Resurrection Home Previous issue Next issue View Original Cover

Computer

RESURRECTION

The Journal of the Computer Conservation Society

ISSN 0958-7403

Number 80

Winter 2017/8

Contents

Society Activity  
News Round-up  
CCS Chair’s Annual Report David Morriss
GPM for a Challenge P.A. Cherepanov
Smoke and Mirrors — The ICL 2905 Dik Leatherdale
A Life in Computing Alan Freke
Obituary : Geoff Tootill Martin Campbell-Kelly
Out of Oslo Peter Short
Yet Another Differential Analyser From The Meccano Magazine - January 1951
Letter to the Editor Mike Robinson
50 Years Ago .... From the Pages of Computer Weekly Brian Aldous
Forthcoming Events  
Committee of the Society 
Aims and Objectives 

Top Next

Society Activity


EDSAC Replica –Andrew Herbert

In recent months we have been beset by valve and mechanical wiring failures tripping out the machine. Frustratingly several days have been lost to fault finding rather than making forward progress. Some adjustments have been made to the sensitivity of the power supply safety circuits and this has improved matters. The main focus for commissioning is Main Control and getting the order fetch and decode cycle working.

We have finally taken delivery of the last batch of metalwork for the remaining chassis from our supplier — a change of ownership has led to significant delays but we now have all we need to finish the machine. With the delivery of the metalwork for the long delay pizza boxes Peter Linington and colleagues are starting on the assembly of the main store delay lines.

Analytical EngineDoron Swade

Work has continued steadily going through Babbage’s manuscript Sketchbooks (his working notebooks) and building up the cross reference database. As reported previously the volume of material is substantial and its organisation is not conducive to systematic study: it not indexed or themed by topic, nor is it rigorously chronological. Time and time again Babbage revisits the same or related topics over a period of decades, and these entries, many of which are cryptic, are dispersed through the 25 volumes comprising some 7-8,000 manuscript sheets. The cross-referenced database is being developed by Tim Robinson as a key research tool to manage this distributed content.

The focus for the last six months has been on the volumes most relevant to the Analytical Engine design started in 1834. Happily this main sequence of books is, for the most part, in chronological order. We have just completed Volume 5 which covers 1841-1844, and includes the most intense period of work by Babbage on Plan 28, the most advanced design for the Analytical Engine.

There are notable gaps corresponding to periods when Babbage was travelling. There is a separate “Travelling Scribbling Book”, which has not yet been studied in detail, and we know of a significant body of manuscript material in the Buxton archive at Oxford which was written while Babbage was travelling in Europe in the early 1840s. The intention is to fill these gaps in due course.

While focussing on building the database, we have resisted the temptation analyse the material as we go as such analysis will be substantially assisted once the full corpus is in searchable form. Having said which, there are already new findings. A landmark drawing is Plan 25, dated 5 August 1840, which depicts the “Great Engine”. This Plan provides the most complete system overview of the state of play at that time and is the most well-known of Babbage’s technical drawings. The version of the machine in Plan 25 is that Babbage presented in Turin in 1840 and on which Menabrea based his published description, which in turn prompted Ada Lovelace’s famous Notes. The new finding is that there is no evidence of “user level” conditional control in this defining version of the design, this despite provision for sophisticated multi-way conditional branching at the “microcode” level. Given that the 1840 version was incapable of user-programmable conditional branching, the machine described in Plan 25 should be regarded technically as a calculator rather than a computer in the modern sense.

The first evidence of user-programmable conditional branching appears in an entry dated 19th March 1842 in what Babbage called the “minor operations” — ascertaining if a variable is zero, and ascertaining if a variable is + or -. This was when Babbage was well into the design of the “Small Analytical Engine” — a reduced version described in Plans 26 and 27. Work on Plan 28 (yet another complete reset) started two months later, but these “minor operations”, essential to making a “universal” machine, were carried over unchanged. It is yet to be established what suddenly triggered the addition of these operations. Babbage himself gives no hint of the reason in the Sketchbooks studied so far.

Some 2,100 pages have so far been cross referenced, indexed, and transcribed into the database. Although this represents only about a quarter of the total manuscript material, we are much further than a quarter of the way through the process: most of the remaining volumes are more fragmentary, contain material not related to the Analytical Engine, or contain rough drafts later transferred to the drawings and/or the notations (textual content already captured). All the remaining volumes will be worked through though these are not regarded as a priority at this time.

Elliott 903Terry Froggatt

Amongst the software available for the Elliott 900-series computers is a program called ‘Workshop’. This program can be used without any knowledge of computer programming to calculate mathematical formulæ on demand. To quote the manual, “Once given mathematical formulæ, the machine writes its own programs, runs them, and provides the results without further attention. The formulę can range in complexity from the simplest addition of a calculating machine to the intricacies of design calculations and statistical tests.” It was written in the Department of Transportation and Environmental Planning at Birmingham University, and was issued in May 1970, by what was then Marconi Elliott Computer Systems Limited at Borehamwood, complete with a manual sporting a front cover in MECSL house style.

In Autumn 2016, Elliott computer enthusiast Dr. Erik Baigar of Munich purchased a “Marconi Avionics Ltd 920ME computer” off eBay, made by Inertial Navigation Division of Rochester in around 1985. After some repairs (to address decoding and to the power supply), Erik was successful in getting it working, just before the Computer Conservation Society’s recent visit to Munich (see https://tinyurl.com/yadnl9m5). The ME is a 920 variant which I’d not previously heard of, so whilst I was in Munich, we ran some tests that established that it was a reimplementation of the late 1960s Elliott 920M, using later technology but running at the original speed, and with the 8K word core store replaced by 16K word battery-supported CMOS expandable to 32K. (Specifically, the 920ME does not support the additional instructions of the 920C or 905 which came slightly after the 920M).

Amongst the programs which I suggested Erik could demonstrate on his 920ME was Workshop. Very soon, using a program which simply tested the mathematical functions back-to-back, he’d found gross errors in the Arctan function (although it was OK for large arguments and around zero). We do not have the Workshop source code, but I was able to locate the string “ATN” in the binary file which led me to the implementation of the function and to the floating point interpreter.

The error was soon obvious. Workshop was using the series
x - x3/3 + x5/(3.5) - x7/(3.5.7) +...
which converges nicely but is wrong, rather than the correct
x - x3/3 + x5/5 - x7/7 +...
which takes ages to converge for x near 1.0. Luckily the code already performed the usual rotation to avoid x near 1.0, and the bug was fixed by simply swapping two instructions (to store the current power of x before, rather than after, dividing by the current odd number, so that the odd divisors do not accumulate).

I’m not sure whether, after 47 years, we can claim the prize for the slowest bug detection, ignoring those which will never be found. One bonus from this investigation was the discovery that Workshop uses a four-word software floating-point format interpreter, which could be a useful addition to the two-word and three-word “QF” formats used by 903 Algol.

SoftwareDavid Holdsworth

Kidsgrove Algol

We have demonstration material showing the compiler in action: kdf9.settle.dtdns.net/KDF9/kalgol/DavidHo/demo.zip.

However, we are not having much success getting the optimiser to work. We have been looking at the few dates that we find in comments and are beginning to doubt that this was a formally released version. After all, my experience is that one only did a big printout when one was baffled as to why things did not work.

David Huxtable has done a sterling job and found a few bugs in our version.

Most of the source code files are the result of Terry Froggat’s OCR, my dictionary of common misreads, and David Huxtable’s proof reading. A key problem area would appear to be brick 41. It is one of the larger bricks and almost devoid of comments. In the hope of eradicating any errors in the source text, I have done a verification by manually copy-typing. This somewhat laborious exercise revealed six previously undetected misreads which affect the binary code. There were also a some misreads in the few comments that the source contains. Sadly, it did not eradicate bugs, but I am embarking on a similar exercise for two other optimiser bricks, which fortunately are rather smaller.

We are planning on getting it working as well as we can, with the source text clearly annotated to show where we have altered the original.

We are actively discussing how best to make the material more widely available.

KDF9 Usercode

As reported earlier, the Kidsgrove compiler compiles into KDF9 Usercode, but we do not have the official brick that translated this into a binary program. We have co-opted the Paper Tape Usercode Compiler as a substitute.

Curiously a bug has been uncovered in this which surely cannot have got out into production, so perhaps we have an imperfect version here also.

