Videos of the lectures and associated seminars of the series Seven Keys to the Digital Future, Edinburgh, UK, September 23^rd- October 6^th (inspired from my courses at Collège de France and organized by Michael Fourman, whom I warmly thank)

Lectures:

0. Revealing Informatics (inaugural lecture)

1. Sequential and Parallel Models of Computation

2. Circuits and Systems on Chips

3. Finite States Systems

4. Synchronous Concurrency

5. Communication and Networking

6. Asynchronous Concurrency

7. Getting Rid of Bugs

Seminars :

1. Neuromorphic Computation for Vision by Yves Frégnac, UNIC / CNRS Gif-sur-Yvette, France

2. Brains in Silico, by Steve Furber, University of Manchester, UK

3. Molecular Programming, by Luca Cardelli, Microsoft Research Cambridge, UK

4. Musical Synchrony, by Arshia Cont, IRCAM Paris

5. Robots that Learn, by Sethu Vijayakumar, University of Edinburgh, UK

6. Diffuse Systems, by Manuel Serrano, INRIA Sophia-Antipolis, France

7. Embracing Uncertainty, by Christopher M. Bishop, Microsoft Research Cambridge and University of Edinburgh, UK

Video and slides of the conference “The Evolution of the Synchronous Model” given in Autrans, France, September 2008.

1.2. En Français

Vidéos du cycle de conférences « Et le monde devint numérique », Universcience, la Cité des sciences, Paris, mai-juin 2011

1. Les racines scientifiques du monde numérique (18 mai 2011)

2. La révolution numérique dans les sciences (25 mai 2011)

3. A la chasse aux bugs, la maladie du certain (8 juin 2011)

Vidéo de la conférence « Les inversions mentales de la révolution numérique » donnée à Sophia-Antipolis le 31 janvier 2011.

Vidéo de la conférence « La révolution numérique dans les sciences » donnée dans le cadre du Colloque de rentrée du Collège de France, 14-15 octobre 2010.

Vidéos et transparents du cours 2009-2010 au Collège de France, « Penser, modéliser et maîtriser le calcul informatique », ainsi que des séminaires associés.

Vidéos et transparents du cours 2007-2008 au Collège de France, « Pourquoi et comment le monde devient numérique », ainsi que des séminaires associés.

Sur le site du Collège de France, les vidéos du cours 2009-2010 sont en format pdf sans les animations. Vous trouverez ici les transparents au format pps, qui préservent toutes les animations. Ils sont lisibles sous PowerPoint, OpenOffice, et d’autres lecteurs gratuits téléchargeables sur le Web

Leçon inaugurale du 19/11/2009 : Penser, modéliser et maîtriser le calcul

Cours 1 du 25/11/2009 : Syntaxe, sémantique, calculabilité :

Cours 2 du 02/12/2009 : Le lambda-calcul : réductions, causalité et déterminisme

Cours 3 du 09/12/2009 : Sémantiques, information et points fixes

Cours 4 du 16/12/2009 : Automates et systèmes de transitions

Cours 5 du 06/01/20010 : Parallélisme asynchrone

Cours 6 du 13/01/20010 : Parallélisme synchrone et vibratoire

Cours 8 du 13/01/20010 : La nécessaire mais délicate coopération de modèles + Les questions

2. Scientific and Industrial Activities

My global research revolves around five fields: mathematical models of computation, associated programming and circuit description languages, implementation of compilers to hardware and software, usage of formal verification systems, and industrial applications.

2.1. Recursive programs, lambda-calculus, and sequentiality

My initial work (1973-1976) at Ecole des Mines de Paris and IRIA was about understanding how to revert the order of recursive computations, for instance by transforming the Fibonnacci iterative algorithm into the iterative one. I proved the optimality in time and space of Rice’s algorithms to compute the Ackerman function as defined by its classical program. In this study, I introduced the notion of a stable function that became fundamental for what followed.

With Jean-Jacques Lévy, I then studied minimal and optimal syntactic computations in recursive programs and in the l-calculus. After that, I switched to the full abstraction problem specified by Plotkin and Milner: build a semantical model of PCF (typed l-calculus + arithmetic) that exactly contains all functions that are definable by syntactic terms. I first established the syntactic properties of stability and sequentiality of the l-calculus, which essentially show that it cannot express parallel computation. I developed the stable model, which preserves stability in the semantics, unlike the classical Scott Model.

