EMRE TELATAR: Thank
you very much. I never liked hearing my own
voice through a microphone, but such is, I
guess, what we have. So I got an email from
John a few months ago, asking if I would be
willing to introduce Bob-- most terrifying email. It's one of these tasks
in which you can't say no, but you know that you
can't do a good job either, because nothing is
good enough for Bob. Whatever I say is not
going to be enough, but here I am anyway. I mentioned my anxieties
to a few people. One of them was Dave Forney,
Erdal Arikan, Shan-Yuan Ho, John Spinelli. Dave came up with
a good suggestion. He said, well, if you have
these anxieties about Bob, simply talk about me. I ended up not
taking the suggestion for, maybe, obvious reasons. But the best advice,
maybe, came from something that I remember
from Bob telling me. He said that anything that
is worth doing, really doing, is worth doing badly. And my own dictum is that-- don't do today what you can do
tomorrow, recipe for disaster, perhaps. But here I am, so
here's to [INAUDIBLE].. So maybe I should say a few
words about Bob as an advisor. I don't want to go through a
list of CV, publication lists, accomplishments, and so forth. This would be, A, too
long, and then, B, you can read them through
Wikipedia or something. My first day at MIT,
I came to visit Bob. He was my supervisor, and he
had given me an assistantship. So I show up in his
office, and Bob tells me, I've given you this
assistantship, and you're paid, but you should not feel
obligated to work for me. If there is some other
professor that you would be more
interested to work with, you should feel completely
free to work with them, and I will actually
still pay you. So you have your
funding from me, but it is perfectly OK if
I work with somebody else. I was naive enough to
actually take this seriously. I went around, asked my
student colleagues to say, this guy, Bob, he's telling me
that I should work for, maybe, somebody else, so shall
I really go around? They said, you're crazy,
so just stay where you are. And I am glad that they told
me so and I avoided disaster, and I should count my blessings
for having stuck with him. When you work with Bob,
you had this feeling that his time was yours. When you met with him, the
impression that you got was, he had nothing
better to do, nothing more
interesting to do, just to listen to you talking
about your research and trying to help you out
as if the rest of the world didn't exist. And now that I am
an academic myself, I don't know how he did
this, because we are all pressed for time. As Sanjoy Mitter
once said, everybody wants a chunk of your time. And somehow, he
managed to find time for us, all of his
students, and it's the uniformly opinion
of everybody else that I have spoken to-- his time was yours. He also had the
very uncanny ability to make you feel
good about yourself. As a graduate student, as
a researcher in general, it's still true for myself-- my steady state is
that I am stuck. So whatever I am trying to
do, nothing I try works. And not only I'm stuck,
but everybody else seems to be doing very well. So maybe it's me, but I
look around me, I always see people who are always doing
some very successful things, very interesting things. And I am there
feeling like an idiot. And my instinct was to hide
from Bob in these things, and then Bob would
come and find me. After a few weeks of not seeing
me Bob would come and find me and say , let's have a talk. And then Bob would explain
that this was in fact a completely normal
thing, that he himself had the same
anxieties, that he would wake up at night-- maybe I shouldn't
tell this thing. I don't know, but
I will anyway-- that he would say
that tomorrow they'll discover that I am a fraud and
they'll take away my degrees they'll take away my
job and I'll be fired and I'll be exposed. So if you hear as a student
that even someone like Bob had his anxieties, it makes
you calm a lot better. Now Bob is, of course, an
extremely intelligent person, very, very smart. Maybe the smartest
man I actually know. But most smart people-- at least they go through
a phase in which, they try to convince
you that they're smart. You meet with them,
and my god, this is the most irritating thing. I mean, they try to
show how smart they are. Most people grow out of this. Some of them never do. I hope not many of
them are in this room. I don't know if Bob had
ever had such a phase. Bob would make you feel smart. So you would be there hopeless. He would say something, he would
say some encouragement point out some particular aspect that
you are not really thinking of. And then, you will leave the
room with lots of hope saying, oh yes, this is great. I think everything is going
to work out, so I'll push, and then, things will be fine. You'll feel really much
better about yourself, and how he will make
you feel better. I don't know anybody
else who can do this. He also trusts his students. I once asked him, what do you
do with someone like Berlekamp. Because Elwyn Berlekamp
was Bob's student. Obviously, a genius. How do you supervise
such people? And Bob told me, you don't. You simply let them be. You don't guide them,
you maybe touch them left and right a little bit,
not with this thing that we were doing
in the morning with trying to get
the pole sticking up. You try to just very
gently guide, and then trust them that they're
going to do the right thing. I think it was a very
important quality. And I try not to do this with
my own students if I can. His wisdom on publishing
was also very sound. He told me that no one
was going to remember me by the length of my CV,
or the publication lists. We will be remembered
by a few good things that we did in nuggets,
he called them. Whether in this day and age,
this can be actually practiced. I don't know, maybe the
current world of academia is not conducive to this,
or at least something we should strive for. And he would tell us, not
to pollute the journals with the latest random
half baked idea. But wait until we have
something interesting to say. After I graduated MIT,
I went to Bell Labs. And there, I realized
how important Bob was. He was so kind and he was
so generous to everybody. And he had built up