Imperfections

It begins to look as though a printout that survives from the 1960s is not certain to be a final working version. This perhaps renders the true production versions inaccessible. Even our Whetstone Algol preservation turned out to contain a bug that Brian Wichmann was certain got fixed subsequently. It is not a reason to abandon attempts to bring surviving 1960s software into execution, but we feel that it is permissible and even desirable to attempt to make changes that would have happened in the field when the software went into production. This is in the hope of accurately reflecting the properties of the software as used in practice. Obviously, such changes should be annotated in the source text.

ICT/ICL 1900Delwyn Holroyd, Brian Spoor, Bill Gallagher

Since the last report our efforts have mainly been centred on the 1901A sub-project, the others either having reached a natural break point or stalled requiring additional information that we have not yet been able to find.

ICL 1904S

The 1904S emulator has been running reliably for some considerable time and we have scheduled January 2018 to review and tidy this emulator ready for release late January/early February, if all goes well.

ICL 1901A

The 1901A emulator is now running both E1HS and the fixed part of E1DS successfully.

We are still missing the 2nd paper tape for E1DS (the overlays), but have recreated some of the missing overlays, enough to be able to currently run and load programs that only use basic peripherals (cards, p/tape and printer).

We are aiming to add the magnetic tape and UDAS disc functionality, but do not have the information necessary to even think about implementing either Automatic Operator or GEORGE 1S.

If anybody has any information, listings etc. of Stevenage Executives — we would welcome it.

PF56

The PF56 emulator has benefitted from the availability of some hardware information regarding the 1901A’s console. Both machines use the same double-height card and the PF56 console now behaves as it should. However the PF56 main diagnostics are still not runnable.

The paper tape reader is still only working for DCP and locks up any diagnostic program used that accesses it.

Again, we would greatly appreciate any PF56 related CPU logics.

Bracknell Archive

We will soon be acquiring material from the former ICL site at Bracknell. This includes design documentation which should prove useful in future emulation efforts, as well as a quantity of magnetic tapes with unknown contents - watch this space!

ICL 2966Delwyn Holroyd

The band rollers from the line printer have returned from the rubber company having been recovered with neoprene rubber. They have now been re-installed in the printer and to our relief it appears to work satisfactorily. There is a small amount of play in the roller containing a bearing — we were warned that the bearing could be damaged in the autoclave, but it was not a straightforward job to remove it so we decided to take the risk. It doesn’t seem to be causing a problem at the moment but it will have an eye kept on it.

The OCP -5.2V problem returned again. This time a pushed back pin was found in a connector involved in the sense circuits — hopefully this is the true cause.

Harwell Dekatron/WITCHDelwyn Holroyd

A program has been developed to calculate and print the Fibonacci series for an event to mark the 5th anniversary of the reboot of the machine. The program consists of two paper tapes. The first tape loads initial variables into store and the second consists of the main calculation repeated three times to make a longer tape. The ends of the latter tape are glued together to form a program loop! To shorten printing time a simplified print format has been plugged up on the patch panel that was installed at Wolverhampton to allow print format customisation.

Argus 700/Bloodhound SimulatorPeter Harry

The Argus 700 has been repaired and the Bloodhound simulator is fully operational again, the fault being the main processor. The processor variant of our Argus 700 is the GX: a five card PCB set with functionality divided between the PCBs, arithmetic, control and a monitor interface. Resolving the fault gave us some difficulty as the symptoms pointed towards a problem with the I/O system and the PeriBus (I/O bus).

An obvious approach to resolving faults with a modular, PCB, constructed computer is to board swap until the fault disappears. Unfortunately, such a fault-finding process depends on a good stock of spares known to be in a working condition and that is the issue facing us in maintaining the Argus 700. Our ‘unknown’ board swap situation is compounded by the damp conditions the Argus was placed in for the many years it was out of use, i.e. edge connectors suffering from oxidisation. Being able to test and repair the Argus 700 PCBs is something we must do. But for this report here are two images to illustrate the start of our repair and test regime.

The first image is of a PeriBus simulator for the Argus 700, developed on a couple of prototyping boards. The cardboard box simply allows for transportation and prevents patch leads from accidently being pulled out! This simulator will be rebuilt once additional functionality is included in the design. Note the Arduino which is providing addressing and data scenarios.

peribus simulator
bench testing lash-up

The second image is a ‘lash up’ to enable limited bench testing of an Argus 700 GL processor. Note our Argus 700 has three processors, the GX and two GL co-processors. Being able to test a processor is a necessity and also a major challenge.

Over the winter months our testing facilities will be developed further and I am hoping to produce a more informative report early next year, subject to available time etc.

Our restoration work on the Bloodhound simulator, including the Argus 700, receives no funding from any external body. All funding is from personal donations made by the team involved. We do receive support in kind from third parties which does make up in part for the lack of financial support — hence the use of a cardboard box!

The 2018 Tony Sale Award for Computer Conservation

The purpose of the Tony Sale Award is to recognise a singular achievement in the area of computer conservation and restoration by a person or group, and to give public recognition to such work worldwide. The emphasis is on highlighting a significant practical project, which may involve hardware and/or software.

Nominations are invited by means of the Tony Sale Award 2018 Entry Form, with a deadline of 30th June 2018.

Assessment and Judging will take place during July/August, with the winner being publicly announced at a meeting of the Computer Conservation Society on the afternoon of Thursday 15th November 2018. The Judging Panel for the 2018 Award will be chaired by Professor Martin Campbell-Kelly.

Assessment and Judging Criteria

The following criteria will help the judges in their assessment of the projects:

Originality: To what extent does the project demonstrate a novel approach to conservation or reconstruction?
Completeness: Has the project achieved the initial goals set?
Ingenuity: What new techniques or processes were developed during the project?
Impact: What contribution has the work made to increasing the understanding of the history of computing?
Outreach: Is the result of the work visible to experts in the field and/or to the general public? To what extent has the work already been publicised or written up?

The Judging Panel reserves the right not to make an award.

The Award

The Award comprises a trophy and the sum of £1,000 to be put towards the winning project. The individual winner, or representative of the winning group, will be expected to attend the meeting on 15th November to accept the Award, and give a presentation on the project, lasting about an hour, and with a question-and-answer session, for which reasonable expenses for travel and accommodation will be paid.

Full details and the entry form can be found at www.sale-award.org/downloads/TheTonySaleAward2018entryform.pdf

Top Previous Next

News Round-Up


2018 marks the 70th anniversary of the first execution of a computer program (in the modern sense) by the “Manchester Baby” computer. 20 years on from the official launch of the Baby replica constructed by a team led by Chris Burton, it is proposed to lay on a modest celebration.

On June 21st just after 11:00 exactly 70 years on from that momentous event the first program will be re-run.

Nearer the time, keep an eye on the CCS website for more details as they emerge.

101010101

narrowboats

The photo of the LEO 3 Range Rover in Resurrection 79 has provoked David Holdsworth to send in a picture of narrowboats at Skipton with not one but TWO famous computer names. Any advance...?

101010101

The annual CCS excursion to faraway places is planned for April 2018 — specifically the weekend of 28th April. A trip to Pisa in Italy to visit a museum there in which we understand there is an extensive collection of computing machinery. No details as yet, but by the time this edition is distributed we hope to have more information on the CCS website.

101010101

Your editor has received several emails asking for payment of invoices issued by the Computer Conservation Society. Obvious scam. If you get one, please ignore it.

101010101

The British Society for the History of Mathematics and the Oxford University Mathematical Institute have announced a conference in March on the history of computing. See https://www.bshm.ac.uk/events/history-computing-beyond-computer for more detail.

Top Previous Next

CCS Chair’s Annual Report

David Morriss
David Morriss

I have thoroughly enjoyed my first year as Chair of the Society. I have been greatly assisted by the Committee, especially the Secretary, Roger Johnson and the past Chair, Rachel Burnett. I am amazed at the vast amount that is done by them and all our volunteers working in the projects. I can safely say that we continue to flourish.

Our membership remains strong, in excess of 1,500 and our finances are in good order thanks both to BCS support and the voluntary donations from members.

Projects

Our 15 projects continue, progress is regularly reported to the committee and can be followed in the pages of Resurrection, our journal.