In 1977, I moved to the new Ecole des Mines site at Sophia-Antipolis. With Pierre-Louis Curien, I developed the sequential algorithm model, where functions are replaced by algorithms that can be presented in two ways: pairs of a stable function and a computation strategy, or programs in a simple kernel programming language. We showed that this model forms a cartesian closed category, hence an non-functional model of the l-calculus. With M. Devin and F. Montagnac, we then developed the CDS programming language. The study of categories as models of the lambda-calculus was also at the root of the categorical abstract machine (CAM) which started the development of CAML.

2.2. Automata, real-time, and Esterel

In 1982, INRIA opened its new site in Sophia-Antipolis, and I created a joint project between Ecole des Mines and INRIA with Gérard Boudol. At that time, I switched to automata theory and real-time programming. With Jean-Paul Rigault and Jean-Paul Marmorat, we gave the initial definition of the new Esterel synchronous programming language dedicated to embedded systems. In cooperation with the Lustre teams of Nicolas Halbwachs and Paul Caspi and the Signal team of Albert Benveniste and Paul Le Guernic, we studied the associated synchronous model of concurrency. I essentially focused the theoretical study on the definition and usage of deterministic preemption structures in this model. With Laurent Cosserat, we developed the fisrt Esterel v1 compiler, soon followed by the Esterel v2 compiler developed with Philippe Couronné. Esterel v2 was based on the adpatation of a fast and beautiful algorithm developed by Ravi Sethi (then at Bell Labs) and myself to translate regular expressions into automata. We conducted industrial experiments at Bertin and Dassault Aviation, where we started a fruitful and longstanding cooperation with Emmanuel Ledinot.

In 1986, we started the development of a much more efficient compiler called Esterel v3, based on decisive fundamental research results by Georges Gonthier. This compiler was then industrialized by the companies Cisi Ingéniérie and ILOG and used by several customers.

2.3. Circuit synthesis, causality, and verification

In 1990, I joined Jean Vuillemin’s group at Digital Equipment as a part-time consultant. The group was developing the Perle FPGA-based machine and using it as a fast co-processor of alpha-based workstations. It was pretty clear how to develop data-flow hardware designs, but much less clear how to develop complex control-intensive designs. EstereI seemed to be a good language to specify them. I created a new translation from Esterel to circuits, which was readily implemented in the new Esterel v4 compiler. This new approach also had a decisive impact on software applications of Esterel: generating a circuit simulator in C instead of an explicit finite state machine made it possible to scale up to any application size, completely avoiding the potential exponential code size blow-up of Esterel v3. It also lead to major progress in understanding causality issues in Esterel and relating them to constructive logic. With Tom Shiple, we could show that a combinational cyclic circuit is electrically stable for all delays if and only if its Boolean equations can always be solved by using only constructive propositional logic, i.e., by disallowing usage of the excluded middle law X or not X = true. This lead to a new Esterel v5 compiler, soon followed by a modular version Esterel v6.

In 1999, I started cooperating with Michael Kishinevsky at Intel on extensions of Esterel to be able to define any kind of for-real synchronous circuits. This required the definition of deep extensions of the language to be able to support arbitrary data path and bit manipulation structures. We defined the new Esterel v7 language and studied how to compile it. The language is characterized by a smart arithmetic and bitvector type-checker that automatically ensures optimal sizing of intermediate variables in data paths.

2.4. Industrial development of Esterel

In January 2001, I joined the Esterel Technologies company created the year before, as its Chief Scientist. I directed the implementation of the Esterel v7 compiler (with A. Boulan, L. Arditi, M. Buchholtz, M. Perreaut, B. Pagano, and others). We solved many fundamental and technical problems needed for production usage: extending the language to support multiclock design; supporting a ECO-enabled flow (ECOs are late patches necessary when it has become too expensive to recompile, re-place and re-route the circuit); ensuring safe translation to VHDL and Verilog that are not well defined languages; linking the compiler with the Esterel Studio graphical interface and with the Prover SL formal verification system; instertion of the tools in complete industrial flows, etc. The Esterel Studio complete design and verification environment is used in production by ST Micro and Texas Instruments and in research by Intel, Philips (now NxP), ST Microelectronics, and Texas Instruments. It has also been widely distributed in Esterel Technologies’ academic program.

