Jean-Raymond Abrial in Grenoble, Christian Jullien testimony

  Extract from

FACS FACTS Issue 2025-2 July 2025 In memory of Jean-Raymond Abrial (1938–2025)
August 31, 2025

https://www.bcs.org/media/yd4ocehl/facs-jul25.pdf 

FACS FACTS Issue 2025-2 July 2025
In memory of Jean-Raymond Abrial (1938–2025)

 

 

 I didn't work directly with Professor Abrial, but I owe him much !


A brilliant and dedicated teacher


Firstly, he was my teacher (two years) during computer science master at IMAG Grenoble 

(1971/72). In 1971 he gave us a strong basis in data management. The classes were 

 fascinating; the lecture hall was truly captivated, the examples being "telling" 

 despite being based on a rigorous theoretical foundation.

I still have  hand-written lecture support  for additionnal courses that he gave for 

 free at university in the evening, after normal hours. 


This document (215 pages) was produced with an alcohol duplicator. He used  a lot of 

figures,  graphs and examples that could have in common with our current comics.\\ 



With these simple and concrete examples, he nevertheless presented essential principles 

 of computer science from the sole perspective of data, its representation, and its 

 processing, all within a rigorous and original approach. Everyone could understand 

 because the explanation was so clear.


It was very pragmatic and exciting. 


IMAG Labs where JR.Abrial was our professor and involved in SOCRATE project .

At that time he had a magnificent  Peugeot 403 coupelike Colombo's!


In 1972, the course was mainly on the theory of automates and computability (much more formal   -see the halting problem !- ), and reflected the particular focus that JR.Abrial  was putting on calculability and  proofs of programs..


SOCRATE

 

In 1968,  JR. Abrial had been commissioned by IMAG to lead a major research project 

on the development of management applications. The topic was data processing. 

 This project, called SOCRATE, gave birth to a software product ( a DBMS - 

data base management system-) in 1970, with the cooperation  of a few  local

 organizations (hospital, social insurance) . Industrialization followed at the  

CII company (which later became BULL), and by ECA Automation, a software house  

whose director, Pierre Thellier, had known JR. Abrial at the CPM (Naval Programming 

Center). The university team had largely joined ECA Automation for the first industrial 

portage for Social Security in Grenoble. Among them, George Beaume and George Vigliano,

 2 of the 3 main  PhD students. At ECA Automation the team was directed by S.Stepanian,

 ingineer who had been also at the CPM.  


A few years later (1976) , after some research activities and teaching in software 

 ingeneering activity, I could join the team involved in SOCRATE. I will  not report

 here all the story 

\footnote{\textcolor{red}{see https://www.dropbox.com/  "30years of DBMS developments 

 at Grenoble"}}, but in the first years, JR.Abrial went regularly visiting the team  

directed by  Stephen Stepanian, Georges Vigliano and Georges Beaume. With them and 

for more than 15 years I could study, develop, and put in software applications  

with the basic notions resulting from the work of JR.Abrial. 

Basically, SOCRATE was a revolution : a practical way for a high productivity in  

programming, and much more than a DBMS. Indeed, since 1970, Socrate was offering a 

 4GL langage which could provide the management of persistent data in a manner that 

 could free programmers from physical management considerations.


This software had been used for more than 30 years for a large number of applications 

in different areas (industry, utilities, defense, bank and insurance, ..) on 

 various platforms (IBM, SIEMENS, mini computers like SOLAR, MITRA, ..., VAX, PCs, ...).


Renamed CLIO-due to a trademark dispute- and anglicized, SOCRATE was sold in 1985 

in the United States, to two customers for use until around 2000. Contacts were 

also made in Singapore, but due to a lack of commercial resources, sales were mainly  

domestic.


Moreover, notions  coming with SOCRATE from the beginning  ( ie collections of data  

like tables, trees, lists, ...)  have been easily mapped to the  relational approach 

in 1992. An  SQL engine has been integrated , conformant to the ISO 92 standards. 

 It was not until 1988 that ORACLE brought PL-SQL  in order to offer a comparable

4GL-type programming language !


Concepts identified by JRA in 1968 were really ahead, and more over, an important 

source of value creation. We owe a great deal to this highly creative geniu


Christian Jullien   Member of Aconit    www.aconit.org

IMAG, Student of JR. Abrial (1970-1972)} 

SYSECA, DBMS Development Engineer 1976-1981

CNET Labs, DBMS Research Engineer 1982-1989

SYSECA, DBMS Project manager 1989-1993

SYSECA, Head of unit Grenoble 1993-1998 

HARDIS Group, Senior IT consultant 1999-2008