From brannon Tue Apr 7 20:48:16 1992 Status: RO X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["30346" "Tue" "7" "April" "92" "20:48:11" "PDT" "Terrence M. Brannon" "brannon " nil "902" "Object Oriented Parallel Environments" "^From:" nil nil "4"]) Received: from stun4c.cs.caltech.edu by jove.cs.caltech.edu (4.1/1.34.1) id AA16932; Tue, 7 Apr 92 20:48:11 PDT Message-Id: <9204080348.AA16932@jove.cs.caltech.edu> Received: by stun4c.cs.caltech.edu (4.1/SMI-4.1) id AA26380; Tue, 7 Apr 92 20:49:17 PDT Comments: Hyperbole mail buttons accepted, v3.02P. From: brannon (Terrence M. Brannon) To: brannon Subject: Object Oriented Parallel Environments Date: Tue, 7 Apr 92 20:48:11 PDT Newsgroups: comp.parallel From: David A Dunn - Utah State University Subject: Concurrent Object Oriented - A Summary (long) Organization: Utah State University Date: 26 Feb 92 17:16:39 MDT I posted a question on Object-Oriented Concurrent Programming to this newsgroup about a month ago, several people asked for a summary of responses. This is a full listing of the responses I received along with the original posting. It is LONG! (About 900 lines due to included bibliography entries...) -------------------------------Original Post---------------------- I am currently writing a paper on Object-Oriented Concurrent Programming. As part of this paper I would like to include some information on what researchers and developers are currently pursuing in this area, both in industry and education. If you are using, or have used, Object-Oriented methods to implement some form of parallelism (distributed, true parallel, or concurrent on one CPU), I would appreciate it if you could answer a few short questions and email me your responses. 1. What kind of system are you developing/using? Is it intended for production or for research? 2. What language/methodology are you using to implement this system? 3. Why did you choose an object-oriented methodology over the more traditional methods for concurrent programming? 4. Anything else you can tell me about your project would be very useful and appreciated. Thank you for your time, David A Dunn Utah State University slhpv@cc.usu.edu Computer Science Dept. (801)755-0675 Logan, UT 84322 ------------------------------------------------------------------------ From: IN%"kthanasi@cs.rochester.edu" 6-FEB-1992 15:21:18.64 Object Oriented Concurrency 1. I am developing an object oriented environment for light weight threads with built in mechanisms for locality management. It is part of my area paper work and hope :-) that it is research 2. The implementation is in C++. 3. Mainly because of the clean interface that I can have to the data. I know that data can be touched only via well define methods which can be augmented to save and use location information. 4. Basicly I want to provide an integrated package with a simple a definition for a computation on top of which one can build a number of different abstractions (lw threads, generators, chores etc). I also want to exploit locality in programs by augmenting objects with location information Leonidas Kontothanassis | Internet: kthanasi@cs.rochester.edu Dept. of Computer Science | UUCP: ..!rutgers!rochester!kthanasi University of Rochester | Phone: (716) 275-2569 Rochester NY 14627-0226 | FAX: (716) 461-2018 ------------------------------------------------------------------------ From: IN%"pcb@gunfight.austin.ibm.com" 6-FEB-1992 15:42:16.19 1. An object-oriented framework for concurrent graphics in heterogeneous environments ph.d. dissertation April 1991 available at UMI 2. The specification syntax was GO|| (see title above: stands for concurrent graphics objects and pronounced "geaux" (or go for the non-cajun)) 3. Demonstrate and specifiy functionality and extensibility not easily found in existing systems 4. combines graphics paradigms (3d,2d,gui,images,...and rendering algorithms) into a consistent framework..suitable for scalability into heter. envts., ala. mach. /*---- Thanks in advance... ----------- These are my own opinions -----------+ | Peter C. Bahrs INTERNET: pcb @ gunfight.austin.ibm.com | | IBM Boca System Design peterb @ bcrvmpc1.vnet.ibm.com | | *IX, OS/X VNET: peterb at bcrvmpc1 | +---------------------------------------------------------------------------*/ ------------------------------------------------------------------------ From: IN%"jbw5g@redwood.cs.Virginia.EDU" 10-FEB-1992 11:58:39.86 Below we include our brief description of what our system, Mentat provides. If you want to receive some papers, please let us know. Mentat: Object-Oriented Parallelism for MIMD Architectures Mentat Objectives o Easy-to-use parallelism o Scalable software architecture o Portable o High performance Mentat Features o Object-oriented (extended C++) o Medium grain o Compiled parallelism o User specified granularity Mentat addresses two problems o Programming MIMD architectures is difficult. o Lack of applications portability between architectures. The Programming Problem - Explicit Control In extended C or FORTRAN with shared memory and semaphores or message passing (send/receive), the programmer has total control over granularity, but: o Programmer must manage o problem decomposition o scheduling o communication o synchronization o Programmer may be overwhelmed by complexity and "Heisenbugs" The Programming Problem - Compiler Control The compiler manages complexity for the programmer, but: o Programmer intervention may be necessary to eliminate artificial dependencies o Granularity is too fine for message passing machines What's Needed? o Safety (correctness) of compiler control o Granularity control of explicit approaches o Portability - a virtual machine that the compiler can work with and that maps well to a variety of MIMD architectures The Solution - Mentat Mentat combines the best aspects of both explicit and compiler based approaches. The programmer makes granularity and encapsulation decisions exploiting domain knowledge. The compiler manages communication and synchronization. Computation Model Underlying Mentat o Not directly visible to programmers - only the compiler o Medium grain o Graph based o Graph expansion via node elaboration o Node elaboration is a local operation, no communication is required o Scalable Mentat Programming Language (MPL) o An extended C++ o Mentat Classes are the units of granularity o Mentat objects used as C++ objects o Compiler manages communication and synchronization o Compiler translates to C++ with calls to perform run-time data flow detection and build program graphs Mentat Run-Time System o Scalable o Portable o Performs run-time data flow detection o Automatic scheduling with optional programmer placement o Layered approach System Availability - FREE o Sun 3 & Sun 4 o Intel iPSC/2 Ports in Progress o Mach (using Mach IPC) o Intel iPSC/860 o IBM RS/6000 o Silicon Graphics Iris Is Your Application o Computation bound? o IO bound? ------------------------------------------------------------------------ From: IN%"Dietrich.Boles@arbi.informatik.uni-oldenburg.de" "Dietrich Boles" 12-FEB-1992 07:41:18. Ol, 12/02/92 Hello, i read your questions in the newsgroup comp.parallel. I am working at the University of Oldenburg (Germany) in the group of Prof. Appelrath. My name is Dietrich Boles. At University of Oldenburg (Germany) we are developing a parallel extension of C++, called QPC++ (Quasi-Parallel C++, because processes will be realised as lightweight-process, which share a common virtual address-space and will run on one CPU). A short summary of the language: aims: ---- + upward-compatibility to C++ + availability of all C++-libraries without recompilation + only a few new language constructs + realising the processes as lightweight-processes process concepts: ----------------- + there exist process classes with inheritance mechanisms (identical to normal object classes) + dynamical process creation + single-threaded processes (with a body) + process sets + implicit scheduling (but it can be done explicit too) communication: -------------- + via shared variables + atomic operations + semaphores (realised as process class) + via message-passing + extended rendezvous-concept (remote procedure calls) + synchronous and asynchronous communication + multicasting facilities. State of work: -------------- There exists a language definition (about 150 pages) (but it is written in the german language). A first run-time-system (prototype) has been written in C++ (on SUN-4, UNIX BSD 4.3, gnu-c++-compiler). We will write a compiler which translates QPC++-code into C++-code using LEX and YACC (or LEX++ and YACC++). Answering your questions: ------------------------- 1. QPC++ is intended for research; we will use it in devellopping an user-interface toolkit and an user-interface management system. (but it can be used in other projects too) 2. The run-time-system of QPC++ is modelled in a full object-oriented way. We use C++ as programming language. 3/4. I think it is a natural way to combine object-oriented with concurrent programming. Objects in object-oriented languages can be seen as processes which can send requests to other objects, waiting for response. But objects are passiv. Processes in an parallel object-oriented language can be seen as active objects, which will communicate with other processes but each process has an own thread of controll. Synchronisation mechanisms are necessary. With an normal object-oriented language it is sometimes very hard to transfer a system which has been modelled in an object-oriented way into code. Just in the field of user-interface toolkits or simulation objects are often active objects. With an parallel object-oriented language such a system can be transfered into code 1:1. Ok, I hope I could help you a little bit. Dietrich. PS: Sorry, but my english is not very good. /* * * * * * * * * * * * * * * * * * * * * * * * */ Dietrich Boles Universitaet Oldenburg Fachbereich Informatik Postfach 2503 D-2900 Oldenburg Germany e-mail: boles@uniol.uucp /* * * * * * * * * * * * * * * * * * * * * * * * */ ------------------------------------------------------------------------ From: IN%"grivas@inf.ethz.ch" 12-FEB-1992 09:03:05.11 Hi David, I am a PhD student, and I use C++ for the Implementation of Symbolic Computation Systems. Especially resp. unifier of such a system. I'm still in the beginning, so I think there are three levels, where I can parallelize: - matching in a many(patterns) to one(expression) way - inside an expression - inside the associative-commutative operation Can you please, email me your list of references about Conc. OOP? I am new in this area. The only reference that I have is from Oscar Nierstraz "The Next 700 Conc. OO Languages" I' m interested also in your paper. Can I get it from a server? Thanks in advance. George Grivas Theoretical Computer Science ETH Zentrum, IFW 8092 Zurich Switzerland Phone: +41 1 254 7391 Fax: +41 1 262 3973 Email: grivas@inf.ethz.ch George Grivas Theoretical Computer Science ETH Zentrum, IFW 8092 Zurich Switzerland Phone: +41 1 254 7391 Fax: +41 1 262 3973 Email: grivas@inf.ethz.ch ------------------------------------------------------------------------ From: IN%"Paulo.Borba@prg.oxford.ac.uk" 14-FEB-1992 20:27:50.00 Dear David, I am currently working with OO concurrent programming. The project consists basically in the extension of FOOPS (see references below) to support concurrency. I would be grateful if you could make me available the results of your research. >1. What kind of system are you developing/using? Is it intended >for production or for research? We are trying to specify communication protocols in FOOPS, since this should be a good experiment to the suitability of the language for this kind of application. >2. What language/methodology are you using to implement this >system? We are using FOOPS and OOZE which are OO languages based on OBJ. The languages are described in the papers below: @incollection(foops, title = "Unifying Functional, Object-Oriented and Relational Programming, with Logical Semantics", author = "Joseph Goguen and Jos\'e Meseguer", booktitle = "Research Directions in Object-Oriented