Programme

Our meetings are held in between September and May, in London at the BCS office in Southampton Street, usually in the afternoon and in Manchester in the early evening. Topics covered this year in our normal meetings have included:

Crises and Lessons From the History of Software, Simulation Models of Historic Computers, Marian Rejewski’s Cyklometr, The Birth and Death of the Basic Language Machine, The Unknown Alan Turing, Computer Films, Babbage and Lovelace, Computing in Air Traffic Management, The Joe Lyons Story and Software as Heritage.

An outstanding event was the Making IT Work conference held in May. Delegates gathered at the BCS London office on the 22nd. May where a series of excellent papers were presented and then the delegates resumed the programme on the 23rd. May at the National Museum of Computing at Bletchley Park. The capacity audience included visitors from overseas.

Awards and Recognition

The 2016 Tony Sale award attracted entries an international field. At the November the 17th London CCS meeting Margaret Sale presented the award to the Heinz-Nixdorf Museum for its reconstruction showing how the 1946 American ENIAC Computer was programmed. This was followed by a most interesting presentation from Johannes Blobel who designed the exhibit together with Rainer Glaschich from Paderborn.

The other major innovation has been the inauguration of our Lifetime Achievement Award, the first recipients being Doron Swade and Chris Burton.

In addition, the society arranged for Hamish Carmichael, sadly now no longer with us, and Len Hewitt to receive the BCS Certificate of Appreciation.

Visits

In April a group of members made a tour of Computers and Microelectronics at the Deutches Museum in Munich.

Communications

Our journal Resurrection is issued every quarter and contains expert articles together with news on the CCS activities and projects. Printed copies are free to CCS members who are also BCS members and costs £10 per year to our other members. It is also on our website www.computerconservationsociety.org which provides very comprehensive information about all our activities and projects.

Committee

Our officers are the Chairman, Secretary and Hon. Treasurer, aided by the rest of the committee comprising the Chairman and Secretary of the North West Group, the Project Leaders, Meetings Secretary, Resurrection and Website Editor, Digital Archivist and Tony Sale Award co-ordinator. Other members focus on specific functions such as Media Officer, Archives Adviser, Membership Secretary and Projects Co-ordinator. A vital part of our committee is the representation from the museums; The Science Museum and the Museum of Science and Industry, The National Museum of Computing and the Bletchley Park Trust.

We are indebted to the providers of our BCS support, notably Mandy Bauer, the Finance team and those booking the facilities.

My thanks to the speakers, to our members and supporters, our volunteers and again to the committee for all their commitment and contributions.

Top Previous Next

GPM for a Challenge

P. A. Cherepanov
A macroprocessor is a program that performs text substitution and is usually used to extend programming languages. An early and influential macroprocessor, Starchey’s GPM, is ported to the C language and tested on GNU/Linux and MS Windows. Although GPM cannot compete with modern tools, it can be appreciated as yet another Turing tarpit that has a deep historic context. A GPM program and GPM interpreter can be packaged into a C file which can be submitted to automated program judging systems. The whole package is distributed from code.google.com/p/ggpm under GNU GPL.

Introduction

GPM (General Purpose Macrogenerator) was an early macroprocessor written by C. Strachey to facilitate the development of a CPL compiler. The detailed description of the GPM implementation made it a popular program, perhaps more popular than CPL itself. GPM was discussed in a number of textbooks, and adapted to rectify the deficiencies of Fortran and Cobol. Finally, GPM inspired the development of a whole line of macroprocessors that culminated in the advent of GNU m4. The latter is present on every standard GNU/Linux installation. Although m4 is more convenient to use, GPM presents the idea of text substitution in a purer form.

Strachey’s description of GPM syntax is still unsurpassed in clarity. This brief review of GPM syntax is included here for completeness.

The only operation supported by GPM is a macro call with the following syntax.

$NAME,arg_1,...,arg_n;

where NAME is the name of a macro and arg_1,...,arg_n are the arguments. A few predefined macros are available at the beginning. New macros are defined as

$DEF,name,<definition>;

where DEF is the name of a predefined macro that defines new macros. Angle brackets <...> are used to delay the evaluation of their enclosed string. A macro call forms a context, with macros defined within the macro call being local to that context. A macro definition can appear anywhere in a macro call. The latest definition with the same name shadows earlier definitions. This is the only technique to perform conditional evaluation.

Arithmetic operations in GPM are limited to addition, subtraction, multiplication, division, and remainder, and only on signed integer numbers.

Preservation status

GPM was widely popular in the 1970s, but it gradually fell into disuse. Although its CPL source is preserved and documented, a CPL compiler is no longer available. There are no working ports of GPM to any modern languages either.

The preservation of software should not be limited to the preservation of its original data. Such software often requires a contemporary environment and only runs on emulated hardware. This is likely to stop all but the most determined users. Ideally, historic software would be available to the user through the same installation techniques as contemporary programs. For instance, KEDIT, a clone of XEDIT, a memorable editor for the IBM CMS system, can be considered from a conservation point of view as a fully operational replica. Well preserved software is not only fully operational in modern environments, but also incorporates a significant part of the original code.

Porting the GPM

Strachey’s article provides a detailed description of a GPM implementation for the Titan computer, including its full source code in the CPL language. However, a CPL compiler is no longer available. Even the character set assumed by CPL and GPM is different from ASCII, which is now prevalent. Finally, modern command line programs are controlled by command line arguments and are expected to have separate output and error streams.

This port of GPM uses the dollar sign ‘$’ instead of the section symbol ‘§§’ as the leading symbol of a macro call. If necessary, a program written with the traditional symbols can easily be transliterated.

Porting the CPL source to C was fairly straightforward. The simultaneous assignments that were widely used in the CPL source were mostly reduced to a series of ordinary assignments. Introducing a temporary variable was necessary in a few cases. Computed goto operators were replaced by case operators without any major change in the program structure.

The parsing of command line parameters is based on GNU’s getopt_long(). A basic implementation of this function is bundled, for the benefit of an MS Windows build. The porting process took about a man-week, which matches existing estimations on the complexity of such a project.

A couple of errors have been encountered during the porting process. The statement in the CPL source code ST[P+7]=′-′→-WW, should be read as ST[P+7]=′-′→-W,W

Many authors have noted that the original definition of the Successor macro is incorrect.

$DEF,Successor,<$˜2,$DEF,˜2,˜1<,$Suc,>˜2<;>;

$DEF,9,<$Suc,>˜1,<;,0>;;>;

Andrew Herbert found that the error is caused by missing quotation marks. However, his version of Successor doesn’t work with the original GPM.

$DEF,Successor,<$˜2,$DEF,˜2,˜1<,>$Suc,>˜2<;>;

$DEF,9,<$Suc,>˜1<;<,>0>;>;

This port just adds an extra quotation layer around the unquoted commas.

$DEF,Successor,<$˜2,$DEF,˜2,˜1<<,>$Suc,>˜2<;>;

$DEF,9,<$Suc,>˜1<;<,>0>;;>;

Traditionally, macroprocessors mix input data and macro operations. This makes them unsuitable to for writing text utilities. This port of GPM solves this problem by adding GETC, GETW, and GETL macros, which read a character, a word, and a line from standard input, regardless of the source of the GPM program.

Programming techniques

The limited syntax of GPM forces the programmer to develop a number of idioms.

Comments and line breaks

GPM has no direct way to add comments to the source file. Comments can be programmed as a macro that takes one or more arguments and consumes them, returning an empty string.

$DEF,REM,;

$REM,<This is a comment.>;

The comment macro can also be used to split long lines in the source file.

Comments can be used $REM,

;to break long lines without affecting the output.

The line can also be split inside parameters that don’t directly contribute to the output. Extra parameters can be added to any macro call.

Quotation characters

The lack of a direct way to output unbalanced quotation symbols, ‘ >’, ‘ <’ is often mentioned as a major oversight in the original design. One of the extended versions, bgpm, has special built-in macros to output them. In fact, this can easily be done through numeral-to-binary conversion.

