IEEE Solid-State Circuits Magazine - Summer 2016 - 24

While I continued doing design for fun,
my largest project was for my digital
design lab at MIT.

derivation. Since I had known Paul
Penfield while I was at MIT, I wrote
to him and asked how they got they
answer. I needed to understand how
to do that step to extend their model
to bound the delay of MOS transistor
capacitor networks.
The answer was disappointing. He
told me that they didn't know how to
do the forward step, but they thought
the answer should be an exponential
and could show that the exponential
works. This answer was no help at all.
So, for a couple of weeks, I took the
derivation apart in every way I could
think of, trying to find some approach
that would work. Persistence finally
paid off. I found a derivation for the
forward path and with it a way to generate bounds for MOS networks too.
Interestingly, this new derivation
allowed me to improve the generated
bound slightly, so I excitedly wrote
to Paul explaining my new derivation and my new improved bounds.
You can image my shock when Paul
wrote back telling me that he had
improved my bounds even further.
When I looked at his results, they were
changes that I had already figured out
but hadn't sent. These were secondorder effects, and I didn't want to
confuse my initial message by adding
these too. My initial reaction was that
Paul was "stealing" my research result.
But I knew Paul and that made no
sense at all. He had always been very
kind to me and helped me in any way
he could. The person I knew would
never steal anything.
I started to analyze this situation
trying to find an explanation, just as
I studied RC network problems trying to find a way that made the solution clear. Soon the answer became
obvious. Paul had invented those
tighter bounds, since I didn't tell him
about them. We both invented them,
since we both independently figured

s u m m E r 2 0 16

out how to get them, and there was
nothing wrong with that. What was
silly was arguing over whose invention it was.
I know many people who have gotten into big fights about who "invented"
something first or are scared to share
results for fear that someone will
"steal" them. I was so lucky to have
this episode very early in my career
with a colleague I completely trusted
since it showed me the fallacy and
danger about arguing over these types
of issues. In the end, Paul and Jorge
were very generous and made me a
coauthor on their final paper, and we
were all much richer for the collaboration. As a result, I have always been
very open about my research ideas
and never argue over whose idea it
was. I have found that almost everyone is honest, and most ownership
issues are really a matter of simultaneous invention. That is not to say that I
haven't run across dishonest people; I
have. But I have a simple solution for
dealing with them-I just never work
with them again-so I still get to enjoy
working with most people.

How I Became a Professor
My older brother was a theoretical
physicist. He wanted to become a professor. I like engineering electronics
and never considered being an academic. I went Stanford since I liked
school, not to become a teacher. But
a funny thing happened on my way
to graduating. A fellow student and
former colleague at Stanford, John
Redford, graduated and started working at Digital Equipment Corp. (DEC)
in the Boston area. They had built a
MOS fab in Hudson, Massachusetts,
and were starting to design their own
custom ICs for their minicomputers. To help train their engineers in
chip design, they wanted to give an
intensive two-week Mead-Conway

IEEE SOLID-STATE CIRCUITS MAGAZINE

VLSI design class and had lined up
Chuck Seitz to teach the initial class.
DEC was also looking for people who
might be interested in teaching it in
the future. They wanted those people
to attend and be teaching assistants
for this initial class. John suggested
that I might be a good candidate. DEC
initially reached out to me and asked
me if I would be willing to come out
and offered to pay for my trip. I agreed
and made plans to attend the class
and then visit my family in Maryland.
A little while later, they called me back
and told me that other, more senior
people expressed interest in teaching
the class, and they no longer needed
my services. Ouch!
Fortunately for me, I don't stay
angry very long, and I can't keep a
grudge to save my life. So after being
upset for a short period of time, I
calmed down and thought about the
situation rationally. Chuck Seitz had
this amazing reputation as a teacher, I
really loved VLSI design, and I already
decided to see my family so I would
be out east anyhow. I thought why not
just ask if I could attend the class if I
paid my own way. While it would be
many years before I joined a startup
and heard the lingo of upside potentials and downside risks, it is a perfect way to describe my reasoning at
the time: the downside risk of going
(if they would let me) would be just
be more bruising of my ego, while the
upside potential was large-I could
interact with a VLSI superstar and
meet real VLSI designers at DEC. So I
called them back and asked if I could
attend if I paid my own way. I think
my offer shocked the person I talked
to, and she said yes. In hindsight, I
think the person I spoke with probably felt she didn't have much choice
because one of the reasons for canceling my invitation was because of
constrained funding.
So I went to the class, and it was
another life changing experience. First
I got to meet and interact with Chuck
Seitz, who ended up being a mentor to
me. One of the high-points of the trip
was that he taught me to juggle, which
I still do for fun. It also introduced me

Table of Contents for the Digital Edition of IEEE Solid-State Circuits Magazine - Summer 2016