Programming", editor = "Bruce Shriver and Peter Wegner", year = 1987, pages = "417-477", publisher = "MIT", note = "Preliminary version in {\it SIGPLAN Notices}, Volume 21, Number 10, pages 153-162, October 1986", comment = "Also, Report CSLI-87-93, Center for the Study of Language and Information, Stanford University, March 1987") @incollection(types-foops, title = "On Types and {FOOPS}", author = "Joseph Goguen and David Wolfram", booktitle = "Object Oriented Databases: Analysis, Design and Construction", editor = "Robert Meersman, William Kent and Samit Khosla", publisher = "North Holland", year = 1991, pages = "1-22", note = "Proceedings, IFIP TC2 Conference, Windermere, UK, 2-6 July 1990") @inproceedings(ooze, author = "Antonio Alencar and Joseph Goguen", title = "{OOZE}: An Object-Oriented {Z} Environment", booktitle = "European Conference on Object Oriented Programming", editor = "Pierre America", note = "Lecture Notes in Computer Science, Volume 512", publisher = "Springer", page = "180-199", year = 1991) >3. Why did you choose an object-oriented methodology over the >more traditional methods for concurrent programming? Actually, we are trying to compare different approaches and verify if there are advantages using an OO methodology. >4. Anything else you can tell me about your project would be >very useful and appreciated. Please, if you want to know anything else don't hesitate to contact me. Thank you, Paulo. ______________________________________ Paulo Borba Oxford University Computing Laboratory Programming Research Group 11 Keble Road, Oxford England, OX1 3QD Paulo.Borba@prg.ox.ac.uk ------------------------------------------------------------------------ From: IN%"Laurent.Thomas@loria.fr" "Laurent Thomas" 18-FEB-1992 00:49:40.52 David, I wrote a paper to appear in the proceedings of the ACM Symposium on Applied Computing at Kansas City in the beginning of March. This paper is : @INPROCEEDINGS{thomas92b, AUTHOR = {L. Thomas}, TITLE = {{A generic object-oriented concurrency mechanism for extensibility and reuse of Synchronization components (Revised Version)}}, BOOKTITLE = {{ACM SIGAPP Symposium on Applied Computing}}, ADDRESS = {Kansa City}, MONTH = {March}, YEAR = 1992, KEYWORDS = {concurrence, synchronisation, heritage, langage objet}} It deals with the interfering aims of synchronization expressions and inheritance. A refined version will appear in the LNCS Series: @INCOLLECTION{thomas92c, AUTHOR = {L. Thomas}, TITLE = {{Extensibility and Reuse of Object-Oriented Synchronization Components}}, BOOKTITLE ="Proceedings of the Parallel Architecture and Language Europe (PARLE'92)", PUBLISHER = {Springer-Verlag, Lecture Notes in Computer Science, Vol. ???}, ADDRESS = {Paris, France}, MONTH = {Juin}, YEAR = 1992, EDITOR = {???}, PAGES = {0--0}, KEYWORDS = {}} I may send you a Postscript version by e-mail, if you want. The basic problem is that the synchronization of a method M in a class C may not be the same as its synchronization should be in a derived subclass C' of C. We have to found some mechanisms to express only the changes between the synchronization of M in C and in C'. > 1. What kind of system are you developing/using? Is it intended > for production or for research? I'm developing a new language wich includes the propositions i made in the two mentionned articles. It is intended (at the moment) for research purpose only. > 3. Why did you choose an object-oriented methodology over the > more traditional methods for concurrent programming? That's a good question ! I feel that the extensibility and reusabilty are of primary interest in the development of quality sofware. At the same time parallel programming needs methodology to not reinvent the wheel ... So OOP seems a promising way to achieve extensibility and reusability in the parallel framework. Please, keep me in touch. It would be kind to have an exchange of our papers on the subject, wouldn'it? PS: I will be on vacation from the 22 to the 29 and then i'am going to go to Kansas City from the 1rst to the 8 of March. Laurent