Symposium on Software, Science and Society

Department of Computer Science and Engineering
_______________________

HOME
Friday, December 5, 2003

AGENDA

SPEAKERS

TESTIMONIALS / COMMENTS

DICK'S ACCOUNT of CSE's early history

DIRECTIONS to World Forestry Center, 4033 S.W. Canyon Road, Portland, Oregon, including MAX information. Event is in Cheatham Hall.

REGISTER ONLINE
______________________
SPEAKER PROFILE
HALF-DAY TECHNICAL SYMPOSIUM ON
SOFTWARE, SCIENCE & SOCIETY

Friday, December 5, 2003
Cheatham Hall • World Forestry Center • Portland, Oregon
SIMON PEYTON JONES
Microsoft Research Ltd.		  

BIOGRAPHY:

Simon Peyton Jones is a researcher at Microsoft Research, Cambridge, where he has been since 1998. Prior to that, he spent nine years as a Professor at the University of Glasgow. His research interests are centred on the design, implementation, and application of programming languages, especially functional languages such as Haskell. He leads the team that develops and supports the open-source Glasgow Haskell Compiler (GHC).
ABSTRACT:

Functional programming languages, one of Dick's enduring interests, are regarded as niche languages, used only by a minority. But in fact the situation is quite the reverse: there are probably more functional programmers in the world than imperative or object-oriented programmers. Why? Because a spreadsheet is a functional programming language.

Cells have values given by formulae, and the spreadsheet computes the order in which to execute, or rather evaluate, the "program". Spreadsheets are programmed by end users (physicists, accountants, teachers, engineers), not by professional programmers. Indeed, they would not describe themselves as "programming" at all; rather, they are "building a model". Only a small minority of these end users make substantial use of formulae, but that minority is still extremely numerous bunch of experienced functional programmers.

Regarded as a programming language, however, spreadsheets are extremely weak, supporting only scalar data and providing no means of abstraction. In my talk I'll sketch a design for adding user-defined functions and bulk data types to Excel. These relatively simple features could dramatically extend the "reach" of end-user programmers.