$DEF,lquote,<$VAL,a,$DEF,a,$BIN,60;;;>;

$DEF,rquote,<$VAL,a,$DEF,a,$BIN,62;;;>;

Conditions

Conditional evaluation based on the newest macro with the same name shadowing the earlier ones can only compare numerals for equality. More advanced predicates can be built from the $sign function; the algorithm can be expressed in Python as:

def sign(x):

return x if x in (0, 1, -1) else return sign(x//2)

Which corresponds to

$DEF,sign,<$˜1,$DEF,˜1,<$sign,$/,˜0,2,

;;>;$DEF,0,0;$DEF,1,1;$DEF,-1,-1;;>;

An arithmetic if operator can be constructed on top of $sign. For instance, $min and $max can be defined using the arithmetic if as:

$DEF,max,<$$sign,$-,˜1,˜2,

;;,˜1,˜2,$DEF,1,˜1;$DEF,0,˜1;$DEF,-1,˜2;;>;

$DEF,min,<$-,$+,˜1,˜2;,$max,˜1,˜2;;>;

For positive numbers, the $max_natural function can be calculated more efficiently.

def max_natural(a, b):

return (a//b*a + b//a*b) // (a//b + b//a)

Which corresponds to:

$DEF,max_natural,<$/,$+,$*,$/,˜1,˜2;,˜1,

;,$*,$/,˜2,˜1;,˜2;;,$+,$/,˜1,˜2;,$/,˜2,˜1;;;>;

Unlimited arguments

GPM is often criticised for restricting the number of arguments in a macro call. GPM uses a single character that follows ‘ ˜’ to identify arguments. It is convenient to name arguments ˜1 to ˜9; the following arguments are named by the ASCII characters that follow ‘9’ .

However, this argument naming convention is not necessary for GPM’s operation. Arguments can be passed as arbitrarily named macros, defined in the superfluous arguments of the function. This technique effectively lifts any limits on the number of arguments. Using this technique, the factorial function can be coded as:

$DEF,Fact,<$$arg;,$DEF,$arg;,<$*,˜0,$Fact,$DEF,arg,$-,˜0,1,

;;;;>;$DEF,0,<1>;;>;

$Fact,$DEF,arg,10;;

Usage

Software doesn’t grow old. This version of GPM is as capable as the original one, but much more usable software has appeared since 1965. The UNIX macroprocessor m4, the extant member of GPM’s lineage, is widely used in GNU configuration scripts. In every respect, it is preferable to GPM as a practical software tool.

However, GPM is simpler and thus illustrates the idea of calculation by textual substitution in the most direct way. It serves as a good example to introduce macrogeneration to students. The small number of primitive operations and limited syntax positions GPM among esoteric programming languages.

Writing non-trivial programs is a challenge. But people program in Intercal, BF, and Malbolge, and climb Everest just because they exist. The recreational value of GPM is comparable with that of other esoteric languages, with the additional aspect of being a real software tool.

GPM to C converter

Competitive and recreational programming often uses automated testing systems known as online judges. The challenge is considered solved if the program passes all the tests.

The sites based on competitive programming offer a limited set of popular programming languages. The sites focused on recreation may have a large selection of languages, but this port of GPM is too new to be included. Although GPM is now ported to C and distributed under the GNU GPL, most likely it will never regain its former popularity, nor the fame of BF or Intercal.

However, this shortcoming is easy to work around. The source code of the GPM interpreter is small enough to fall within the size restrictions of most sites. Thus the GPM program and interpreter can be bundled into a single file that can be submitted as a C program.

To facilitate this process, this project includes ggpm2c, which takes a GPM program and converts it into a self-contained C+GPM program.

Conclusions

Strachey’s GPM has been ported to the C language and extended to comply with the GNU guidelines. The new predefined macros that read standard input decouple the program and the data streams making it possible to write text handling programs in GPM. The source of the GPM interpreter and a GPM program can be bundled into a self-contained C file that is suitable for online testing. The project is distributed under the GNU GPL v.3 from https://code.google.com/archive/p/ggpm/ for everyone to share and enjoy.

Contact details

Readers wishing to contact the Editor may do so by email to
, or by post to 124 Stanley Road, Teddington, TW11 8TX.

Members who move house or change email address should notify Membership Secretary Dave Goodwin () of their new address or go to
www.computerconservationsociety.org/membership/membership_general.htm. Those who are also members of BCS, however, need only notify their change of address to BCS, separate notification to the CCS being unnecessary.

Queries about all other CCS matters should be addressed to the Secretary, Roger Johnson at , or by post to 9 Chipstead Park Close, Sevenoaks, TN13 2SJ.

North West Group contact details

Chairman  Tom Hinchliffe:  Tel: 01663 765040.
Email:  
Secretary  Gordon Adshead  Tel: 01625 549770.
Email:  

Top Previous Next

Smoke and Mirrors — The ICL 2905

Dik Leatherdale
In the days when ICL was headquartered in Putney, there were three buildings anybody crossing Putney Bridge couldn’t really miss. On the South (Putney) side stood ICT/ICL House formerly Electrolux House and before that, the site of a bus garage operated by the London General Omnibus Company. ICL House
Trevor Philpott introduces the ICT 1900 Series with ICT House in the background

ICL2905
The Eight Bells

On the north (Fulham) side were Bridge House South and Bridge House North. And behind these two buildings was the lunchtime watering hole The Eight Bells, often referred to as Bridge House East. For it was here that gossip was exchanged, brains were stormed and deals were done between ICL staffers who might not otherwise meet.

Many readers will be familiar with the story of the ICL 2903 launched in 1973. A dramatically cost-reduced version of the successful, but by then ageing, ICL 1900 it was firmly aimed at entry-level computer users. The revenues from nearly 3,000 sales kept ICL afloat during the difficult period of transition from 1900 to “New Range” (a.k.a. ICL 2900.1). In 1975 the introduction of the 2904 allowed a growth path for 2903 users whose workload had expanded (as is often the case).

By the early 1980s the second generation of the 2900 range included the 2950. The 2950, like the 290x, was a microprogrammed computer which was capable of supporting not only the 2900 order code but also 1900 programs, including those running on the 290x by using the DME-2 version of the microcode. In theory this gave 290x users a further growth path.

But theory isn’t always reflected in practice as readers will be all too aware. The mid to late 1970s had been a painful experience for first-generation 2900 users. In particular the VME operating system was, with some justice, considered complex and unreliable. By 1980, this had been triumphantly overcome but VME’s early reputation clung to the whole 2900 range like a bad smell. So it was that despite the evident need for more powerful 290x machines, sales of the 2950 running DME-2 were pretty dismal.

Enter at this point one Brian Cook, a consummate marketing man working in Bridge House South. It was Cook’s inspiration to rebrand the 2950 running DME-2 as the 2905. It was, as it turned, out a notion of genius. No hardware or software development was required so the costs were zero as was the timescale. Just as important was the avoidance of internal company politics which would certainly have delayed matters and might well have kicked the idea into touch altogether. It seems that Product Development Group were not even informed, much less consulted. Had they been, I have it on good authority that serious objections would have been raised. Extending the life of the 1900 order code was then against established company policy.

Not even the factory-fitted identification logos were changed — replacements were applied as part of the installation procedure.

Sales picked up immediately. 2905s, while not exactly flying off the shelves, sold in respectable numbers to 290x users who might otherwise have been lost to ICL entirely and who had steadfastly refused to entertain the prospect of owning a 2950. And this at a time when ICL’s fortunes were at a seriously low point.

The Eight Bells
An ICL 2905 (photo courtesy of TNMoC

So, was this idea conjured out of the air in The Eight Bells one lunchtime? It seems likely that it was. That was Brian Cook’s wont. Did he save the Company? That might be a bit of a stretch, but his cunning sleight of hand certainly contributed a great deal when it counted most.

Not long afterwards, a meeting involving ICL’s then new chief executive Robb Wilmot was seriously delayed when somebody, assuming Bridge House South would be in the south side of the river in Putney, spent ages failing to find it. Bridge House South was summarily renamed Bridge House.

Bridge Houses
Bridge House North and South Today

In the 1990s ICL moved out of Putney and Fulham altogether. A heavily rebuilt ICL House is now luxury flats, Bridge House South, an hotel and Bridge House North houses a travel agent. Only Bridge House East remains, echoing to the sound of ghostly chatter and the gentle clink of lunchtime glasses.

Thanks to the late Chris Cheetham, Steve Black, Rod Brown, Virgillio Pasquali and Brian Aldous.

1. The nomenclature of the 2900 range was (deliberately?) confusing. The low-numbered models (290x) were 1900 order code machines. Higher numbered systems (“proper” 2900s) were designed for the New Range order code, but the 2960 and all of the second-generation S-Series machines were capable of supporting either.

