June 28th, 2004

classic cylon

(no subject)

I've gotten through about half of the 6.829 lectures. It's been fun
reacquainting myself with the subject matter -- much more fun than
working on Right Management modules or workshops. (As an aside, I put
my program on hold for a week, and am thinking of extending it to a
month.) After I finish the lectures, I'll hopefully have enough
background to contribute to one of the network performance-related
IETF working groups.

One of the lectures covers Van Jacobson's slow-start algorithm for
TCP. It's interesting that there is now a large body of research that
sprung from his work. I remember when he first came up with the idea
back in 1987. There had been lots of congestion in the Internet at
the time, and it wasn't very well understood what to do about it. His
solution was not universally adopted right off the bat. He had to
strongly persuade the 4.3 BSD Unix developers to incorporate his

When I first read his arguments on the IETF mailing list, I was
immediately excited because I understood everything he was talking
about. (It was like being in 6.003 again.) It started me thinking
about going to grad school, because I wanted to learn more about how
to do that type of analysis, and how to apply it to other types of
networks and traffic. I also figured that with the Internet growing,
it would be a useful skill.