Thomas EMAIL : thomas@loria.fr POST : Centre de Recherche en Informatique de Nancy. Campus Universitaire. BP 239. F-54506 Vandoeuvre-les-Nancy CEDEX, France FAX : (+33) 83413079 ------------------------------------------------------------------------ From: IN%"thomas@cs.rochester.edu" 6-FEB-1992 08:48:00.22 I too, am interested in OO parallelism. Although it is not exactly what you asked for in your posting, I am sending you the (slightly) annotated bibliography for a paper I recently wrote on this topic. I would very much appreciate it if you could either post or mail me a summary of the responses you recieve. Thanks, -Jeff @BOOK{Agha86, AUTHOR = "Gul A. Agha", TITLE = "ACTORS: A Model of Concurrent Computation in Distributed Systems", PUBLISHER = "MIT Press", YEAR = 1986, ANNOTE = "ACTORS" } @INCOLLECTION{Agha87, AUTHOR = "Gul A. Agha and Carl Hewitt", TITLE = "Concurrent Programming Using Actors", EDITOR = "Akinori Yonezawa and Mario Tokoro", BOOKTITLE = "Object-Oriented Concurrent Programming", PAGES = "9--36", PUBLISHER = "MIT Press", YEAR = 1987, ANNOTE = "Act3" } @PROCEEDINGS{Agha88, TITLE = "Proceedings of the ACM SIGPLAN Workshop on Object-Based Concurrent Programming", EDITOR = "Gul Agha and Peter Wegner and Akinori Yonezawa", PUBLISHER = "ACM Press", MONTH = "September", YEAR = 1988 } @ARTICLE{Ahuja86, AUTHOR = "Sudhir Ahuja and Nicholas Carriero and David Gelernter", TITLE = "Linda and Friends", JOURNAL = "IEEE Computer Magazine", PAGES = "26--34", VOLUME = 19, NUMBER = 8, MONTH = "August", YEAR = 1986, ANNOTE = "Linda" } @ARTICLE{Andrews83, AUTHOR = "Gregory R. Andrews and Fred B. Schneider", TITLE = "Concepts and Notations for Concurrent Programming", JOURNAL = "ACM Computing Surveys", PAGES = "3--44", VOLUME = 15, NUMBER = 1, MONTH = "March", YEAR = 1983, ANNOTE = "Previously Cornell TR 82-520" } @INPROCEEDINGS{Arvind86, AUTHOR = "R. S. Arvind and K. K. Pingali", TITLE = "I-Structures: Data Structures for Parallel Computing", BOOKTITLE = "Proceedings of the Workshop on Graph Reduction", PUBLISHER = "Springer-Verlag", MONTH = "September/October", YEAR = 1986 } @BOOK{Birtwistle80, AUTHOR = "Graham M. Birtwistle and Ole-John Dahl and Bjorn Myhrhaug and Kristen Nygaard", TITLE = "Simula Begin", PUBLISHER = "Studentlitteratur", YEAR = 1980, ANNOTE = "Simula Language Description and Tutorial" } @BOOK{Budd87, AUTHOR = "Timothy Budd", TITLE = "A Little Smalltalk", PUBLISHER = "Addison-Wesley", YEAR = 1987, ANNOTE = "Little Smalltalk" } @BOOK{Budd91, AUTHOR = "Timothy Budd", TITLE = "An Introduction to Object-Oriented Programming", PUBLISHER = "Addison-Wesley", YEAR = 1991, ANNOTE = "Smalltalk, C++, Object Pascal (Macintosh), and Objective-C (NeXT)" } @TECHREPORT{Cox90, AUTHOR = "Alan L. Cox and Robert J. Fowler and Jack E. Veenstra", TITLE = "Interprocessor Invocation on a NUMA Multiprocessor", INSTITUTION = "University of Rochester Department of Computer Science", NUMBER = 356, MONTH = "October", YEAR = 1990, ANNOTE = "OSMIUM" } @ARTICLE{Caromel89, AUTHOR = "Denis Caromel", TITLE = "Service, Asynchrony, and Wait-by-Necessity", JOURNAL = "Journal of Object-Oriented Programming", PAGES = "12--22", MONTH = "November/December", YEAR = 1989, ANNOTE = "Futures for Eiffel" } @INPROCEEDINGS{Crowl88, AUTHOR = "Lawrence A. Crowl", TITLE = "A Uniform Model for Parallel Programming", BOOKTITLE = "Proceedings of the ACM SIGPLAN Workshop on Object-Based Concurrent Programming", PAGES = "25--27", YEAR = 1988, ANNOTE = "Matroshka" } @PHDTHESIS{Crowl91, AUTHOR = "Lawrence A. Crowl", TITLE = "Architectural Adaptability in Parallel Programming", SCHOOL = "University of Rochester", MONTH = "May", YEAR = 1991, ANNOTE = "Matroshka" } @ARTICLE{Flynn72, AUTHOR = "M. J. Flynn", TITLE = "Some Computer Organizations and Their Effectiveness", JOURNAL = "IEEE Transactions on Computers", PAGES = "948--960", MONTH = "September", YEAR = 1972, ANNOTE = "Flynn's Taxonomy" } @ARTICLE{Gelernter86, AUTHOR = "David Gelernter", TITLE = "Domesticating Parallelism", JOURNAL = "IEEE Computer Magazine", PAGES = "12--16", VOLUME = 