The aim of DIADEM was to re-implement the whole system using a modern programming language in order to become hardware independent. So far the system, written in 360-Assembler, was bound to the IBM-360 and the Siemens 4004 series. The programming language DUTAP was realized as Macro-Assembler language. Now the idea was to implement the system on different so called mini-computers. I intended to do this using Mumps. However, the experts consulting the ministry (BMFT) forced me to use Fortran instead. They said that would be available for every computer. I hade to give in, in spite of massive concerns due to my knowledge of Fortran‘s weaknesses in text processing. (By the way: Shortly later the ZI was allowed to use Mumps for the realisation of DOMINIG-III, an irony of IT-history!)
Three different mini-computers were selected for the demonstration: Siemens 404/3, Dietz Mincal 621, and Intertechnique Multi 20.
On the mini-computer Siemens 404/3 with max 48 KB working storage (nobody can imagine that today!) the Fortran compiler was ok. But the realization of the complex DUTAP-system required many ‚overlays‘ (more than twenty, if my memory is right). Consequently the generation of text was slow.
But the industry wished to market this system: Triumph-Adler used it in its ‚DOC-System‘. It is a pity that this first commercial DOC-System did not survive the failure of Triumph-Adler. It is now a star in our collection of old computers (see FITG-Buch 'Bollerwagen mit Dynamo', TA-DOC, pp. 129 and 226).
The Fortran compilers of Dietz and Intertechnique were prototypes and we became victims of what we ironically called ‚Fortability‘, the lack of compatibility of the internal data representation and treatment. Portability of our Fortran programs for text-processing and -generation without changes in the program code was impossible. (With Mumps that would not have been a problem!)