Systems Software Research is Dead
October 20, 2009 at 8:50 am 3 comments
Beki and I continue to have a conversation through our blog posts. (Normally, Beki is down the hall from me, but this semester, she is teaching at Georgia Tech’s campus in Metz, France. Barb and I are visiting Beki and her husband, Keith Edwards, when I go to speak at Informatics Education Europe the first week of November. Perhaps we should do a joint blog post that day…) Today, Beki is talking about Rob Pike and his concerns about “science” taking over what’s interesting about “Computer Science.”
Measurement as a “misguided” focus on science, but then he adds:
“By contrast, a new language or OS can make the machine feel different, give excitement, novelty. But today that’s done by a cool Web site or a higher CPU clock rate or some cute little device that should be a computer but isn’t.
The art is gone.
But art is not science, and that’s part of the point. Systems research cannot be just science; there must be engineering, design, and art.”
Entry filed under: Uncategorized. Tags: meaning of computer science.
Computing Ed Research - Guzdial's Take 
1.
Alan Fekete | October 20, 2009 at 3:04 pm
But Rob’s paper comes from 2000! In the years since then, there has been a tremendous resurgence of interesting and important research work doing innovative designs in systems software; not focused mainly on the desktop operating system, but instead on debugging and fault-tolerance (this years hot area at SOSP according to http://matt-welsh.blogspot.com/2009/10/sosp-2009-day-two.html), infrastructure for cloud computing (eg the Amazon Dynamo paper which was at SOSP07), peer-to-peer infrastructure (which Welsh says is now seen as “so-2003”), even sensor networks.
2.
beki70 | October 20, 2009 at 4:10 pm
Hi Alan, That’s a very fair comment… and it was designed to be a provocative paper. I pulled out points, one of them is the one that Mark refers to, I don’t know whether we’ve made much progress on embracing an art, design, engineering and SCIENCE of systems. I think that there’s still merit in this comment.
3.
Mark Miller | October 21, 2009 at 8:54 am
At the Rebooting Computing summit Gene Spafford referenced a blog post he wrote that sounds a similar note to Rob Pike’s article, except he approached it from the angle of system security. It’s called Rethinking Computing Insanity, Practice and Research. He had a novel idea that we should return to (I believe he said) a design practice that existed in the 1960s, where computer systems were designed for specific purposes, rather than as general-purpose systems. He said that general-purpose systems introduce security risks that do not exist in system with a more focused purpose. Different uses carry with them different avenues for attack, and so it’s easier to design special-purpose systems such that they are secure.