19, NUMBER = 8, MONTH = "August", YEAR = 1986, ANNOTE = "Parallel languages vs. parallelizing transformations" } @BOOK{Goldberg83, AUTHOR = "A. Goldberg and D. Robson", TITLE = "Smalltalk-80: The Language and its Implementation", PUBLISHER = "Addison-Wesley", YEAR = 1983, ANNOTE = "Smalltalk-80" } @ARTICLE{Halstead86, AUTHOR = "Robert H. Halstead, Jr.", TITLE = "Parallel Symbolic Computing", JOURNAL = "IEEE Computer Magazine", PAGES = "35--43", VOLUME = 19, NUMBER = 8, MONTH = "August", YEAR = 1986, ANNOTE = "Multilisp" } @ARTICLE{Hoare74, AUTHOR = "C. A. R. Hoare", TITLE = "Monitors: An Operating System Structuring Concept", JOURNAL = "Communications of the ACM", PAGES = "549--557", VOLUME = 17, NUMBER = 10, MONTH = "October", YEAR = 1974, ANNOTE = "Monitors" } @INCOLLECTION{Lieberman87, AUTHOR = "Henry Lieberman", TITLE = "Concurrent Object-Oriented Programming in Act 1", EDITOR = "Akinori Yonezawa and Mario Tokoro", BOOKTITLE = "Object-Oriented Concurrent Programming", PAGES = "9--36", PUBLISHER = "MIT Press", YEAR = 1987, ANNOTE = "Act 1" } @BOOK{Kernighan78, AUTHOR = "Brian W. Kernighan and Dennis M. Ritchie", TITLE = "The C Programming Language", PUBLISHER = "Prentice-Hall", YEAR = 1978, ANNOTE = "C" } @ARTICLE{Nelson91, AUTHOR = "Michael L. Nelson", TITLE = "Concurrency and Object-Oriented Programming", JOURNAL = "ACM SIGPLAN Notices", PAGES = "63--73", VOLUME = 26, NUMBER = 10, MONTH = "October", YEAR = 1991, ANNOTE = "Survey of concurrent and distributed object-oriented languages" } @ARTICLE{Pancake91, AUTHOR = "Cherri M. Pancake", TITLE = "Software Support for Parallel Computing: Where Are We Headed?", JOURNAL = "Communications of the ACM", PAGES = "53--64", VOLUME = 34, NUMBER = 11, MONTH = "November", YEAR = 1991, ANNOTE = "Scientific computing" } @ARTICLE{Scott90, AUTHOR = "Michael L. Scott", TITLE = "The LYNX Distributed Programming Language: Motivation, Design, and Experience", JOURNAL = "Computer Languages", PAGES = "209--233", VOLUME = 16, NUMBER = "3/4", YEAR = 1990, ANNOTE = "LYNX" } @BOOK{Silberschatz91, AUTHOR = "Abraham Silberschatz and James L. Peterson and Peter Galvin", TITLE = "Operating Systems Concepts", EDITION = "Third Edition", PUBLISHER = "Addison-Wesley", YEAR = 1991, ANNOTE = "Introductory Operating Systems Text" } @BOOK{Ellis90, AUTHOR = "Margaret A. Ellis and Bjarne Stroustrup", TITLE = "The Annotated C++ Reference Manual", PUBLISHER = "Addison-Wesley", YEAR = 1990, ANNOTE = "C++ ANSI Base Document" } @BOOK{Stroustrup91, AUTHOR = "Bjarne Stroustrup", TITLE = "The C++ Programming Language", EDITION = "Second Edition", PUBLISHER = "Addison-Wesley", YEAR = 1991, ANNOTE = "C++" } @BOOK{Yonezawa87, EDITOR = "Akinori Yonezawa and Mario Tokoro", TITLE = "Object-Oriented Concurrent Programming", PUBLISHER = "MIT Press", YEAR = 1987, ANNOTE = "Actor based languages" } ------------------------------------------------------------------------ From: IN%"tb06@PL118f.CC.Lehigh.EDU" 7-FEB-1992 04:49:46.23 Yasuhiko Yokote wrote a dissertation on Concurrent Smalltalk. He is currently at the Sony Computer Science Laboratories. After emailing to him, he mailed me a copy of his dissertation. He also has contacts at his Ph.D university, Keio University where I understand a multiprocessor version of his uniprocessor ConcurrentSmalltalk is being developed. One of these email addresses should work and the other should bounce: ykt@csl.sony.co.jp jkt@csl.sony.co.jp One question: where do you plan to publish the paper and how will you notify the general public of its publication? \ ! / Terrence Brannon, PO Box 5027, Bethlehem, PA 18015 , USA -- O -- internet: tb06@pl118f.cc.lehigh.edu (215) 758-1720 (215) 758-2104 / ! \ computational biology . yoga . acupuncture . qi gong . tai chi Happiness! Happiness! Happiness! Happiness! Happiness! Happiness! ------------------------------------------------------------------------ From: IN%"kasper@iesd.auc.dk" "Kasper Osterbye" 10-FEB-1992 04:20:46.35 A side note, It you hear from anyone who examines inheritance in connection with concurrency, I am most