In 2003, Esterel Technologies acquired SCADE, the leading tool for synchronous avionics development, based on the Lustre Technology developed in Verimag, Grenoble, and used by major avionics companies such as Airbus. I participated in the development of the SCADE language and environment, and, in particular, of the new Scade 6 language that integrates the best of Lustre and Esterel. See www.esterel-technologies.com for more information.

2.5. Courses at Collège de France

In 2009-2010, I have been selected again by Collège de France to occupy the first year of the new chair Informatique et sciences numériques (Computer Science and Digital Sciences). I will teach a course called Penser, modéliser et maîtriser le calcul informatique (Thinking about, modeling, and mastering computation), where I will discuss why and how to model automatic computation, from the original sequential models to the current models of parallel computation.

In 2007-2008, I have been selected by Collège de France to occupy the yearly Technology Innovation Chair Liliane Bettencourt, sponsored by the Bettencourt-Schueller Foundation. I decided to give a general audience computer science course called « Pourquoi et comment le monde devient numérique » (Why and How the World Becomes Digital). The inaugural lecture analyzed why the digitization idea and the fantastic expansion of computers changed the world, and how this change will continue. It also explains the scientific background of this revolution. The inaugural lecture was followed by 7 2-hour course / seminar sessions, respectively on algorithms, circuits, programming languages, embedded systems, program verification, networks, images, cryptography, and a conclusion. For each subject, I invited experts from research and industry to give the seminars. The inaugural lecture and all courses and seminars (except cryptography) are available as videos (in French). The inaugural lecture is also available as a book in French.

2.6. Current research : Hop and HipHop for Web programming

Since March 2009, I am a member of the Indes INRIA project, which is dedicated to diffuse programming, i.e., the global programming of the set of computerized objects we now have at our disposal: computers, telephones, TVs, domestic appliances, etc. Safe programming of this network of devices will require tight cooperation between many sequential and parallel programming models. My own work is about understanding which models are needed where and how to make the models cooperate. We are currently developing HipHop, an Esterel-based extension of Manuel Serrano’s Hop language and system. Hop is a Scheme-based multi-tier language to develop complex Web applications with a single source code for the server and client, making code migration and client / server communication fully transparent. HipHop adds the possibility of orchestrating complex concurrent behaviors from with Hop. Compared to Esterel, it is a much more dynamic language, whose programs are Hop values that can be constructed on the fly and run by an interpreter that implements the constructive causality of Esterel. HipHop can be used both on the server and client side for maximal flexibility.

3. Carreer

3.1. Positions

2009 - : Director of Research at INRIA Sophia Antipolis.

2009 - 2010 : Professor at Collège de France, Chaire informatique et sciences numériques 2009 / 2010

Course: Penser, modéliser et maîtriser le calcul (Thinking about, modeling, and mastering computation). Videos and slides of the lectures and seminars (in French) will be available here.

2008 : Professor at Collège de France, Chaire d'Innovation technologique 2007 / 2008

Course: Pourquoi et comment le monde devient numérique (Why and How the World Becomes Digital). Videos and slides of the lectures and seminars (in French) are available here.

2001 - 2009 : Chief Scientist of Esterel Technologies.

1990 - 2001 : Regular consultant at Digital Equipment, Schneider Electric, Synopsys, Cadence Design Systems, and Intel.

1977 - 2001 : Director of Research at Ecole des Mines de Paris, Sophia Antipolis site. Co-head of a joint project with INRIA Sophia Antipolis.

1979 : Docteur d’état in Mathematics at Université Paris VII, option Computer Science.

1970 - 1977 : Researcher at Ecole des Mines de Paris, Paris site.

1970 : Entered Corps National des Ingénieurs des Mines.

1967 : Entered Ecole Polytechnique.

3.2. Additional responsibilities

2010 - : Member of the board (conseil d’administration) of ANR (Agence Nationale de la Recherche).

2011 : Member of the Scientific Council of IRCAM (Institut de recherche et coordination acoustique / musique).

2010 : President of the Scientific Council of the Computer Science Department at Ecole Normale Supérieure.

2005 - 2009 : delegate of the Inter-Section of Applications at Académie des sciences.

3.3. Awards

2008 : Chevalier de l'Ordre du mérite.

2005 : Great Prize of the EADS Foundation for applications of sciences to industry.

