Announcement

Programming Tools, Technology, and Teaching the Masses to Fish

Daniel Weise

Microsoft Research

How do we get ideas and methods from the last three decades of programming

language research into the daily practice of commercial programmers? New

mainstream languages only appear once a decade, so language design is out.

Generic technology is also a non-starter, as the invented technology all too

easily misses the mark, being either too abstract or too removed from the

problems at hand. Our research group has taken the approach of giving

programmers in the product groups the ability to do their own

meta-programming-the ability to extend their programming languages and to

construct programs that themselves analyze and construct programs. This

approach allows good language ideas to be re-invented and implemented by

practitioners, and changes the relationship that practitioners have with

their programming language and development environment. In this talk I

describe the initial results of this experiment, with a focus on how product

groups have been using meta-programming to write simple verification tools

that have become valuable to them.

(Daniel Weise was born and raised in Reseda, CA, which gained fame as the

town in The Karate Kid, and which has been mentioned in songs by Frank Zappa

and Tom Petty. He left home for Cambridge, MA to wheedle a couple of

post-graduate degrees from MIT. Then he moved to Palo Alto to abuse his

position on the faculty at Stanford to corrupt impressionable students with

functional programming ideas. Daniel now lives in Washington state, and is

amazed to learn that there is a place in the US with more cloudy days per

year than Seattle, namely, Ithaca, NY. Is his spare time, he does research

on program analysis, program comprehension, program correctness, and static