cfry@watdcsu.waterloo.edu (C.Fry - Inst. Computer Research) (05/03/88)
The Future of UNIX in the Year 2001
by
Michael Tilson
of
HCR Corporation
130 Bloor Street West, 10th Floor
Toronto, Ontario M5S 1N5
Abstract
UNIX has been available outside Bell Labs since the mid-70's; the
first UNIX system at the University of Toronto was installed in
early 1975. Thirteen years ago the system was new, still experi-
mental, and rarely used. Today, UNIX is mature, becoming stand-
ardized, and widely used. In thirteen years we will see the dawn
of a new (calendar) millenium. What can we expect from UNIX?
This colloquium discusses the technology trends that will deter-
mine the status of UNIX at the turn of the century, with special
reference to the role UNIX may play in ``wiring the world''.
UNIX has become a standard. The lifetime of standards is
surprisingly long. On the other hand, technology continues to
advance at a rapid rate. There is no reason to believe that the
rate of change will slow between now and the end of the century.
However, there are good reasons to believe that ``conventional''
architectures will continue to dominate general-purpose computing
for the next decade, and if this is true, then UNIX has a long
and bright future as a higher-level machine definition.
In the next thirteen years UNIX will open the door to possibili-
ties for distributed processing and distributed applications that
go far beyond anything we can do today. In particular, LAN net-
working issues will be solved soon, but long haul connectivity of
diverse international systems presents some new technical chal-
lenges. UNIX, standards, high performance machines, networking
(especially ISDN), and ``internationalization'' will combine to
create a basis for practical, everyday world-wide network comput-
ing.
A goal of this talk is to understand why a typical obsolete C ap-
plication written in the mid-80's might be still running on an
incredibly advanced networked architecture, moving data tran-
sparently from New York to Tokyo in the year 2001, and why that
might actually represent a good solution.
This talk is based upon a paper presented at the Summer '87
Usenix Conference and new material presented at the recent Uni-
Forum '88 conference.
DATE: Wednesday, May 11, 1988
TIME: 3:30 p.m.
PLACE: DC 1302
Everyone is welcome. Refreshments served.