Top Previous Next

A Life in Computing

Alan Freke

I joined English Electric at Kidsgrove in October 1964 as a peripheral test engineer. I had no previous computer experience, but had just completed an electrical engineering apprenticeship with the Bristol Aeroplane Company where I had a slight acquaintance with their DEUCE.

I worked mainly on KDF9 peripherals, but also KDP10, KDF7, and KDN2. Later I worked on peripheral design and development for System 4 peripherals. I was in a very junior capacity there, but it taught me a lot.

In September of 1966 I saw an advertisement for a job at Bristol Siddley Engines in Bristol for someone to design a special purpose interface for their KDF9. I applied and got the job.

The plan was to link their KDF9 to a newly acquired Digital Equipment Corp. PDP-8 to speed up job entry. All input was via the paper tape reader on the KDF9, and the plan was to have the data prep done online using the PDP-8 with Teletype 33s, and then the PDP-8 scheduling the work and feeding it to the KDF9 instead of the paper tape reader.

Work had already started when I arrived, and the plan was to use an “standard” interface being promoted by Derek Barber at the NPL, but adapted to suit our purpose. In particular rather than 8 data bits we wanted to use 12 data bits in parallel on data transfers (PDP-8 12-bit word, KDF9 48-bit word, and KDF9 paper tape code 6-bit). Also NPL wanted to use +/- 12 volts for interface signals, we just wanted to use integrated circuit logic levels of +5 volts nominal and zero. In this we were supported by the folk at Cullam Labs. of the AEA who were also keen to use the interface. (Eventually, in its NPL form, it became a British Standard — but I don’t think that very many people used it)

The PDP-8/KDF9 project was successful, and I was then asked to link engine testbed data gathering equipment to the PDP-8, so that data could be transferred directly to the KDF9 for analysis. This was again done successfully.

A new ambitious plan was hatched to use a DEC PDP-10 (with a large amount of disc storage) as an online facility for programmers, and then link that to the KDF9 to run jobs. The system was known as AMOS, and presented several problems on the engineering level, as well as software problems. The PDP-10 had a 36-bit word length, thus confirming our 12-bit view of the world. However, we didn’t use 12 bits for the PDP-10—KDF9 link, but 144. (4 × 36-bit PDP-10 words to 3 × 48-bit KDF9 words)

This 144-bit ring buffer was filled and emptied by the respective machines on a Direct Memory Access basis allowing high speed data transfers. The programmers’ lives were transformed, as instead of two POST runs a day for debugging their code, they keyed in KDF9 code on their Teletype 33s attached to the PDP-10, then posted it into the KDF9 entry, and almost immediately had a one minute run slot on the KDF9, with results back to them via the PDP-10. If their job took more than the minute allowed the whole task was swapped out of the KDF9 to the PDP-10’s disks, and then put back in the queue at a lower priority, so it would take a lot longer.

The system was successful, and we acquired a second KDF9 from BAC at Filton to add more power. The need for more KDF9 power arose from the Rolls Royce take-over of Bristol Siddley. Their DP people confidently thought that they could convert all our advanced KDF9 software to run on IBM hardware (it was their preferred computer), but it was soon apparent that they couldn’t. So advanced was our application software that the development engineers at Derby wanted access to it, and so a special dedicated line was installed from Derby to Bristol to allow their engineers to access our system, and a second PDP-10 was added.

About this time DEC launched the PDP-11 which was 8-bit byte based, and which we viewed as an oddity. However, we did add many PDP-11s to the ever-growing network of computers linking testbeds, laboratories etc. The NPL-like interface was used for all sorts of peripherals, including high speed printers, plotters, paper tape equipment of all kinds, A-D converters etc. We even used a PDP-11 to control a motorised milling machine to test out an idea for a “touch probe”. This was an idea of a mechanical engineer who wanted to computerise inspection of components. Eventually Rolls Royce dropped the project, so the engineer, Dave McMurtry, left taking one of my staff. He later called his company Renishaw.

In 1975 the final addition to the AMOS system was a CDC Cyber74. It had a 60-bit word length, thus confirmed our 12-bit view of the world — how wrong we were!! The Cyber74 needed no peripherals, but CDC insisted on a tape drive and printer, or their engineers couldn’t run diagnostics! Again, CDC programmers worked on the Teletypes — keying Cyber programs, which were posted across to the Cyber for running, and results returned to the PDP-10s in the same manner as with the KDF9.

DEC produced a nice colour brochure about the system, as they saw their PDP-10s as the heart of it. I wish I’d kept a copy!

In 1977 I was offered a job with a small Canadian company called Geac, and I left what was then Rolls Royce, thus ending my involvement with the KDF9.

Top Previous Next

Obituary : Geoff Tootill

Martin Campbell-Kelly
geoff Tootill
Tom Kilburn & Geoff Tootill in 1998.

On 21st June 1998 Manchester University and the City of Manchester celebrated the 50th anniversary of the “Baby” — the world’s first electronic stored-program computer. For the occasion, a team of volunteer engineers led by Chris Burton had built a working replica of the original computer. The replica was inaugurated by Tom Kilburn and Geoff Tootill, its two surviving creators. The replica is now a permanent exhibit in Manchester’s Museum of Science and Industry.

Geoff Tootill was born in Chadderton, Lancs, 1922, but grew up in Birmingham where his father was midlands manager for the Co-operative Press. He was educated at King Edward’s High School, Birmingham, and won a scholarship to read mathematics at Cambridge University in 1940. Studies at Cambridge were intense and it was at this point he realised his vocation was not in mathematics. He sat in on some engineering lectures and realised that was where his true interests lay.

After graduation he was directed to work of national importance, initially as a mathematician in operations research, but after a few weeks he managed to switch to an engineering role at TRE, Malvern, installing and trouble-shooting airborne radar systems.

In another laboratory at TRE, F. C. Williams and his assistant Tom Kilburn were working on a CRT-based electronic data storage system intended for use in radar. When the war was over, Williams was appointed to the chair of electro-technics at Manchester University, and he brought in Kilburn and Tootill to work on a computer memory project.

In order to test the memory when it was constructed, Kilburn and Tootill designed an elementary computer with just 32 words of memory, officially known as the Small-Scale Experimental Machine, but better known as the “Baby.” The machine first worked on Monday 21st June 1948, taking 52 minutes to find the highest proper factor of 218, involving about 3.5 million arithmetic operations.

In August 1949, Tootill transferred to Ferranti, Moston Manchester, to specify a full-scale computer based on the Manchester University design. The first Ferranti Mark I was delivered to the University in February 1951, the world’s first commercially-available computer.

Next, Tootill took up a senior lectureship at the Military College of Science, Shrivenham. The college offered no opportunity for research, so he leapt at the offer of a research position at the Royal Aircraft Establishment, which came from Stuart Hollingdale, head of the mathematics division.

Hollingdale and Tootill

He and Hollingdale collaborated on a book for the layperson “Electronic Computers”, published in 1965 as a Pelican Original. In an era when the public were becoming aware of computers, but accessible books on the subject were rare, it performed an important rôle. In the late 1960s it seemed as if a copy could be found in almost every UK public library.

From 1963-1969 Tootill was seconded to the European Space Research Organisation, where he established a network of computerised ground stations. Returning from this project, he spent four years in a bureaucratic role (which he detested) before escaping to the NPL in 1973 to work on the European Informatics Network.

He retired in 1982 but remained active in computing and academic pursuits. He died from pneumonia at the age of 95 following a short illness.

Top Previous Next

Out of Oslo

Peter Short
The IBM Hursley Museum was fortunate to receive a large donation of historical artefacts from IBM Norway this year. This is the story of how that came about and of some of the unique artefacts that were included.

Within IBM Europe there is a department, RESO (Real Estate Strategy & Operations), that organises and coordinates office moves. Hursley’s curators first learned about this department about three years ago, and the contact proved very fruitful. At the time, the last IBM office in Basingstoke was being cleared, and we were able to access the building to rescue anything of interest. Shortly afterwards we were able to raid the Weybridge offices when they too closed.

Hardware from Basingstoke and Weybridge included several different iterations of the AS/400, including a rack containing three tape library machines (a 9309 and two 3590s). We also had some RS/6000 and PC / PS2 boxes, printers, miscellaneous cables and adapter cards, a couple of racks together with eServer & RS/6000 rack mounted units. There was also a Sequent-branded rack containing two servers, now on display in the museum.

