interested, similar to people in the Boards - who are
frequently not specialist and I have found, in the preparation of papers, that this extra
look has caused papers to be substantially changed and this has been extraordinarily valuable
when I have to come to present them to the Boards. And so, in that context, papers that go
on Atlas work to the Science Board will have careful scrutiny within the campus, within
the Division Heads Committee, before they go up; that will be an advantage.
Might it not be more relevant to look at the possible domination by High Energy Physics
interest for the actual computing scene on the 195. It is clearly different from the
management scene of the whole of the Chilton campus. There, there is certainly a
preponderance of High Energy Physicists and let's say, nuclear physics interests.
There, it is quite possible, it seems to me that there could be domination.
Why has this changed?
I don't think it is going to get a chance to change, and yet maybe needs to.
But you have been living with it?
And not enjoying living with it.
Let me answer it then, if you have not been enjoying it. I'm very sorry about that!
The provision of a substantial fraction of the actual requirements by another laboratory,
with rather remote contact, in the past, has obvioulsy led to the possibility of conflict.
This situation, in the future, is open to correction because you will have the head of the
195 division and the acting head of the Atlas Division being able to discuss their
requirements at one common place, and so, if there are, in fact, difficulties, I will
certainly be able to be aware of them much more easily, and directly, than I was before.
And so that is a positive advantage, if there are difficulties there is a means now
whereby one can identify them and resolve them, hopefully.
Can I comment a little perhaps some people are not be aware that the 195 was in fact
purchased because of the argument that High Energy Physicists and Nuclear Physicists required
it. That was the argument for it. The original discussion at Council, in fact, once spoke
about possibly, in the long term, 10% being used by other users. I can quote the minutes to
you if you wish, but that is absolutely correct. In practice, they have used much more.
I think, ignoring all those aspects, the SRC has before it a great demand in the future for
computing, not only for Nuclear Physics but also for other activities, and I think the thing
one should address: what is the most efficient way of getting increased computing power,
and how can one most effectively get it into use. If one tries to present it from the
viewpoint that there is a 195, why can't we have a bigger share, one is deliberately
creating the thing that you are stating that you don't wish to have - a conflict.
I am very interested in your comment here but its very clear that, if there are problems,
then it's going to be much easier now to identify what those problems are and resolve them.
One hasn't got as much computing as one needs, in fact, for all the jobs. The 195 is now
pretty well fully loaded. The formal share, as Geoff said, that you have been using is
somewhat higher than was notionally agreed, it was only notional, and it is for Council
to decide how to divide these things up anyhow, eventually. But the predicted growth from
High Energy Physics is doubling in five years, its a prediction - it may be excessive, but
certainly with the new activities at CERN, it is certainly on the increase, and we certainly
have to take on some of the Daresbury work so there is obviously going to be pressure,
and we have to share out, as best we can. On the other hand I hope that one might see ways
of getting more computing power in the future. That would solve the problem.
Can I just make a comment on the difficulties with the 195. The 195 was really only agreed to,
providing that the Atlas had a very substantial share. This is the statement that was made
in the minutes.
The fact of the matter is that in dealing with the problem, where we had a substantial part
of the 195 to dispose of to our users, we had to agree who was going to manage this and,
of course, this management has to be in one hand. It can't really be in two sets of hands
and so it was agreed that we would in fact keep the management under one control.
This has a number of disadvantages and I think this is what Jed was refering to: that if you
don't have absolute control, then you can't do the things that you want to do that is against
the interests of the 80% share. I think we do run into real difficulties about not being
able to do the things that we would wish to do and somebody has to manage and, in this case,
it is Nuclear Physics. That situation could continue but we would rather see it change
so that a better management of the 195 from our point of view was seen to be set up.
If there are difficulties I think that, now, there is a better means for us to achieve
coordination
In the long term do you see any difficulty in justifying the existence of two separate
computer centres on the same site
No
You've got to distinguish here I think between operations and the scientific activities.
And clearly I think, long term, particularly as any changes in computing are likely to be
towards larger size of the computer rather than a number of smaller ones
{I suppose that may not necessarily be true),
but the trend in the past, let me put it this way, has been towards larger and larger computers.
Therefore it seems to me that, if you had a centre like Chilton, to have two very large
computers, as this would imply, next to each other under entirely separate management
doesn't really make much sense. Now you could imagine, for example, that we might go towards
a single large computer, perhaps with peripheral computers of some sort, where the operations
were done as a unit and the scientific activities
were divided in some other way. After all the operation of a computer is a service as
distinct from the scientific activities that go on.
For example Nimrod is a machine which provides a service for High Energy Physicists and it is
operated as an independent unit, with the proper coordination between the man that runs that
and the High Energy Physics community
Could I make a comment about that point as well. I think that this is a major difference
between the C&A Division's interests and the Atlas Laboratory's interests. The C&A Divison's
interests are providing facilities whereas the interests of this Laboratory are in terms of
not only providing facilities, but in terms of computation as well and this is a very
important point. It means that our interests are, in terms of computation, are much wider.
