panel discussions in the Bay area

[David G. Hough at validgh]

Date: Mon, 1 May 1995 08:45:23 +0800
From: Peter.NurkseaEng.Sun.COM (Peter Nurkse)
 
     + please forward and post as appropriate within the Bay area +

                Bay Area Computer History Perspectives

              "Floating Point---Past, Present, and Future"

                        A panel discussion with

 David Bailey, NASA Ames                   David Hough, Sun Microsystems
 Joel Boney, HaL Computer                  Harsh Sharangpani, Intel
 Jerome Coonen, formerly Apple Computer    Brian Wong, Sun Microsystems
                                 and
                 Professor William Kahan, UC Berkeley

    Note: this program will be presented twice, at Berkeley and at
    Stanford. The two programs will not be identical, and persons
    with a special interest in floating point questions may want to
    attend both (directions at end).

        5:30 PM, Tuesday, May 9   |   5:30 PM, Tuesday, May 23
        Soda Hall                 |   Room 370, Bldg. 370
        UC Berkeley               |   Stanford University
        Berkeley                  |   Stanford


By 1985, the time of the official adoption of the IEEE 754 standard for
Binary Floating-Point Arithmetic, a casual observer might have supposed
that there was nothing left to be done in the field of computer
arithmetic. Yet correctness and performance of computer arithmetic
continue to challenge computer system manufacturers, sometimes even to
the extent of becoming front-page news, with financial implications
that nobody would confuse with roundoff error.

What made computer arithmetic so problematic in the past, and what
issues continue to challenge system designers? Technical matters are
sometimes the least part of the problem.  Various aspects of these
questions will be addressed, from a variety of points of view, by David
Bailey, Joel Boney, Jerome Coonen, David Hough, Harsh Sharangpani, and
Brian Wong.

A commentary on these contributions will then be provided by Prof.
William Kahan, of the Dept of Electrical Engineering and Computer
Science at UC Berkeley, who taught many of the panelists. Prof. Kahan
received the Turing Award from the ACM in 1989, the highest honor in
computer science research, for his research in numerical analysis and
his leadership in developing standards for floating-point computation.
He was a principal instigator and champion of the IEEE 754 standard for
binary floating-point arithmetic. Among the other panel members, Boney,
Coonen, and Hough were also involved in the development of IEEE 754.

------------------------------------------------------------------------

Bay Area Computer History Perspectives is an occasional series of
talks emphasizing events in the history of the local computer industry.
Suggestions for topics and speakers are always welcome, and can be
emailed to Peter Nurkse, nurkseaeng.sun.com, or faxed to Jeanie Treichel,
at 415/691-0756. Next program will be Alan Kay discussing the complete
history of Dynabook, on Tues., Sept. 26, at Sun Microsystems, Bldg. 11,
Willow Rd. at Bayshore Expwy, in Menlo Park. 

------------------------------------------------------------------------

Berkeley directions:

Take the University Ave. exit from interstate 80 in Berkeley, and go
east, toward the hills, about 1.5 miles. Turn left at Oxford St., then
turn right on Hearst Ave. Go up Hearst Ave. past Euclid Ave., and on
your left you will see a green-tiled building, which is Soda Hall.

Stanford directions:

In Palo Alto, take University Ave. west, toward the mountains. The
street will becom Palm Drive, and ends at a large oval in front of the
Stanford main quad.  Walk west to the front of the main quad, and turn
right (north) into the covered walkway that surrounds the exterior of
the quad. Follow it to the north corner of the quad, then turn left
(west), still under the covered walkway. About 50 yards down enter
through the Room 370 doorway on the left side of the walkway (avoid the
main Bldg. 370 entrance just to the left of that doorway).