2005 : Member, Académie des technologies

2002 : Member, Académie des sciences.

2001 : Nominee, World Technology Awards.

1999 : Science and Defense award.

1994 : Chevalier de l'Ordre des palmes académiques.

1993 : Member, Academia Europaea

1990 : Monpetit Prize of Académie des sciences.

1979 : Bronze medal of C.N.R.S.

3.4. Invited Talks

Invited speaker for Seven Keys to the Digital Future, set of 8 lectures of Collège de France in English at Informatics Forum, University of Edinburgh, and the Royal Society of Edinburgh, October 2010, see Section 1.1. above.

Invited speaker in 37 international conferences or workshops :

EMSOFT 2011 (Taipei), DCM 2010 (Edinburgh), LOLA 2010 (Edinburgh), Bits&Chips 2007 (Eindhoven), FMICS 2007 (Berlin), Ada Europe 2007 (Geneva), FDL 2006 (Darmstadt), CSR 2006 (St Petersburg), TACAS'2005 (Edinburgh), VLSI'2004 (Mumbai), ICCAD'2003 (San Jose), HLDVT'2002 (Cannes), SAME'2001 (Sophia Antipolis), CHARME'2001 (Edinburgh), FEMSYS'2001 (Munich), FST&TCS'2000 (New Delhi), ASE'2000 (Grenoble), CHARME'99 (Bad Herrenalb), CONCUR'98 (Nice), POPL'98 (San Diego), DSL'97 (Santa Barbara), CAV'97 (Haifa), CHDL'97 (Toledo), TACAS'97 (Twente), AMAST'96 (Munich), ICSC'95, (Bangkok), IWLS'95 (Lake Tahoe), MFPS'94 (Manhattan, Kansas), LICS'94 (Paris), POPL'94 (Portland), FSTTCS'93 (Bombay), REX'93 (Amsterdam), CONPAR'92 (Lyon), ESEC'91 (Milano), TAPSOFT'90 (Brighton), IFIP'89 (San Francisco), STACS'86 (Paris).

Invited speaker for the Tata Consultancy Services Distinguished Lecture, 2006 (Pune, India).

Keynote speaker for the IBM Programming Languages Day, 2006 (Yorktownheigths, USA).

Invited speaker at the Kieburtz Symposium, 2003 (OGI Portland, USA).

Invited speaker at the first Milner Lecture, Edinburgh Université, 1996.

Invited speaker at several summer schools : Marktoberdorf 1992 et 2008, Laser 2007, ArtistDesign 2008, and several schools in India, Malaisia, Thailand, etc.

3.5. PhD supervision and examination

Advisor of 15 PhD theses. President of the Jury of 8 PhD of Habilitation theses. Reviewer of 10 Habilitation theses and of 36 PhD thesie, including 4 foreign ones. Jury member of 33 other PhD theses, including 4 foreign theses.

4. Scientific Publications

4.1. Books

Penser, modéliser et maîtriser le calcul informatique

G. Berry, Fayard, Collection Collège de France, Nov. 2009.

Pourquoi et comment le monde devient numérique

G. Berry, Fayard, Collection Collège de France, Jan. 2008.

Compiling Esterel

D. Potop, S. Edwards, and G. Berry. Springer, 2007.

The Constructive Semantics of Esterel

G. Berry. Draft book, current version 3.0, Dec. 16th, 2002.

The Esterel Language Primer, version v5_91

G. Berry. Current version Esterel v5_91, june 2000.

4.2. Theses

Modèles complètement adéquats et stables des lambda-calculs typés

G.Berry. Thèse de Doctorat d'Etat, Université Paris VII (1979).

Calculs Ascendants des Programmes Récursifs.

G.Berry. Thèse de Troisième Cycle, Université Paris VII, 1976.

4.3. Journal and Book Articles

Constructive Boolean Circuits and the Exactness of Timed Ternary Simulation

M. Mendler, T. Shiple et G. Berry, accepted for publication in Formal Methods for Systems Design (2012).

Penser, modéliser et maîtriser le calcul informatique, résumé des cours

G. Berry. In Cours et travaux du Collège de France, année 2009-2010. Collège de France, 2010.

Pourquoi et comment le monde devient numérique, résumé des cours.

