CERN Accelerating science

Abstract

This small article is triggered by the historical note written by René Brun, "The Design and Implementation of Large Packages: Hbook, GEANT3, PAW, ROOT", in the last "Special Edition - 35th Anniversary" CNL, in order to give some detail regarding the GISMO package and people who have been involved in that project.

René Brun wrote, in the last CNL:

On the software side, we had a pleasant workshop in Erice at the end of 1989. The first solid attacks against Fortran were made and the first advocates of Object Oriented Programming appeared in our field. In 1990, Toby Burnett from SLAC had an interesting prototype called GISMO of a simulation program written in Objective C and developed on a very advanced workstation (Next).

Following are some historical details (with most people involved) concerning this project.

Graphics used to mean Tektronix storage scopes until the ERASME film measurement people (Christer Ljuslin and I, based on ideas from Hans Drevermann and others) developed hardware and software for a dynamic 3D display with a grey-level overlay. Jürgen Bettels and I then went on to develop the 3D dynamic displays used by UA1 and UA2 with Jean-Pierre Vialle. You will have seen the photos which ultimately appeared in New Scientist and Scientific American from UA1 (or maybe you are too young!).

We were joined by Leon Osterijk and started on the PIONS system, which was certainly the first Object-Oriented development at CERN. Leon wrote an Object-Oriented Fortran compiler, analogous to Objective-C, and PIONS was also a very early client-server system (like X windows) and had many other avant-garde facilities five years or more ahead of its time. (It included many ideas which later appeared in PHIGS). GKS was introduced for low-end 2-D applications, which must have been when you came on the scene, basically because we could buy software and drivers from outside.

I went on to start GISMO, with Mick Storr and Bill Atwood, later joined by Toby Burnet and many others. When this did not get enough interest at CERN I joined Wolfgang von Rueden and developed the front-end controls for the neutrino beam in C++, something which is still being used in NA48, COMPASS, HARP, and so forth. During that period much more interest developed in O-O, and projects like MOOSE were started, and later, GEANT4.