perm filename TECHN.JST[COM,LSP] blob sn#813021
filedate 1986-03-18 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00003 PAGES
C REC PAGE DESCRIPTION
C00002 00002 Proposed Justification Letter
C00010 00003 How we make decisions is up to us, but how we appear to the rest of the
Proposed Justification Letter
I'm not sure sending something like this is the right idea, but here
is my attempt at a justification message:
The purpose of this message is to shed some light on the decision-making
processs behind the selection of the technical committee.
I will present the criteria along with additional commentary enclosed in
(1) The member must have a deep knowledge of Lisp, with experience both in
the implementation and design of at least one serious Lisp implementation.
[Although one could argue that a non-Lisp individual with a deep
understanding of language design would be a good choice (and the ad hoc
selection committee did consider such individuals), we decided that only
selecting people from the Lisp community were essential. The candidate
must have struggled with at least one design and implementation of a Lisp
that is recognized as an important Lisp dialect: Lisp implementation
projects are not sufficient. The individual must also be a user of Lisp.
In some cases, the criterion of technical expertise was held very high,
and special consideration was made where a candidate demonstrated
remarkable technical judgement.]
(2) The member must be a well-recognized, prominent individual. He or she
must be someone whose reputation is recognized internationally.
[The X3J13 technical committee is the US representative in the
international arena. When Common Lisp comes up for a vote at the ISO
level, there is one vote per country. The US, the USSR, Monaco, Belgium,
and Liechtenstein, for example, get one vote each. The US committee should
contain mostly well-known, well-respected people, especially people
respected outside the US, so that the recommendations of the American
committee are taken seriously. There can be no compromises made for
internal US politics as far as the committee membership is concerned. We
can settle our political differences during the deliberations of the X3J13
committee work, but the membership of the committee must be above
reproach. For this reason, people who have done creditable work but who
are not known outside the US in the Lisp world had to have more than
compensatory technical merit to qualify for membeship.]
(3) The group, considered as a whole, must reflect the wide variety of
viewpoints and backgrounds that are present in the Common Lisp community.
[The committee cannot appear to have been stacked in favor of Common Lisp.
There is much interest in Scheme and Scheme-like languages, especially
outside the US. If only dyed-in-the-wool Common Lispers were represented,
the committee would not appear to be one whose main interest was promoting
a good Lisp for international standarization. EuLisp, a European
contender for ISO standardization, is heavily influenced by Scheme and by
3-Lisp. Our committe also must be above reproach here.]
(4) The committee must not be so large that it is unable to reach
decisions with reasonable speed.
[Many people are not willing to tolerate decisions by mass debate and
consensus. The outlines of Common Lisp are well-established, and the
decisions to be made are relatively minor compared with the main portions
of Common Lisp design. The decision-making procedures for the technical
committee are designed to give all organizations and responsible Common
Lisp individuals the US a voice concerning the issues, but not to cramp
(5) The members are chosen as individuals with a commitment to the success
of Common Lisp as a widely used standard, and not as representatives of
their respective companies or organizations.
[US politics, again, must be kept out of the committee.]
Another major consideration in evaluating potential members was their
ability to express their thoughts to each other and to the public. There
are several major Lisp implementors who were dropped from consideration
because they could not express themselves well enough to be public
spokesmen or to express their technical judgements.
Several times I have indicated that US politics must have no place on the
technical committee, as seen by the rest of the world. That doesn't mean
that US politics cannot have an influence on the decisions of the US
membership of the committee. Remember that X3J13 is open to public
membership; remember that each implementation group shall designate a
contact to make sure all relevant input is received regarding decisions.
I think it would demonstrate questionable judgement to state the
deliberations that went into choosing the individual committee members;
many lengthy discussions, vigorous debate, and soul-searching went into
the process of decision-making that the Common Lisp community asked us to
undertake last December. I believe that this committee is an excellent
one, and that ample means for guaranteeing input from all interested
parties are provided, not only by the X3 mechanisms, but also by the
additional structure we have set up.
How we make decisions is up to us, but how we appear to the rest of the
world depends on the membership of the technical committee.
Now I will move on to the part that is most distasteful about this message:
justifying each individual according to the criteria outlined above.
Alan Bawden, Massachusetts Institute of Technology
He was active in the early Common Lisp design effort. He has worked on the
MIT Lisp machine Lisp implementation, on MacLisp, and on Scheme. He did
serious design work for Connection machine Lisp dialects. His technical
judgement and taste is well-recognized.
Daniel G. Bobrow, Xerox Corporation
Danny wrote the precursor of InterLisp (and of BBN Lisp), he co-invented
spaghetti stacks, co-invented Tenex and paging, co-invented reference
count garbage collection schemes, and co-invented CDR-coding. He was
active in innovative programming language design during the 1970's,
especially with KRL. He was responsible for LOOPS, the InterLisp
object-oriented programming, and for Common Loops.
Scott E. Fahlman, Carnegie-Mellon University
Scott Fahlman is the de facto leader of the Common Lisp world.
He headed the SPICE Lisp design and implementation effort. He
brought SPICE out as the public domain Common Lisp implementation.
The fact that Common Lisp was accepted by many computer manufacturers
can be traced directly to him.
Richard P. Gabriel, Lucid, Inc.
Dick Gabriel began the Common Lisp movement, along with JonL White
and Guy Steele Jr. He headed the design and development of the S-1 Lisp
system. He wrote the book on Lisp performance evaluation. He headed
the design and development of Lucid Common Lisp. He and John McCarthy
invented Qlisp, a parallel dialect of Lisp.
Martin L. Griss, Hewlett-Packard, Inc.
Martin Griss pioneered the modern Lisp-on-stock hardware movement.
He headed the design and implementation of PSL, he pioneered portable
compiler technology for Lisp. He headed the HP Lab group that is
bringing Lisp into the HP culture.
David A. Moon, Symbolics, Inc.
David Moon is perhaps the best Lisp hacker in the world. The current
design of Symbolics Lisp and Symbolics Common Lisp are his products.
He is one of the foremost people in Lisp machine hardware.
Jonathan A. Rees, Massachusetts Institute of Technology
He is one of the premier Scheme people. He has designed and implemented
several dialects of Scheme. His work on Scheme compilers is innovative and
Guy L. Steele Jr., Thinking Machines, Inc.
Guy Steele Jr is one of the most important Lisp people in the world. He
authored the Common Lisp book, with help from his friends. he co-invented
Scheme, he co-designed NIL and S-1 Lisp.