We have, in fact, an interest not only in computers, but in the science of computing and in
the question of computation and I think those three things are important in distinguishing.
Perhaps Rutherford's attitude is that this is a facility just like Nimrod. The scientific
work is another thing. Here we have a very different attitude, and it must be clearly
understood that that is the case, because otherwise we will get a confused state,
if we assume we act in the same way as C&A Division
I appreciated this Jim; I did understand that you didn't work in the way that C&A did.
That there were significant differences.
What plans are being made for the 1906A?
There are no plans being made at all.
It is one of the problems that we will have to look at.
You said early on that there would be some changes in the Administration.
The Atlas Laboratory had a very good reputation for the cleanliness of the Lab.
Good
As you are going to make it one unit, does that mean you are going to bring Contract Cleaners
over this side
That's a very detailed question and I don't know in fact what the situation is at the present
time here.
Nor do I
Have you got your own cleaners here?
YES!!
We rejected contract cleaning so long ago that we just can't remember it. It just doesn't
work, in our opinion. Which is really Kitty's point I think, a number of the Administrative
staff are in fact on complement whereas in the case of the Rutherford this is not true.
I might say you are talking to the converted.
In the Administration, and the effect of users and the 195 (nothing to do with me, I am not
in that part of the Lab at all) I always get the impression that the real problem is that
you have Rutherford sitting here, and you've got the user in Belfast with a little 7070
terminal sitting here, and poor old Atlas in the middle just being shoved from both sides,
and basically there are just too many layers down there. You don't agree?
I don't really follow your remark about poor old Atlas.
The problem is that Rutherford or C&A Division hears about the problems our users are getting
second-hand, and, hence, they don't really feel the urgency of them, or the degree of
annoyance that the user is trying to put forward, because it is being relayed it is buffered
all the time. C&A tends to think of Atlas as A user rather than as a big complicated group
of users, with varying needs. When Rutherford say: well we introduced this facility because
A, B and C of our people here find it useful, and we say: yes, but it causes difficulties
for us; then, Rutherford thinks there are 3 people who like it and one person who doesn't,
which isn't the real case at all. Comparatively speaking many of our users are very small
in a sense, in terms of the computing they use, but none the less, in terms of shear numbers,
it does cause a problem.
You are really defining the situation which has existed up till now. You have to, be because
you can't define the situation in the future. But the situation is now in a very favourable
position for easing this problem in the following way that: Geoff Manning will, I trust, in a
very short term, have full and complete understanding of these sort of problems and,
therefore, straightforward for him to present and clarify these problems at the Division
Head's Committee and no significant changes of policy can take place on the 195 without a
full discussion at the Division Head's Committee; and so, now, this is the crucial point,
there is a man from here talking at a common body that I run as distinct from the past,
in which, with a separate Director, there was no discussion on these problems at that sort
of level. And there was no means of reaching a decision. If you have two Directors,
the two Directors can eventually agree to differ, if it ever came to that, and can only be
resolved by the Chairman and that of course is a tricky business. Now we have a means, on
site here, of trying to do the best for everybody, and, if there are therefore particular
problems like that, at least we have a means for airing them and understanding them;
and, having understood them, we can then come back and tell you what we have decided
and, hopefully, you will understand the reasoning behind it.
Do you see any possibility of physical staff movement between our buildings and Rutherford
buildings at all for any reasons. I don't know how much the Divisions in Rutherford are
essentially in separate units and how much they are intermingled.
We aim to keep the Divisions as separate as possible although they are somewhat mixed up.
I don't see much movement; I would in fact, frankly, have liked to have seen closer physical
proximity between the computing work here and the 195. Can't see it happening. Frankly, I
would liked it to have happened. Then you are in a position to talk to more of those people
This division is rather artificial.
Will there be any attempt to, as it were, crystalise the two computing roles that are
presently going on in this campus. At the moment you will find this Division and C&A
Division doing in part exactly the same job.
There will certainly be hopefully a lot of effort going in to getting this much more
coordinated. I am sure that's very important and, again, the vehicle for doing that,
again, is very clear now, through common responsibility
You see, at any time, I can ask for a discussion between Geoff and anyone else who is
appropriate in the Laboratory here, and Walkinshaw, and anybody else in his Division,
and so there is a means of getting, if there are problems, on either side. If Bill has
problems or if Atlas has problems then we can find a way of understanding them.
Will the C&A Division most likely take responsibility for the actual operations of the
interactive facility, the operations.
That's not decided at all yet. And the reason is that the interactive facility will require
some computing. The scale has been defined by Geoff but what this is going to be is not at
all understood yet or where it is going to be - what fraction is on this campus and what
fraction is computers in the universities.