so much goodwill, that this reflected
partly on his students. So when I showed up, I
had instant credibility, which is maybe less important
because I think eventually, you will establish
yourself anyway. But you had instant goodwill. People really want
to take care of you because they know that
you were a Bob student. Because he had somehow, touched
many people's lives there, or I guess, anybody
in our community. And this was, I benefited this
thing from this enormously. I am deeply in his gratitude. One person was very
important at Bell Labs, was Aaron Wyner, who
was my first employer. And Aaron told me
a story about Bob. And I think, late 1960s,
67, John Wiley and Sons had contacted Aaron,
and they sent him a preprint or a prepublishing
copy of Bob's information theory text with
the instructions that he was supposed to
read through this book and find typos, and
he was paid for this. They had given some
significant amount of money, I think to this review job. And Aaron said that, he
felt extremely guilty about this particular
assignment because he couldn't find a single error-- he
went to the whole book, I think, he found
one typo he said, and then he got
paid the full thing. And Wiley was convinced that he
had actually not read the book, and just found some random
stuff and sent it back. This was I think,
how meticulous Bob is when it comes to writing. And he was also
meticulous with us. For his students, he took our
writing very, very seriously. He would make sure that the-- even though nobody was going
to read what we are writing, at least he did. And the writing should
be for posterity, whether somebody reads
it or not, you should you should be very careful. Bob's research, I think, is
driven by a very strong sense of intellectual curiosity. I think, he hears something,
he gets interested, and he wants to
understand whatever he has heard very deeply. I think, his trajectory from
coding information theory networks, later on, wireless,
stochastic processes, communication systems, I
think, all driven by curiosity, and a need to
understand and explain. And when he explains,
he explains it so well that you don't want to
read anything else ever again. Now if I want to
learning anything, if Bob has written about
this I go and look there. And I think it's the best
place that we can do. He looks, for
simplification, unification. He had told me once that
our job as researchers was not to add leaves
to the knowledge tree, but to prune it. To somehow, find that
these five things are, in fact, the same thing. You cut them through, and make
them into a single entity. And that was more valuable
than adding isolated knowledge to the branch. Similarly, teaching
and explanation also drove his pure research too. I asked him once about, how he
came up with this 1965 paper on a simple proof
of decoding theorem. And he said, he has
supposed to teach it a week after in this classroom,
and he was looking desperately to find a good way to
present it to the students. And outcome was precisely
this very simple, very clean, derivation of the theorem. I should also say perhaps, that
among all the award that he received, these include things
like the IEEE Medal of Honor, the [INAUDIBLE] medal. Things which I had in my
mind just before I walked to this room, I
promptly forgot when I'm standing in front of you. Among these things,
the one who admired or many appreciated most,
was a teaching award given by the graduate students. I think, we can do very
well by trying to emulate Bob in some small degree. I tried to do this with a
small amount of success, real little success
but even that I think, makes me a better person
than I would have been, have I never bet Bob myself. And we could all
benefit from his wisdom, and I'm very grateful
for his concern for us, as students, this care for
us, his respect for us. And thank you very much, Bob. Thank you. I think Bob has a rebuttal. ROBERT GALLAGER: Well,
thank you very much Emre for that
beautiful introduction. I wish it were
closer to the truth. But things like this should
be exaggerated a little bit. But I'm glad and I'm
very happy about it. I also want to thank
John and his committee for organizing this whole thing. I'm very grateful to that. But in the few minutes
I have to talk, I want to reminisce
a little bit. As you get older, reminiscing
becomes very important to you. You just love to talk. But since I haven't given
any lectures for a while, I'm going to use notes
to talk and to explain a little bit about what
happened in my career. And in fact, how I
have treated students, and how I view students, and how
I view research because I think that might be worthwhile. I first arrived at
MIT back in 1956, which was a very long time ago. It was mostly a
scam at that point, because I wanted to
get out of the army. I had been drafted as a
private, 2 years before. And found the only
way I could escape, was to go to graduate school. So I thought, well,
I will go to MIT, and I will try to get a
master's degree there. I wasn't very
confident of my ability to do it because
well, I just knew that MIT was a great place, and
I had very little confidence in myself. But I went there anyway. I was working for
Pete Elias and Bob Fano, who were really some of
the best information theorists around at the time. Information theory was
really a hot field back then. And the graduate students going
into it were really remarkable. Most of them knew a great
deal more than I did. And it always affected
me from that time on, that they were very
kind to someone like me who was a little slower,
a little more backward, and had a great deal to learn. But they were so
kind and so decent. So it was a wonderful thing. Well, I went on for a while. I started to try to
do a master's thesis. Bob Fano encouraged
to me in doing it. And I thought, gee,
this is sort of fun. I looked back at it a while ago. It was absolute garbage. But anyway, I felt
good enough after that that I thought, well, why not
try to get a doctor's degree. I mean, there's opportunities
here, MIT at that point, was the center of the
information theory world. Claude Shannon came
a year after I did. Now, that's-- I mean, to say
Claude Shannon came a year after I did, is really