In June 2017 we received an email from RESO to say that they were relocating IBM Norway’s Oslo headquarters later this year. There were several artefacts that they would not be able to accommodate. Would we be interested in acquiring these for Hursley?

The voluntary museum curators have a strong ethic for preservation, but are also acutely aware of the space limitations in Hursley. Curator Peter Short made contact with Oslo to find out what was on offer. The reply came from an IBM Norway retiree, Kjell Markhus, who had been tasked with assisting with the clear out. He had visited the storage area and sent a Dropbox link showing a number of photographs of the contents. Apparently these machines had been there since 1985, following celebrations for IBM Norway’s 50th anniversary in that year. We agreed that the best approach would be to set up a Skype video call to go over the contents of the room in detail and decide what artefacts we would like to be sent over.

632 tabulator

As you can see from the sample of photographs, there was a huge jumble of old equipment, much of it immediately recognisable, but with some not so obvious. On the left we can see a 632 tabulator looking a bit worse for wear, the accompanying typewriter and keypad, and an 011 electric hand punch.

On the right, a Copier 1, 029 and 5496 card punch machines, a dial recorder, master clock, 557 Interpreter, 550 Interpreter and 416 Tabulator.

copier, 029 & card punch
PC/PS2 PC/PS2/Typewriter 011

Here on the left is mainly PC / PS2, but of great interest were the S/370 front panel and the unusual-looking monitor with diskette drives. The centre photo is a jumble of PC, PS2 and typewriter hardware, which ended up as EE waste, as the picture shows, and on the right the 011 in more detail. As you can see from these pictures, which only show part of the room, there was an awful lot of old kit to look through.

A couple of Skype calls later and we had a list of what we would like to add to the Hursley collection. Much of what was on offer we already have in the museum, so we were easily able to weed those items out. Kjell then proceeded to photograph everything we had selected and created a spreadsheet with full details for final confirmation. There then followed the inevitable problem of finding funding for the shipment.

By mid-August IBM Norway was able to confirm they would fund the transport costs as long as we could provide the manpower to unload at our end. This we agreed, and in September the transport company collected the shipment, which finally arrived during October. We were able to get the lorry reasonably close to the House to keep the moving distance to a minimum.

When the lorry arrived, we found that the trailer was not full, but would have been fuller if everything had been shipped! It turned out that quite a few artefacts are still in Oslo, and will need to be shipped at a later date.

chaos damage

It became immediately obvious that the load had not been tied down properly. On the left is the chaos caused as heavy hardware up to the hundreds of kilos moved around and tipped over. Sadly, this caused quite a bit of damage, although we are hopeful we can fix most of it. Hursley’s on site contractor unloaded the lorry with a fork-lift and the photo on the right shows everything unloaded in front of the stable block and attracting attention from passers-by.

temporary home temporary home

Once everything had been unloaded, we three curators were left to move everything into the museum ourselves, as it was Friday afternoon and there was nobody else available. By late afternoon we were somewhat knackered (understatement) and had to admit defeat before everything had been moved. We managed to seek some assistance to move the remaining machines into the stable block archway until after the weekend, as you can see in these two photos. On the Monday the Hursley contractor helped us to get the remaining hardware inside.

416 tabulator 416 tabulator again

The 416 tabulator gave us an interesting problem. Not only had it lost a wheel, but it was also too wide to get through the west door of Hursley House. We had to remove anything sticking out the front, and it was wheeled in with millimetres to spare. Rolling a three-wheeled dead weight of several hundred kilos did provide some interesting moments too!

reproducing punch numeric interpreter

Most of the bigger items were put straight out on display, with some dynamic planning centred on what would fit through which doors. Wooden blocks replaced missing legs and wheels for now. On the left is the 519 Document Originating Punch, aka Reproducing Punch, and the 557 Alphabetic Interpreter. On the right, the 550 Numeric Interpreter sits next to the 091 Hollerith Interpreter. The latter came from the Lyngby offices in Denmark a while ago.

029 key punch System 3 punch

The 029 card punch was somewhat legless, so had to be repaired before putting out on display next to the 5496. The 5496 was announced with System/3, and was a combined punch and verifier for the small format 96 column punch cards. This was probably the last ever iteration of the punch card before it became obsolete. The remaining items were stored in the back office, and many have subsequently also gone out on display. Our oldest artefact is now an original Hollerith pantograph card punch, which is dated 1889.

C R T
‘     T     ’ ‘C         R’

In particular the ‘Origins of IBM’ story has now been rearranged into two rooms. Before being renamed IBM in 1924, the company was called CTR, the Computing, Tabulating and Recording Company. In one room there is punch card equipment (the ‘T’ of CTR) and in the other the computing scales (‘C’) and clocks and time recorders (‘R’). We did receive many time recorders, which were built in the manufacturing plant in Norway until the late 50s. One, not yet on display, is brand new in its original packaging.

master clocks collator

Amongst the items not out on display are two IBM master clocks, the 077 collator and dial recorder. One of the clocks is in excellent condition, and we are hoping to get it back to full working order.

024 card punch 370/155

The 024 card punch may be part of the 632 tabulator, but we haven’t yet checked it to see if it has an I/O interface. If not then it will be a stand-alone device. Annoyingly it is missing the table top, so before it can be displayed it will need the same treatment as the 024 and 056 on loan from the Army, a home-made chipboard table top.

The 370 panel is from a model 155. The 370-135 was developed in Hursley, so we may have to pretend it’s one of those when it goes out on display in the Hursley room. Other hardware stored in the back office include a DisplayWriter, a 5110 with tape drive and 8” diskette unit, a plasma display, some prototype ThinkPad keyboard-less tablets, and a PC-AT and dictating machine with Perspex covers.

We now have our work cut out for the foreseeable future. Apart from other museum reorganisation projects, we will be looking at which of these machines can be brought back to life. Interestingly, the older machines look in much better condition than the not so old.

There’s a slide show with more photos of the shipment arriving on the web site at: hursley.slx-online.biz/pictures.asp?pictures=pics-movies/galleries/oslo/oslo.html.

Top Previous Next

Yet Another Differential Analyser

From The Meccano Magazine — January 1951

A Meccano Calculating Machine

Solving Complex Mathematical Equations

A WONDERFUL Meccano model constructed by Professor D. R. Hartree, then of Manchester University, was described and illustrated in the “M.M.” for June 1934. The purpose of this was to solve complicated mathematical equations, a task that it performed accurately and far more quickly than was possible by ordinary calculations. Its original, to which the name differential analyser was given, had been designed and constructed by Professor V. Bush at the Massachusetts Institute of Technology, in the

Meccanno D.A.

United States After seeing photographs of this machine Professor Hartree had the impression that someone had been enjoying himself with a large Meccano Set. This suggested building a differential analyser of Meccano Parts, and the demonstration model that he constructed actually proved of use in his own research work. Other successful Meccano reproductions of the machine followed, and on this page we illustrate one that has been constructed at the new University of Malaya, Singapore, by Professor J. C. Cooke.

This remarkable machine can add, subtract and multiply, and in addition it carries out a more complicated process known as integration. It works with a series of shafts turning at various speeds. These shafts, built up of Meccano Axle Rods, can be seen on the left of the photograph of Professor Cooke’s model. For adding two terms in an equation shafts are made to rotate at speeds corresponding to them, and by means of differential gearing these turn a third shaft at a rate corresponding to the sum. Multiplication is effected by simple gearing in a similar manner.

For integration, which is the central purpose of the machine, the round tables seen on the right are used. These are made of bakelite. The mechanism of each table is connected by cross shafts to the appropriate shafts on the left, the drives being given by bevel or helical gearing. One of the cross shafts rotates the table, and a second acting through a Screwed Rod moves it towards or away from the rest of the machine. Resting on the table is a Bush Wheel and the shaft of this is connected to a third shaft on the left. As the table rotates it turns the Bush Wheel, at a low speed when this is near the centre of the table and at a higher speed when it is nearer the circumference. This variation in speed enables the device to accomplish the type of summation that mathematicians call integration, and the result is transmitted to one of the shafts on the left.

The results of all these operations are represented by the rates at which the final shafts in the model turn. In some cases the turning of these shafts is made to draw a curve that represents the “answer.” In others revolution counters are turned to give direct numerical result.

Professor Cooke’s model is not just a toy, or even a demonstration model. It is a mathematical calculating machine capable of serious work, and its success emphasises the judgment of Professor Hartree, that such machines are a striking tribute to the practical value of the Meccano system.