We keep talking, or people keep talking, about federation or federated campus or site or
Laboratory or what have you but all I have ever heard is the SRC and the D Of I. Federations
to me are things like six states and two empires. Just SRC and the D Of I don't seem to be
much to make a great big fuss about. Are there any plans for any other government or
non-government organisations to come on to the site.
Yes, they are very vague but there are other possibilities
Any idea of numbers two or three or thirty?
At least three other Groups, in fact some of those being within D Of I, separate from the
CAD Centre. but these they are very very vague and they are not agreed and not agreed by
their parent bodies.
Is SRC going round looking for people to say: please would you like to come and join us,
or is there someone in the higher level in the hierarchy up above SRC that is trying to
coordinate all these things?
It is more than just the SRC just going around saying will you join us. It was at a high level.
There was very strong pressure in the discussion by the Chief Scientist of D of I.
He wanted this to happen and he wanted it to happen here.
You were talking here about reduction of manpower in the campus. Do you think this will
effect the position of students involved in the campus, a cut back in numbers?
It has nothing to do with that. Students are funded by SRC Grants and so it is an internal
university matter dependent on the number of grants that the Council decides to offer.
I was not talking about that at all but purely SRC employed staff.
What about the position of sandwich students?
That won't have an effect too. It is independent of this.
Will the staff that go to Daresbury form a separate Division there or will they be
amalgamated in a Division within Daresbury
I really can't answer that. I'd hate to interfere in the affairs of yet another Laboratory
All the plans that we have heard so far are based on an asssumed budget over a number of years.
If we are to believe Professor Edwards seriously on the remarks on Friday about the cuts in
public spending, are there contingency plans for the activities put forward.
I think for the benefits of the audience, Peter, you should say what you read in the paper
I haven't read anything, I have been hearing remarks, today, about a talk that he gave at the
British Association on, I think, Friday. He talked about posible reductions in funds for
government activities and, if this happens to come about, are there alternative plans
for any of the changes?
It is extremely difficult to answer that question. At the discussions that have gone on over
the last several years at the Science Research Council, there have been a series of plans
made on various assumptions about the budgets. The current one is based on the assessment of
what the financial situation is likely to be. And the best knowledge that we have is that the
worst situation there, is the one which will apply, but it is not significantly worse than
the plans were put forward last year on what was called Programme C.
But the regrouping exercise is essentially part of that Programme C. Now having said that,
clearly, if the financial situation gets worse for the SRC, then it makes the squeeze that
I was talking about earlier even more relevant, but I am sure, I don't want to make this a
gloomy meeting, One has to be aware of the fact that the SRC has been under attack for many
years financially, and the High Energy Physics community has been under more severe attack
than any others, so it is no surprise to me to have a situation where one has to fend off
the worst of the financial attacks that may be put on us.
Simple, my job is to get the best deal I can for the Staff of the two Laboratories.
It is probably worth stating that almost certainly what Edwards refered to were the cuts
which we were already aware of and he was making the Public aware about something that we
knew about some time ago. If you are asking: is tomorrow's news going to be worse still,
your guess is as good as mine
Let me just say what I read in the paper: it was a talk by Sir Sam Edwards at the British
Association. His theme was the pressure on the research funds in the country and as the SRC
was the largest of all the Councils, taking 60-70% of all the money, that is the one that
came under attack. Its the biggest one. Similarly, High Energy Physics, which was the biggest
spender of the lot, came under attack. So the treatment, that has been handed out to me in the
past, is no differentto the treatment that is now being handed out to Sam Edwards by his seniors.
What he said then was that, with the cuts that were being pressed on one, the High Energy
Physics situation was very bad. In fact, all the larger spenders, High Energy Physics, and
then Space Research, those pure research topics came under the severest of attack.
He then went on to say things about the problems of funding a national programme against a
background of having a first rate proposal.
I would like to see the full text of his speech because I've read two reports, one in the
Times and one in the Guardian. The Guardian was slightly more detailed and presented a
slightly different slant. The Times had missed out a few key words. The Guardian had included:
the High Energy Physics remains stubbonly high, due to the devaluation of the pound and the
contribution at CERN. Whereas, in the Times, it just sounds as if, inspite of Edwards efforts,
the High Energy Physics remained high and he couldn't do anything about it,
but he did qualify that by saying it was a great pity.
Two of us were actually there. The High Energy Physics was stubbonly high because one is in
an inflationary society and one is at the mercy of the exchange rate of the pound. In a
large international programme, High Energy Physics and the Space Programme,you are bound to be
in serious trouble and you might land up with no liquidity at all for your national projects.
And he actually said he forsaw the possibility of having, the worst comment he made,
no domestic High Energy Physics Programme.
He meant that as a serious comment. There could be a situation where this could actually
happen, if ones international commitments became so violent because of the pound going so low.
So the Guardian Report was more accurate than the Times.
It missed out some things as well
There you are, if we spend a lot of money we are always under attack. Anything else?
Well we have had an hour and a half. So I suggest if there are no more questions we will call
it a day. Thank you all very much for listening.