about as--as--[LAUGHTER] But anyway, he had come also
and taught a course here, the term before I got
here, in the fall of 1955. He taught a course,
where every lecture he came in, and delivered
some new result that he'd thought of the night before. And every one of
them was beautiful. He didn't write any of them up. Scribes recorded what he said. But that kind of thing just
led to an atmosphere, which was something you couldn't ignore. It was something,
where I felt well if I can stay on a little
longer, it's worth doing that. I can always go back to Bell
Labs and do switching work. Switching work back then,
was like doing software now. It's something that's
mindless for the most part. Well, actually I
mean, software where you're doing
meaningless things, not software where you're
doing important things. But anyway, I wasn't
ambitious at that point. I don't know whether
I've ever been ambitious. But the intellectual
atmosphere around here was just very addictive,
more so than alcohol, more so than tobacco. I suspect more than cocaine,
although I've never tried that. But anyway, I started out
looking for a doctoral thesis. Finally settled
on a fun problem. It was totally
impractical, it was something called LDPC,
which you were just talking about earlier,
which has been very important 40 years later. I would like to tell you
that I was smart enough to know that that
would become important after a long enough time. I really wasn't. I had no idea that that was ever
going to be a practical thing. I thought it was a fun thing. It was a neat
intellectual puzzle. Peter Lyons encouraged me. So I went on, and finally
did it for my PhD thesis. It was rejected by
the major conference in information theory in 1960. And that was a little
bit discouraging. But anyway, I was offered an
assistant professorship at MIT. I finally realized why. Peter Elias was becoming
department head at the age of 36, he was somebody who was
involved in so many different areas and so widely respected
that he had a chance of holding the whole field together,
which most people didn't. Bob Fano, who had been another
pioneer in information theory was just starting something,
called Project MAC, which was a computer
project dealing with time sharing
computers, which was really what took computation. And turned it from a small field
into a very large, and very active field. So they needed somebody to teach
the information theory course, which is why they hired me. Well anyway, I was very
proud to have been hired. I thought, well, I probably
can't do this very well. But I can always go back to
Bell Labs and do switching work. So well, anyway, I kept doing
research, I kept doing my best. But after I'd been teaching for
about a year, and gradually, slowly building up
my confidence, who walks in the door
but Elwyn Berlekamp. Now, I don't know how many
of you know Elwyn Berlekamp, but he is a certified genius. And always, the only way
I could deal with him, every time he came
in to talk, I wanted him to talk about his thesis. I felt it was important for
him to get his thesis done, and get out and
do something else. He insisted on
playing games with me. He always had a game to play. He always had the
optimal strategy, he could always beat me. And I got used to being beaten. I said, well, this is
the price I have to pay. And I lived with it. And finally, when Elwyn
left, I started to build up my confidence again. And who should walk in
the door but Dave Forney. Who you've just heard
a little bit about it, you'll hear a great
deal more about him. Confidence goes down again,
and I said, oh my god, I can't do this. I am totally lost here. But then Dave graduated
and I said, well this is all right anyway. But I finally, realized, that
I've been extremely lucky because I learned how to be
a decent professor in all of that. And the answer was, you
don't teach students. What you do, is you help
them, you support them, and you try to bring
them to the point where they can learn things. And basically, you realize
that you and they are working together as learners. And in a place where
the other people there are trying to learn also. Well, I had an enormous number
of really fantastic students after that. All of them much
smarter than I was. People like Emre who
you just talked to, David over there who's going
to give it a talk later, Erdal, who has invented
polar codes recently, and which is the hot
thing in coding now. Many, many wonderful students. But I have always found that
the way to deal with students is to say, I am so lucky to have
this great person to work with. And that is the secret to
being a good professor. People claim these
days that there's a dichotomy between doing
research and teaching. That's a very false dichotomy. Doing research is
learning in one sense. Not really learning. Doing research is trying
to understand things. And teaching, is not
what people think of from the old days
of going up to a board, and writing on it in a pre
15th century mode of thinking. That you write things,
somebody else copies them down. And that's the way
learning is passed on. You can't teach if
you're not also trying to understand things yourself. So teaching is also the process
of a community of people, trying to understand things. What you're trying to understand
is something we all grope with. We move from one
thing to another. I had started out
in RLE which is where most of the communication
was being done back in 1960. I then moved to LIDS
in the middle of 1970. And that is where the networking
was starting to be done. It's where there was this
beautiful community of people, interested in optimization,
interested in mathematics, interested in communication,
interested in control. And all of that
just worked together to make this community,
where all of us could understand things. And working with students there
were these delightful times when you woke up in
the morning, and said, the light has gone on. And every once in a
while, the light goes on, and that's what keeps you going. And the one thing I miss
being in Florida now, is not having this wonderful
community around me of other people who are also
trying to understand things. Thank you.