CCS Website Information

The Society has its own website, which is located at www.computerconservationsociety.org. It contains news items, details of forthcoming events, and also electronic copies of all past issues of Resurrection, in both HTML and PDF formats, which can be downloaded for printing. At www.computerconservationsociety.org/software/software-index.htm, can be found emulators for historic machines together with associated software and related documents all of which may be downloaded.

Top Previous Next

Letter to the Editor

Mike Robinson

Never Believe the Publicity

Dear Dik,

A smile came to my face when I read the news byte “ICT move into Giant Machine Market” from Computer Weekly 50 Years Ago in Resurrection 79.

Firstly “Giant” was one instruction per microsecond and 64KW of main store. How times have changed!

More importantly, ICT were exaggerating. A prototype 1906A was not “in operation” in the Autumn of 1967.

I returned from installing a 1906 in RAE Farnborough in September 1967 to work with Archie MacBeth developing the cooling and power control systems of the First Prototype 1906A. There were no integrated circuits in the machine. In fact there were no platters in place in the doors, only blanks to keep the air in.

The first 12 layer platter with incorrect artwork came across the yard from the Multi-layer department to the labs in 1968.

I still have it. Unfortunately it has no script etched into a layer.

Request for Information

I started writing a SIMH emulation of MU5 last Christmas and I have made good progress on the CPU. I have had a lot of assistance from Prof. Ibbett and others, but I would still like to locate further MU5 material that would help to make a better emulation and to get it actually running some software. The holy grail would be to get a copy of the operating system, MUSS. There may be a copy of it on an ICL EDS8 disk, but currently there is no way to read it. I wonder if there is anyone who has any listings, paper tape, magnetic media, anything really, with software for MU5? Other manuals could also be of interest, although I have the Basic Programming Manual and can get access to most of the Hardware Manual should I need it. There is a missing chapter of the Hardware Manual that talks about the console, if anyone has that chapter it would be great to get a copy of it.

Information on the project at robs-old-computers.com/projects/mu5/.

Rob Jarrett ()

Top Previous Next

50 Years ago .... From the Pages of Computer Weekly

Brian Aldous

GUS Plans On-Line Mail Order Service: The veil of secrecy over the plan of Great Universal Stores to computerise their mail order and other businesses is gradually cracking. The latest step has been to order a Honeywell 1200 system as an interim measure until the installation of the planned English Electric System 4/50. (CW64 p1)

English Electric/E-A link-up activities: : Another step in the integration of English Electric and Elliott Automation has been taken with the linking up of the computer activities of the two companies. (CW64 p2)

Cybernetic Model Built by Russians: : The Russians claim to have built a cybernetic model of purposive behaviour which exhibits the human personality attributes of instincts for self-preservation, reproduction and social behaviour. (CW64 p8)

DEC’s Rapid Growth as PDP-8i is Introduced: : Digital Equipment Corporation has recently announced the integrated circuit version of their PDP-8 range of 12-bit-word computers which will be available in this country at a price of £6,200 for the rack-mounted version. At about the same time the American journal EDP Industry Report published an issue devoted to an examination of DEC which is today’s leading supplier of smaller computers. (CW65 p7)

CAD Program Speeds Engineering Projects: : A computer program which, it is claimed, could have saved this country £500,000 in design drawing operations in 1966, has been introduced by the international engineering contractors, Humphreys and Glasgow of Victoria, London. The program enables isometric projections of piping layouts to be automatically produced by a computer, and a variant of the program is now available for use on electrical cable sizing problems. (CW65 p16)

360/30 Real Time System On-Line: : A real-time order control system using an IBM 360/30 and seven display terminals goes on-line on January 1 at the Worcester offices of Kay and Co, a large mail order company within the Great Universal Stores group. (CW66/67 p16)

NRDC Support for Mini Machine: : A new computer, the Micro 16, which meets the Ministry of Technology’s Type C specification for small computers, has been launched by Digico Ltd, of Welwyn Garden City, Herts. Development of the new machine has received support from the NRDC and it competes directly with the DEC PDP-8S for the scientific market. (CW68 p1)

EAL-Honeywell Hybrid for MRC Project: : The first hybrid computer for use in medical research in the UK is to be installed by the Medical Research Council at the Mill Hill, North London, centre of the National Institute of Medical Research. The system will be used to extend work on biological models. The analogue side will use the centre’s existing EAL TR-48 machine, and digital facilities will be provided by a Honeywell DDP 516 with an interface system jointly devised by the MRC and Honeywell. (CW69 p1)

1900A Range Leads Bid for New Markets: : Deliveries at the rate of five a week by the end of 1968 are forecast for the ICT 1901A, the smallest of the new fully compatible 1900A range of computers announced this week. (CW70 p1)

360/91 on Space Project: : The fastest and most powerful IBM computer ever delivered to an outside user, a 360/91, is now fully operational on research problems at the US National Aeronautics and Space Administration’s Goddard Space Flight Centre, near Washington DC. It is being used in a project designed to increase the knowledge of the earth’s magnetic field, high and low energy particles, ionosphere meteorology, micro-meteorites, solar physics and astronomy. (CW70 p1)

UK Small Machine to Meet Growing Demand: : The first deliveries of Modular One, the compact computer designed and manufactured by Computer Technology Ltd at Hemel Hempstead, are scheduled for May. Currently the company has between £150,000 and £250,000 of work in hand and expects to deliver 30 systems this year. (CW70 p1)

Plessey’s N/C Orders: : With orders totalling £250,000 as a result of the company’s participation in the recent Machine Tool Exhibition at Hanover, a numerical control success story has come from the Plessey Automation Group. They have received orders for 26 N/C systems for customers in the UK, France and Germany. Most of the units sold under these contracts are of the Plessey BR3100 type — a basic contouring design with a wide range of options making a very flexible and efficient system. (CW71 p5)

Remote Access Program for KDF9: : In anticipation of the increasing demand for large scale remote access systems based on a central computer, the Operations Research Department of English Electric Computers has developed a program for the Bureau Division’s KDF9 at Kidsgrove which calculates the optimum methods of installing such a system. (CW71 p15)

360/85 Added to IBM Range: : The most powerful machine to be made generally available in the 360 range, the Model 85, has now been announced by IBM. The 360/85, which is likely to be of most interest for scientific applications, is capable of 12.5 million operations a second and can multiply two 10-digit numbers two million times in one second. (CW72 p1)

Sheffield gets Third University 1907: : The last of the three ICT 1907 computers to be ordered by a British university is now being installed at Sheffield, and should be operational by the late spring or early summer. The others are already installed at the Queen’s University, Belfast, and at Southampton University. (CW72 p1)

Devaluation Swings Order to System 4: : Devaluation has brought a windfall to English Electric Computers in the way of a £102,000 order for a System 4/30 in place of an IBM 360/30 which gasket and jointing compound makers Richard Klinger has had on order since December, 1966. The new contract has followed the price increases arising out of devaluation announced by IBM last month. The total saving amounts to about £20,000, roughly half of which is due to the change in the £. (CW73 p1)

Classroom of Tomorrow: : Two recent events foreshadowing the rapid advance of computer aided instruction in US schools is the expansion of the Stanford-Kentucky Teletype network and the RCA demonstration at the HemisFair ’68 International Exposition in Texas, which features an electronic system for the “classroom of tomorrow”. Thirty new Teletype machines went into operation at the Breckenridge School of Morehead State College, Kentucky, providing students with more places for the 15-minute practice sessions at these terminals. (CW73 p7)

Real time plan of Mirror Group Company: : In the six months since its establishment, International Data Highways, the Mirror Group’s subsidiary formed to take advantage of computers, has had time to clarify its aims. Large scale recruitment which is now starting indicates that the go ahead has been given and it is expected that computer staff will double this year. In addition to its four existing projects, Computaprint, Scan, Scan Plan 2, and Computer Data Processing, the company is now ready to tackle four further major projects, but is keeping details of them under guard at the moment. (CW73 p16)

Multi-access Plan at Hospital: : The first hospital multi-access system in the UK is to be based on one of two ICT computers that have been approved by the Ministry of Health for installation at London hospitals. The multi-access, using a 1905E, will be at King’s College Hospital and will be used to process the day-to-day activities of the hospital on a real time basis. Among the first applications will be outpatient appointments, inpatient admissions, bed allocation, nurse allocation, drug and diet control, and the analysis of laboratory and other tests. (CW74 p1)