G. Berry. In Cours et travaux du Collège de France, année 2007-2008. Collège de France, 2008.

Esterel: a Formal Method Applied to Avionic Software Development

G. Berry, A. Bouali, X. Fornari, E. Ledinot, E. Nassor, R. de Simone. Science of Computer Programming 36 (2000) 5-25.

An Implementation of Constructive Synchronous Constructive Programs in Polis

G. Berry, E. Sentovich. Formal Methods in Systems Design 17(2), October 2000, Kluwer Academic Publisher.

The Foundations of Esterel

G. Berry. In Proof, Language and Interaction: Essays in Honour of Robin Milner, G. Plotkin, C. Stirling and M. Tofte, editors, MIT Press, Foundations of Computing Series, 2000.

A Hardware Implementation of Pure Esterel

G. Berry. Sadhana, Academy Proceedings in Engineering Sciences, Indian Academy of Sciences, vol. 17, part 1 (1992) 95-130.

Esterel on Hardware

G. Berry. Philosophical Transactions Royal Society of London A, vol. 339 (1992) 87-104.

The Synchronous Programming Language ESTEREL: Design, Semantics, Implementation

G. Berry and G. Gonthier. Science of Computer Programming, vol. 19, no. 2 (1993) 87-152.

The Chemical Abstract Machine

G. Berry and G. Boudol. Theoretical Computer Science, vol. 96 (1992) 217-248.

Incremental development of an HDLC entity in Esterel

G. Berry and G. Gonthier. Comp. Networks and ISDN Systems 22, (1991) 35-49.

The Synchronous Approach to Reactive and Real-Time Systems

A. Benveniste and G. Berry. Proceedings of the IEEE, vol. 79, no. 9 (1991) 1270-1282.

From Regular Expressions to Deterministic Automata

G. Berry and R. Sethi. Theoretical Computer Science 48 (1986) 117-126.

Theory and Practice of Sequential Algorithms: the Kernel of the Programming Language CDS

G. Berry and P-L. Curien. In Algebraic Methods in Semantics, Cambridge University Press (1985) 35-88.

Full Abstraction for Sequential Languages: the State of the Art

G. Berry, P-L. Curien, and J-J. Lévy. In Algebraic Methods in Semantics, Cambridge University Press (1985) 89-132.

Sequential Algorithms on Concrete Data Structures

G. Berry and P-L. Curien. Theoretical Computer Science, vol. 20 (1982) 265-321.

Minimal and Optimal Computations of Recursive Programs

G. Berry and J-J. Lévy. Journal of ACM, vol. 26, no. 1 (1979) 148-175.

Bottom-up Computations of Recursive Programs

G. Berry. RAIRO Informatique théorique, vol. 10, n. 3 (1976).

4.4. Conference Articles

HipHop: A Synchronous Reactive Extension for Hop

G. Berry, C. Nicolas, and M. Serrano. Proc. PLASTIC'11 Workshop, Portland 2011.

SCADE: Synchronous design and validation of embedded control software

G. Berry. Proc. General Motors India Lab Workshop, Bangalore, 2007, Springer-Verlag Lecture Notes, 2007.

An Esterel-based Formal Specification Methodology for Power Management Development

G. Dubost, S. Granier, and G. Berry. Proc SAME'07, Sophia-Antipolis, France, 2007.

Clocking Schemes in Esterel

L. Arditi, G. Berry, M. Kishinevsky, and M. Perreaut. Proc. Designing Correct Circuits DCC'06, Vienna, Austria.

Late Design Changes (ECOs) for Sequentially Optimized Esterel Designs

L. Arditi, G. Berry, and M. Kishinevsky. Proc. Formal Methods in Computer Aided Design (FMCAD'04), Austin, Texas, USA.

System Level Design and Verification using a Synchronous Language.

G. Berry, M. Kishinevsky, and S. Singh. Proc. ICCAD'03, San Jose.

Toplevel Validation of System on Chip.

G. Berry, A. Bouali, J. Dormoy, and L. Blanc. Proc. HLDVT'2001, Cannes.

Multiclock Esterel

G. Berry, E. Sentovich. Proc. CHARME'2001, Edinburgh, Correct Hardware Design and Verification Methods, Springer-Verlag LNCS 2144.