interested. I attended the workshop on concurrency in connection with the ECOOP'91, and from the 30-40 researchers gathered there, there were none doing research with respect to processes and inheritance. So in my view there is not any work in Object-oriented concurrency, only in object based concurrency (Difference according to Wegner) --Kasper ------------------------------------------------------------------------ From: IN%"harbison@bert.pinecreek.com" 7-FEB-1992 14:20:10.81 FYI, the Modula-3 language has facilities for both object-oriented and concurrency, although the two are not connected in any conceptual way. Sam Harbison Pine Creek Software; Suite 300; 305 South Craig Street; Pittsburgh, PA 15213; USA. Phone&FAX: +1 412 681 9811. E-mail: harbison@bert.pinecreek.com. ------------------------------------------------------------------------ From: IN%"colbrook@concerto.lcs.mit.edu" 5-FEB-1992 15:53:44.02 I have sent you a copy of a paper that will appear in the PARLE'92 conference. You might find it of interest. - Adrian Colbrook ------------------------------------------------------------------------ From: IN%"levine@segur.ICS.UCI.EDU" "David L. Levine" 5-FEB-1992 13:20:03.63 Would I be able to get a copy when it is finished, even if you don't plan to publish? Thanks, -- to accept the universal responsibility of doing | David L. Levine good to all sentient beings, and if that is not | Internet: levine@ics.uci.edu possible, to do no harm | UUCP: uunet!ucivax!levine -- paraphrased from the Fourteenth Dalai Lama | BITNET: levine@uci ------------------------------------------------------------------------ From: IN%"deshpande-ashish@CS.YALE.EDU" "Ashish Deshpande" 8-FEB-1992 19:49:12.78 Would you please, please forward any information you receive to me too? Especially, any offers of ftp'able or e-mail'able papers or anything of the sort ... I'd really appreciate it. I'm seriously considering the field as a possible research area and need to find out as much as possible about it as quickly as possible. This would help a great deal. Thanks, Ashish ------------------------------------------------------------------------ From: IN%"dm_devaney@pnlg.pnl.gov" 5-FEB-1992 14:25:11.53 Please forward or post a summary of responses. Thanks in advance - Mike ******************************************************** * Mike DeVaney * (509) 375-2435 * * Pacific Northwest Laboratory * * * 906 Battelle Boulevard * FAX (509) 375-6631 * * mailstop K1-87 * * * Richland, WA 99352 * dm_devaney@pnl.gov * ******************************************************** ------------------------------------------------------------------------ From: IN%"terry@wag.caltech.edu" "Terry R. Coley" 5-FEB-1992 15:25:10.09 I would be interested in a copy of your net-summary when the responses come in. Also - your paper sounds very interesting, if you would care to forward a copy, I would enjoy reading it. My interest is in using Object oriented programming and parallism for high performance computational chemistry problems. Thanks. - Terry terry@wag.caltech.edu 818-397-2724 818-585-0918 (FAX) Terry Coley, Caltech 139-74, Pasadena, CA 91125 ------------------------------------------------------------------------ From: IN%"marsh@cec.mtu.edu" 6-FEB-1992 06:18:55.50 Could you please forward any responses that you get on this topic to me? I am working on my Master's thesis in this area, and I would greatly appreciate this. Thank you very much. Dianne Marsh Michigan Technological University Houghton, Michigan 49931 -- marsh@cec.mtu.edu ------------------------------------------------------------------------ Return-path: Dave, Would you mind sharing a copy of your paper when it becomes available? Thanks, Sukan ******************************************************************************* Sukan Makmuri Phone: (415) 543-0900 FAX: (415) 543-3716 Interactive Development Environment sukan@ide.com 595 Market Street, San Fransisco, CA 94105 uucp: ...!{sun,uunet}!ide!sukan *******************************************************************************  .