System 4 Boosted by 4/40: : The position of English Electric Computers has been strengthened this week by the addition to the System 4 range of the Model 40 which fits into the family between the 4/30 and 4/50, and is directly competitive with the new ICT 1903A. This is the first addition to System 4 since the introduction of the 4/75 multi-access machine. (CW75 p1)

LEO 326 to Optimise Mail Vans: : A scheme to optimise the use of the 600 mail vans operated in the Central London area has been developed by the GPO, in conjunction with English Electric Computers, using one of the GPO’s seven LEO 326 computers. The fleet performs some 30,000 time critical trips a week, mainly within a five mile radius of King’s Cross and mainly against regular commitments. (CW75 p7)

GPO Extends Service with Datel 300: : The announcement of the Datel 300 service has taken the GPO into the business of marketing a computer peripheral. It is, however, according to Mr Geoffrey Dale, of the GPO’s telecommunications HQ, an exceptional case and in no way indicates a change of policy, which has been and continues to be, to supply a communications service, while industry supplies the peripheral hardware. Datel 300 is a service for the collection of industrial data, and it is expected that users will have their terminal equipment at several locations over a wide area. In these circumstances the decision was taken to supply the necessary card readers, tape readers and keyboards in addition to the modems, thus enabling the GPO maintenance service to be used. Customers with suitable peripherals, however, will be able to hire the modulators alone. (CW76 p1)

Memorex Enter Disc Market: : Closely following the prediction made by several computer manufacturers, particularly in the US, that disc stores will have an increasingly important role to play in data processing over the next three to five years, Memorex, best known for their high precision magnetic tapes, have announced their entry in to the disc market. Until recently the only disc pack available with full third generation compatibility has been that manufactured by IBM, principally for their own 1311 and 2311 disc drives. (CW76 p16)

Top Previous Next

Forthcoming Events


London Seminar Programme

18th Jan 2018 Re-Imagining the PDP8 Oscar Vermuelen
15th Feb 2018 What Really Happened in Y2K Martyn Thomas
15th Mar 2018 Founding of the Numerical Algorithms Group (NAG),
its Early Days and its Rôle Today
Provisional listing. Please see the CCS website to confirm.
Brian Ford
and colleagues

London meetings take place at the BCS in Southampton Street, Covent Garden starting at 14:30. Southampton Street is immediately south of (downhill from) Covent Garden market. The door can be found under an ornate Victorian clock.

You are strongly advised to use the BCS event booking service to reserve a place at CCS London seminars. Web links can be found at www.computerconservationsociety.org/lecture.htm.

For queries about London meetings please contact Roger Johnson at .

Manchester Seminar Programme

16th Jan 2018 KDF8 A Retrospective Neil Macphail
20th Feb 2018 EMIDEC and its Software Dave Parsons
20th Mar 2018 Restoring the 2988 Delwyn Holroyd

North West Group meetings take place in the conference room of the Royal Northern College of Music, Booth St East M13 9RD: 17:00 for 17:30.

For queries about Manchester meetings please contact Gordon Adshead at .

Details are subject to change. Members wishing to attend any meeting are advised to check the events page on the Society website at www.computerconservationsociety.org/lecture.htm.

Museums

MSI : Demonstrations of the replica Small-Scale Experimental Machine at the Museum of Science and Industry in Manchester are run every Tuesday, Wednesday, Thursday and Sunday between 12:00 and 14:00. Admission is free. See www.msimanchester.org.uk for more details

Bletchley Park : daily. Exhibition of wartime code-breaking equipment and procedures, including the replica Bombe, plus tours of the wartime buildings. Go to www.bletchleypark.org.uk to check details of times, admission charges and special events.

The National Museum of Computing : Thursday, Saturday and Sunday from 13:00. Situated within Bletchley Park, the Museum covers the development of computing from the “rebuilt” Colossus codebreaking machine via the Harwell Dekatron (the world’s oldest working computer) to the present day. From ICL mainframes to hand-held computers. Note that there is a separate admission charge to TNMoC which is either standalone or can be combined with the charge for Bletchley Park. See www.tnmoc.org for more details.

Science Museum : There is an excellent display of computing and mathematics machines on the second floor. The new Information Age gallery explores “Six Networks which Changed the World” and includes a CDC 6600 computer and its Russian equivalent, the BESM-6 as well as Pilot ACE, arguably the world’s third oldest surviving computer.

The new Mathematics Gallery has the Elliott 401 and the Julius Totaliser, both of which were the subject of CCS projects in years past, and much else besides.

Other galleries include displays of ICT card-sorters and Cray supercomputers. Admission is free. See www.sciencemuseum.org.uk for more details.

Other Museums : At www.computerconservationsociety.org/museums.htm can be found brief descriptions of various UK computing museums which may be of interest to members.

Top Previous Next

Committee of the Society


Chair  David Morriss FBCS CEng CITP:
Secretary  Dr Roger Johnson FBCS:
Treasurer  Dr David Hartley FBCS CEng CITP:
Chairman, North West Group   Prof. Tom Hinchliffe FBCS FIEE CEng:
Secretary, North West Group  Gordon Adshead MBCS:
Editor, Resurrection  Dik Leatherdale MBCS:
Website Editor  Dik Leatherdale MBCS:
Meetings Secretary  Kevin Murrell FBCS:
Membership Secretary  Dave Goodwin MBCS:
Media Officer  Dan Hayton MBCS FRSA:
Digital Archivist  Prof. Simon Lavington FBCS FIEE CEng:
Awards Sub-Committee Co-ordinator:  Peta Walmisley:

Awards Sub-Committee

Rachel Burnett, Roger Johnson, David Hartley

Museum Representatives
Bletchley Park Trust  Peronel Craddock:
National Museum of Computing  Dr Andrew Herbert OBE FREng FBCS:

Project Leaders
SSEM  Chris Burton CEng FIEE FBCS:
Bombe  John Harper Hon FBCS CEng MIEE:
Elliott 8/900 Series  Terry Froggatt CEng MBCS:
Software Conservation  Dr Dave Holdsworth CEng Hon FBCS:
Elliott 401 & ICT 1301  Rod Brown:
Harwell Dekatron Computer  Delwyn Holroyd:
Our Computer Heritage  Prof. Simon Lavington FBCS FIEE CEng:
ICL 2966/ICL 1900  Delwyn Holroyd:
Analytical Engine  Dr Doron Swade MBE FBCS:
EDSAC  Dr Andrew Herbert OBE FREng FBCS:
Bloodhound Missile/Argus  Peter Harry:
IBM Hursley Museum  Peter Short MBCS:

Co-opted Members
      Prof. Martin Campbell-Kelly FBCS CITP FLSW:
Past Chair  Rachel Burnett FBCS CITP Hon D. Tech:
Top Previous


science museum logo
MSI logo
TNMoC logo

Computer Conservation Society

Aims and objectives

The Computer Conservation Society (CCS) is a co-operative venture between BCS, The Chartered Institute for IT; the Science Museum of London; and the Museum of Science and Industry (MSI) in Manchester.

The CCS was constituted in September 1989 as a Specialist Group of the British Computer Society (BCS). It thus is covered by the Royal Charter and charitable status of BCS.

The aims of the CCS are to

  • To promote the conservation of historic computers and to identify existing computers which may need to be archived in the future
  • To develop awareness of the importance of historic computers
  • To develop expertise in the conservation and restoration of historic computers,
  • To represent the interests of Computer Conservation Society members with other bodies,
  • To promote the study of historic computers, their use and the history of the computer industry,
  • To publish information of relevance to these objectives for the information of Computer Conservation Society members and the wider public.

Membership is open to anyone interested in computer conservation and the history of computing.

The CCS is funded and supported by voluntary subscriptions from members, a grant from BCS, fees from corporate membership, donations, and by the free use of the facilities of our founding museums. Some charges may be made for publications and attendance at seminars and conferences.

There are a number of active Projects on specific computer restorations and early computer technologies and software. Younger people are especially encouraged to take part in order to achieve skills transfer.

The CCS also enjoys a close relationship with the National Museum of Computing.


Resurrection is the journal of the Computer Conservation Society.
Editor – Dik Leatherdale
Printed by – BCS, The Chartered Institute for IT
© Computer Conservation Society