Data Path Generation for Esterel Programs

A. Girault and G. Berry. Proc. International Symposium on Signals, Circuits, and Systems, Iasi, Roumanie (1999).

Efficient Latch Optimization Using Incompatible Sets

E. Sentovich, H. Toma, and G. Berry. International Digital Automation Conference DAC'97, Anaheim, USA, 1997.

Latch Optimization in Circuits Generated from High-Level Descriptions

H. Toma, E. Sentovich, and G. Berry. Proc. International Conf. on Computer-Aided Design ICCAD'96.

Constructive Analysis of Cyclic Circuits

T. Shiple, G. Berry, and H. Touati. Proc. International Design and Testing Conference IDTC'96, Paris, France (1996).

Preemption in Concurrent Systems

G.Berry. Proc. FSTTCS'93, Springer-Verlag Lecture Notes in Computer Science 761 (1993) 72-93.

Communicating Reactive Processes

G. Berry, S. Ramesh, and R. K. Shyamasundar. Proc. 20th ACM Conf. on Principles of Programming Languages (POPL), Charleston, Virginia (1993).

Optimized Controller Synthesis Using Esterel

G. Berry and H. Touati. Proc. Intl. Workshop on Logic Synthesis, Lake Tahoe, USA (1993).

Real Time Programming: Special Purpose or General Purpose Languages

G. Berry. Information Processing 89, G.X. Ritter (Ed.), Elsevier Science Publishers B.V., North-Holland (1989) 11-18.

The Synchronous Programming Language ESTEREL and its Mathematical Semantics

G. Berry and L. Cosserat. In Seminar on Concurrency, Springer-Verlag LNCS 197 (1984) 389-448.

ESTEREL: Towards a Synchronous and Semantically Sound High-Level Language for Real-Time Applications

G. Berry, S. Moisan, and J-P. Rigault. Proc. IEEE Real-Time Systems Symposium, Arlington, Virginia, IEEE Catalog 83CH1941-4 (1983) 30-40.

Programming with Concrete Data Structures and Sequential Algorithms

G. Berry. Proc. ACM Conf. on Functional Programming Languages and Computer Architecture, Wentworth-by-the-sea, USA (1981).

On the Definition of Lambda-calculus Models

G. Berry. Proc. Int. Coll. on Formalization of Programming Concepts, Peñiscola, Spain, Lecture Notes in Computer Science 107, Springer-Verlag (1981) 218-230.

A Survey of Some Syntactic Results in the Lambda-calculus

G. Berry and J-J. Lévy. Proc. Ann. Conf. on Mathematical Foundations of Computer Science, Olomouc, Tchecoslovaquia, Lecture Notes in Computer Science 74, Springer-Verlag (1979).

Stable Models of Typed Lambda-calculi

G. Berry. Proc. 5th Coll. on Automata, Languages and Programming (ICALP), Lectures Notes in Computer Science 62, Springer-Verlag (1978) 72-89.

Séquentialité de l'évaluation formelle des Lambda-expressions

G. Berry. In Program Transformations, 3eme Colloque International sur la Programmation, DUNOD, Paris, B. Robinet ed. (1978) 67-80.

Program Equivalence and Canonical Forms in Stable Discrete Interpretations

G. Berry and B. Courcelle. Proc 3rd Coll. on Automata, Languages and Programming (ICALP), Edinburgh University Press (1976).

TIF, File Interrogation and Transfer System

G. Berry and P. Pietri. Proc. TIMS/ORSA Conference, Atlantic City (1972)

5. Publications ‘Pataphysiques

Depuis le 22 novembre 2009, jour de ma leçon inaugurale donnée à la Cité des sciences à Paris, je suis Régent de Déformatique au Collège de ‘Pataphysique. La Déformatique se définit ainsi :

L’informatique c’est la science de l’information, la Déformatique, c’est le contraire.

Sa devise est empruntée à Oscar Wilde :

Appuyez vous sur les principes, ils finiront bien par céder.

Publications :

Manifeste pour la réhabilitation du pavillon des poids et mesures

G. Berry, Viridis Candela, Correspondancier du Collège de ‘Pataphysique, n^o 1, pages 33-60 (2007)

Les Shadoks sont-ils décervelables ?

G. Berry, Viridis Candela, Le Correspondancier du Collège de ‘Pataphysique, n^o 7, pages 57-70 (2009)

Un mauvais Cardeur

G. Berry, Viridis Candela, Le Correspondancier du Collège de ‘Pataphysique, n^o 8, pages 51-52 (2009)