12

Gothenburg Universities Computer Centre (in Sweden) developed a timesharing system for IBM mainframes, known as GUTS (variously expanded either as ''Gothenburg University Timesharing System'', or as ''Gothenburg Universities Terminal System''). It appears some other universities adopted it; one source mentions its use at University College London in the 1970s; another mentions its use at Trinity College Dublin, on an IBM 360/44 which was the first timesharing system in Ireland, as a replacement for IBM RAX and ITF (Interactive Terminal Facility, a predecessor to TSO).

Westgard and Groth's 1981 paper, "Design and evaluation of statistical control procedures: applications of a computer 'quality control simulator' program" describes chemistry simulation software written in "FORTRAN IV for an IBM 370/158 computer run under OS/MVT and the Gothenburg University Terminal System", providing a citation to "Gothenburg University terminal system. TS-GUTS reference manual. Gothenburg University Data Center, Gothenburg, Sweden, 1980".

From what I understand, rather than an operating system in its own right, it was a timesharing subsystem for MVS (and its predecessors such as OS/360 MVT), an alternative to TSO.

The Kermit Software Source Code Archive preserves "Kermit/GUTS", "Kermit for the Gothenburg University Timesharing System (GUTS)", version 1.0, dated 1985-04-05. It is noted as being for the IBM 370, written in BAL (IBM Basic Assembly Language), and the OS name is given as "MVS/GUTS". It preserves the source code and some associated documentation and emails. One particularly interesting file it preserves, is what appears to be an example of its procedure language. The help message embedded in that procedure contains an interesting tidbit about the nature of GUTS, that it had two different types of files, OS and GUTS:

 %WRLINE            KERMIT-GUTS (which is the name of the GUTS version  00032290
 %WRLINE            of KERMIT) can only send and receive OS-files.      00032300
 %WRLINE            If you want to send a GUTS file you will have to    00032310
 %WRLINE            copy if first to a OS-file. You can do that with    00032320
 %WRLINE            the /TSOS procedure. You can use the /OSTS procedure00032330
 %WRLINE            to copy a OS file to a GUTS file.                   00032340

Another source I've stumbled across, in the IAEA International Nuclear Information System (INIS), is a 1986 progress report (catalog entry, PDF on the "NAMMU-HYPAC" system for nuclear waste hydrology, then under development at Stockholm University Computer Center. Appendix 6 (page 34 onwards) also provides some interesting information on GUTS:

HYPAC... is implemented on a MVS operating system together with GUTS (Gothenburg Universities Terminal System).

GUTS provides among other things a text editor, a file system (the files stored there are called GUTS-files as opposed to ordinary MVS-files which are called OS-files ), a possibility to submit jobs to MVS and to retrieve job output to GUTS files, tools for conversion of GUTS-files to OS-files and back and also a possibility to run programs interactively.

To use the HYPAC-NAMMU package at QZs Amdahl computer you should have a basic knowledge of

  • JCL (Job Control Language) for the MVS operating system
  • GUTS (Gothenburg Universities Terminal System)
  • FEMGEN (mesh generator program)

You should also have access to various GUTS procedures and JCL-decks

Page 37 (Appendix 7) of the same report gives a couple of examples of GUTS commands. Firstly, to print the contents of an MVS/OS file (actually a PDS member):

OSPRINT 'SD.ABC123S.COIBAN(abcd)'

And, to convert an OS/MVS file to a GUTS file:

OSTS 'SD.ABC123S.COIBAN(abcd)',gutsfilename

An interesting message sent to IBM-MAIN mailing list in July 1999, by Martin Leist, Technical Analyst, Norfolk County Council, UK:

All this talk of ROSCOEand COM-PLETE etc, but is anyone else out
there still using GUTS (Gothenburg University Timesharing System) ?
It was quite popular over here in the UK during the 1980's as it ran on
VS1 and MVS as well as the earlier incarnations of those systems.
We are now the only UK site (we think) still actively running the
system and have to do our own (source) modifications at each new
level of OS/390. GUTS is *very* TSO compatible, so much that you
can actually run ISPF under it, although that defeats the object as
GUTS uses one address space for all its editing and file
management, only using other address spaces when invoking an
external program, those being shared among all users. I think the
system was used by some parts of the US IRS but have no idea if its
still used.

