Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!ucsd!ucsbcsl!eiffel!bertrand
From: bertrand@eiffel.UUCP (Bertrand Meyer)
Newsgroups: comp.lang.c++
Subject: Re: Eiffel vs. C++
Summary: One language designer's view (part 2)
Message-ID: <151@eiffel.UUCP>
Date: 4 Jun 89 23:54:54 GMT
References: <2689@ssc-vax.UUCP>
Organization: Interactive Software Engineering, Santa Barbara CA
Lines: 101


                    ``Ia pamiatnik siebie vosdvig nierukotvornyi...''
                                Aleksandr Pushkin

                    ``A la septieme fois, les murailles tomberent.''
                                Victor Hugo

                    ``Eppur si muove.''
                                Galileo Galilei

                    ``You can fool, etc.''
                                Abraham Lincoln


From message <2689@ssc-vax.UUCP> by dmg@ssc-vax.UUCP (David Geary):

> I would like to start a discussion of Eiffel vs. C++.

    A previous note described my view of the differences between the
two languages. There is another side to the story, however.
Mr. Geary's message painfully reminded me that the kind of
comparative discussion he calls for
(a desire many other people undoubtedly share)
should normally have been held live next October in a public forum.

    At the 1988 edition of OOPSLA (the ACM conference on
object-oriented programming), a well-known and respected
Professor from a university on the East Coast invited me
to participate in a panel at the next conference
(New Orleans, early October 1989). The idea for the panel was
to discuss the designs of major object-oriented languages, with
the Eiffel, Smalltalk, C++ and Objective-C viewpoints presented
respectively by myself, Dr. Adele Goldberg from ParcPlace, Dr. Bjarne
Stroustrup from AT&T and Dr. Brad Cox from Stepstone. This looked liked a
great idea and I agreed immediately.

    Some time later I was told that I was out of the panel. The reason:
Dr. Stroustrup had declared that he would not participate if I did.
Perhaps even more stunning than Dr. Stroustrup's reaction was the panel
organizer's apparent inference that, if A refuses to be on a panel with B
(who himself welcomes the opportunity), then B should be excluded.
Then as now I did not understand why the inference is not that A must be
taken to his word.

    I do not know the current status of the panel proposal and
have not been contacted again for that panel. A letter and an e-mail
message to the organizer remained unanswered. Clearly there will be
no opportunity for attendees of OOPSLA '89 to hear about
the comparative designs of Eiffel and C++.

    In a way, this whole story confirms that we must be doing
something right. If the developer of C++, backed by the immense
resources of one of the largest and richest companies in the world,
by constant one-sided propaganda in the technical media, by a strong
position in the organization of OOPSLA, and by countless
other assets having to do with the sheer size and wealth of
the supporting organization, can find no better argument
to defend C++ than exclusion of any reference to Eiffel in public forums,
then he must be rather scared indeed.

    There is no doubt that censorship of this nature will have damaging
immediate effects on the recognition of Eiffel's contribution to the
cause of software engineering. The history of such cases over the past
three years amply demonstrates this. But it is a folly to believe that
good ideas can be suppressed for very long. Many people,
some of whom were quoted at the beginning of this note,
have said it much better than I ever could.

    Money, power and intimidation cannot buy everything. Once the lights
are out, the hype is gone, the pre-settled panels have been held and
the biased articles have been read, sooner or later people will go
home and start to work. Then they will judge on costs and benefits,
not promises and politics.

    Among the readers of comp.lang.c++, there are two non-disjoint
groups whose reaction to the present situation is of particular interest.
One is AT&T's employees and particularly its researchers, whose contributions
to computing science have been so impressive over the past 30 years.
Blaming a large organization for the individual behavior of each of
its workers would be unfair and stupid. But when the misdirected
efforts of one or more employees of a multi-billion dollar company
leave the (right or wrong) impression that the company is trying
to crush a group of independent, free-spirited scientists, simply
because they have had another idea and work hard to push it -
I doubt their colleagues feel very proud.

    The other group is C++ users and enthusiasts. Many of them undoubtedly
love C++ for highly respectable reasons and neither I nor anyone else
may ever be able to convince them that Eiffel is better for what they do.
Having everyone agree on any single topic, even if it is just the
programming language - what a horrible prospect! But regardless
of any difference in technical opinions, I would be surprised if
they did not keep open minds. Am I wrong in assuming that they, too,
want to base their decisions on rational arguments clearly expressed,
not on the autocratic suppression of alternative approaches?

-- Bertrand Meyer
bertrand@eiffel.com