One of the active (UK) developers later went on to IBM and used
some of the techniques that had been used in GUTS to improve the
performance of CICS. The "end" of GUTS was signalled when
Gothenburg University moved away from a mainframe based system.

Still, in terms of actual technical information, all I have been able to find is these few tidbits. Is anyone aware of any further sources?

1
  • 2
    @TomasBy thanks. What I am keenest on finding is technical details – although information on who used it is also interesting. parlament.gv.at/PAKT/VHG/XVI/III/III_00121/imfname_553284.pdf is 1984 report by the Austrian Government, on computer hardware/software use in Austria. PDF page 291 mentions GUTS was used by University of Linz Oct 23, 2022 at 9:31

4 Answers 4

11

I know for certain that Trinity College Dublin used GUTS on their IBM 360/44. I studied there from 1975-1979 and personally used it. It was phased out around 1978-1979 when they got a new DECSystem 20. We undergraduates used it to edit our programs. We used the PL/C compiler. I remember finding a bug in GUTS where I mistakenly entered a negative line number in an edit and crashed the whole mainframe!

8

Ingemar Dahlstrand who was running Industridata operations in Gothenburg after the demise of Facit mentions GUTS in his article "The Development of University Computing in Sweden 1965–1985" page 130, in "IFIP Advances in Information and Communication Technology, History of Nordic Computing 2, Second IFIP WG 9.7 Conference, HiNC2 Turku, Finland, August 21-23, 2007 Revised Selected Papers". Ingemar is 91 years old by now but he can be contacted and since he was working in Gotheburg at the time he might know more about this.

3
  • 1
    Added links to book and chapter. Hope this is fine with you.
    – Raffzahn
    Jan 22 at 15:14
  • 1
    That article carries a historically very interesting note supporting that machines were not always ordered to fit the needs of their user, but rather what management thought makes them look good: "At one time, I suggested exchanging the Univac 1108 with one module of 64 K words for a somewhat slower Univac 1106 with three modules, which would have suited us better, at the same cost. The reaction was very negative. Speed still carried prestige, dating from the origin of computing, irrespective of whether we could sell it or not."
    – Raffzahn
    Jan 22 at 15:18
  • The article is only available thru Springer (or if you have the proceeding) but there is a transcript in Swedish of a talk about Higher Computer Science education in a swedish perspective. That transcript is available from diva-portal. diva-portal.org/smash/… Jan 23 at 12:01
7

Linz (Austria) University used GUTS in the early 1980s.

2
  • 2
    That's useful information. Could you elaborate? Given how little we have right now, anecdotes (or anything else you have) would be helpful. You can edit your answer to add more.
    – wizzwizz4
    Oct 27, 2022 at 18:24
  • 1
    If you've lost access to the account you used to post this answer, you could make a new one and request them merged using the [contact] form. (You don't have to, though; so long as you don't vote on your own stuff, using multiple accounts is allowed.)
    – wizzwizz4
    Oct 27, 2022 at 18:25
2

It is a very long time ago but I went to Trinity College Dublin from 1969 and used GUTS via their display screens, they had eight 2260s I think. By todays standards the hardware was pathetic, I think 256K later extended to 512K and operations taking 4nsec. But you could edit files and run FORTRAN programs and the response was pretty good! That side was used during the day for teaching students. The compiler was held in store and FORTRAN programs would be given a small amount of time to run to completion. A different system was run at other times which allowed one or two programs to run at once and programs had to be put in using boxes of cards. I'm scraping my memory now but I think the compiler used with GUTS came from Waterloo in Canada and if you used the value 0xdeaddead it would stop your run saying the value was uninitialised.

3
  • Welcome. Was this the 360/44 mentioned in the question? And I'd suppose a FORTRAN compiler from Waterloo to be WATFOR?
    – dave
    Feb 14 at 23:07
  • Had a look around and a few bits misremembered, see treasures.scss.tcd.ie/hardware/TCD-SCSS-T.20121208.018/… GUTS wasn't there at the start, it was put in later, and the machine had 128K to start with later increased to 256k! Feb 15 at 0:00
  • And yes it was WATFOR Feb 28 at 17:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .