1
00:00:00,000 --> 00:00:07,527
I'll show you the Berners-Lee footage,
it's I have to tell you that I'm a giant
2
00:00:07,527 --> 00:00:12,338
fan of yours. ≫> Oh really? ≫>
Yes I am. I mean I don't know and I've
3
00:00:12,338 --> 00:00:18,558
never met you but I, it seems to me as
though maybe, maybe I'm just, it's just
4
00:00:18,558 --> 00:00:22,290
the Larry Smarr image rather than the
real, Larry Smarr. ≫> You'll have to
5
00:00:22,290 --> 00:00:26,693
tell me what that is. [laugh] >>
>> I'm the one person that never
6
00:00:26,693 --> 00:00:31,044
gets to see that. ≫> The Larry Smarr
image? ≫> Yeah, right. ≫> When
7
00:00:31,044 --> 00:00:35,670
you're inside it. ≫> Yeah. I didn't
know you were an astronomer. You know Bob
8
00:00:35,670 --> 00:00:39,361
Stein by any chance? ≫> Sure.
≫> Bob Stein is my mentor. ≫>
9
00:00:39,361 --> 00:00:43,356
Really? Oh yeah, we gave Bob lots of time.
≫> Yeah, I heard he's like in top
10
00:00:43,356 --> 00:00:47,758
five here, ob is, Bob is the guy that
taught me high-performance computing.
11
00:00:47,758 --> 00:00:51,262
≫> I go way back with him. ≫>
And >> Tom Wolf used to be a good
12
00:00:51,262 --> 00:00:54,725
friend of ours, before he had his stroke
or. ≫> Yeah. ≫> Heart attack,
13
00:00:54,725 --> 00:00:59,479
or whatever it was. ≫> And but I
mean to me it, after. ≫> So we don't
14
00:00:59,479 --> 00:01:04,881
get disrupted. Why don't you take this and
put it over there? Tell Tira to hold my
15
00:01:04,881 --> 00:01:09,204
calls and it's, it's the why an astronomer
got into this. ≫> Yeah, go ahead.
16
00:01:09,204 --> 00:01:13,094
Right. ≫> Okay. ≫> I mean tell
me, tell me how this NCSA started? I mean,
17
00:01:13,094 --> 00:01:16,715
the, the whole purpose of this. ≫>
Mm-hm. ≫> Is, is, it's sort of about
18
00:01:16,715 --> 00:01:23,062
NCSA. It's not as much about Netscape and
Mosaic. I mean it, to me I'm curious,
19
00:01:23,062 --> 00:01:27,039
[inaudible] on how, why it happened here?
≫> Right. ≫> Which is, I think
20
00:01:27,039 --> 00:01:31,575
>> Yeah. I can tell you about that.
≫>. Okay. So go ahead. Tell me about
21
00:01:31,575 --> 00:01:35,665
NCSA. ≫> Okay. Does the little red
light go on when you're? ≫> Yep. I'm
22
00:01:35,665 --> 00:01:41,524
recording. ≫> Okay good. Well, I am
a trained relavistic astrophysicist. And I
23
00:01:41,524 --> 00:01:47,056
got my P.h.D. Always in physics
departments, working on astrophysics
24
00:01:47,056 --> 00:01:53,091
problems involving generalitivity, gas
dynamics, and so forth. So in the 70s,
25
00:01:53,091 --> 00:01:58,777
when I was developing what is now called
numerical gen eralitivity, which is the
26
00:01:58,777 --> 00:02:04,395
way to solve Einstein's equations for
dynamics of black holes, colliding black
27
00:02:04,395 --> 00:02:10,159
holes, or an astrophysics that require
generativity like supernova events and so
28
00:02:10,159 --> 00:02:15,695
on. I had to go and get a top-secret
nuclear weapons clearance to get access to
29
00:02:15,695 --> 00:02:22,746
super computers, because in the mid 70s
the only place to get a supercomputer was
30
00:02:23,003 --> 00:02:28,097
either at Livermore, Los Alamos, or in one
area which was weather. Now astronomy is
31
00:02:28,097 --> 00:02:33,919
always been a driver or the super
computing and in fact on Johnny von Neumann's
32
00:02:33,919 --> 00:02:41,023
computer that was built by the institute
half of that went to army to Aberdeen.
33
00:02:41,023 --> 00:02:46,380
Half of that was used for the army for
trajectory calculations but the other half
34
00:02:46,380 --> 00:02:51,081
was used for stellar revolution. So
astronomy and von Neumann's interest in
35
00:02:51,081 --> 00:02:58,710
weather meant that astronomy and weather
for 50 years basically have been dominant
36
00:02:58,710 --> 00:03:04,858
drivers of supercomputing usage. And yet
to do pure astrophysics. You're having to
37
00:03:04,858 --> 00:03:08,473
get a top secret nuclear weapons
clearance. Now, nobody seemed to think
38
00:03:08,473 --> 00:03:13,815
there was anything unusual about this. And
so I just went ahead as a post doc and I
39
00:03:13,815 --> 00:03:18,143
would get a few months in the summer, work
100 hours a week. And then, the rest of
40
00:03:18,143 --> 00:03:22,906
the year I'd had to live off of that. Go
back to Harvard where I was a junior
41
00:03:22,906 --> 00:03:28,038
fellow. Try to explain to them about, you
know one could solve the laws of physics
42
00:03:28,038 --> 00:03:32,872
that we had been laying down for 300 years
in incredible detail. For engineering
43
00:03:32,872 --> 00:03:37,413
devices like nuclear weapons. That put on
Earth temperatures of the center of the
44
00:03:37,413 --> 00:03:42,628
sun stresses beyond anything that we could
imagine in academic problems that we were
45
00:03:42,628 --> 00:03:46,755
trying to solve and lots of disciplines.
So, this thing could revolutionize
46
00:03:46,755 --> 00:03:51,718
academic research. Well, nobody got it. I
mean it was like, I really felt like I was
47
00:03:51,718 --> 00:03:56,524
transitioning in a flying saucer between
this advanced civilization at Livermore,
48
00:03:56,524 --> 00:04:01,586
and the stone age culture at Harvard, and
Harvard was as advanced as any place in
49
00:04:01,586 --> 00:04:06,319
thinking about this, so it was a, what, I
didn't figure this out until I was in the
50
00:04:06,319 --> 00:04:11,136
early 80s and by then the fir st cray had
gone into the continent of Europe at an
51
00:04:11,136 --> 00:04:16,549
open scientific institute, the Max Planck Institute
of Physics and Astrophysics. So
52
00:04:16,549 --> 00:04:21,126
I'm over there in the summer along with
people like Dave Arnet, who was one of the
53
00:04:21,126 --> 00:04:25,238
great supernova supercomputer guys in our
country and, people in chemistry and it
54
00:04:25,238 --> 00:04:29,308
was like Paris in the 20s with all these
expatriates sitting over there. And we're
55
00:04:29,308 --> 00:04:33,934
like trying to figure this out, like this
is an American built supercomputer, right.
56
00:04:33,934 --> 00:04:40,055
Why are we in Munich, right? I mean this
is very strange so but, you know in
57
00:04:40,055 --> 00:04:46,033
America we don't question the
infrastructure somehow. I mean it's just
58
00:04:46,033 --> 00:04:51,576
like its either there or it isn't there. That's
just the way it is but I was having a mas [German word for mug?]
59
00:04:51,576 --> 00:04:56,897
of beer like one night. Actually I think
it was the second mas of beer with my
60
00:04:56,897 --> 00:05:02,685
German host who had also been born like I
was, postwar and he finally turns to me
61
00:05:02,685 --> 00:05:07,226
and he says, aren't you ashamed of
yourself? You big, rich occupying country.
62
00:05:07,226 --> 00:05:11,818
You come over here in our little country
and, and we finally get enough money
63
00:05:11,818 --> 00:05:16,995
scraped together after World War II to
buy one of these supercomputers, and you
64
00:05:16,995 --> 00:05:21,449
Americans come over here and use up our
time. He says, how did you guys ever win
65
00:05:21,449 --> 00:05:26,523
the war? You know? What is going on here?
And this finally just sort of stimulated
66
00:05:26,523 --> 00:05:31,503
me into saying, what is going on here?
This is nuts. And I went back and I found
67
00:05:31,503 --> 00:05:35,863
out, for instance, that after the Sputnik
Program, the Federal Government had funded
68
00:05:35,863 --> 00:05:40,293
the universities, built the science
buildings, started the supercomputer
69
00:05:40,293 --> 00:05:45,174
centers. IBM would go around and give away
almost the mainframes and, and so the
70
00:05:45,174 --> 00:05:50,207
scientists in the 60s took it for granted
that they had the fastest computers in
71
00:05:50,207 --> 00:05:54,842
the world in academia. But about 1970, with
the starting of the Vietnam War, and with
72
00:05:54,842 --> 00:05:59,696
all kinds of guns and butters issues and
everything else, that stopped. And in
73
00:05:59,696 --> 00:06:04,105
fact, to give an example, by 78, half the
number of PhDs and engineers in
74
00:06:04,105 --> 00:06:09,695
engineering was being generated in our
country as there were in 1970. So, there
75
00:06:09,695 --> 00:06:16,124
was this complete severing of the Sputnik
era partnership with the federal
76
00:06:16,124 --> 00:06:21,625
government and it was particularly bad in
computing. To give you an example, when I
77
00:06:21,625 --> 00:06:27,715
was at Livermore in the 70s, there
were four CDC 7600s, which were just one of
78
00:06:27,715 --> 00:06:33,034
the finest supercomputers ever built. No
American university ever took delivery of
79
00:06:33,034 --> 00:06:38,048
a single CDC 7600. In fact, the University
of Illinois, when I first got here, had a
80
00:06:38,048 --> 00:06:43,034
Cyber 175 which was a retread, a second
design manufacturing of, of this thing.
81
00:06:43,034 --> 00:06:48,046
And we were one of the first universities
that had it, and people all thought that,
82
00:06:48,046 --> 00:06:53,051
you know, Illinois naturally was way ahead
of everybody else. So it's like, we were
83
00:06:53,051 --> 00:06:58,082
just completely divorced from the private
sector that was generating these wonderful
84
00:06:58,082 --> 00:07:03,087
machines, because of federal policy, which
was to say, these things are only, are so
85
00:07:03,087 --> 00:07:10,086
valuable that we can only afford to put
them into war environments. So, after this
86
00:07:10,086 --> 00:07:16,734
German encounter I came back and I said, I
wonder how many other scientists like me
87
00:07:16,734 --> 00:07:21,873
are there? So, at the University of
Illinois, I started calling, cold calling
88
00:07:21,873 --> 00:07:26,837
my colleagues, and saying hi, you don't
know me. I'm a little assistant professor
89
00:07:26,837 --> 00:07:31,440
but I bet you that your research is
blocked by lack of access to
90
00:07:31,440 --> 00:07:36,288
supercomputers, and they'd sort of say
like, who? Crack, crack. Who is this? You
91
00:07:36,288 --> 00:07:41,270
know, [laugh] crank call and but we'd
start talking in chemistry and biology and
92
00:07:41,270 --> 00:07:46,043
agriculture and so on and, and sure
enough, it turns out that, that, that was
93
00:07:46,043 --> 00:07:50,293
true. They knew how to do the science,
they just didn't have access. So I said
94
00:07:50,293 --> 00:07:55,611
well, send me a little prospectus of what
science you could do if you had a
95
00:07:55,611 --> 00:08:00,625
supercomputer. Well, I ended up with 65
faculty and fifteen departments on one
96
00:08:00,625 --> 00:08:05,693
campus and I thought, this is going to be
all this way all over the country so it
97
00:08:05,693 --> 00:08:11,163
was, I really said well jeez somebody has
got to raise this issue, And about that
98
00:08:11,163 --> 00:08:17,452
time there had been a Lax Report, that had
the federal government had done that began
99
00:08:17,452 --> 00:08:23,096
to undercover of the stuff and they still
weren't, I mean Peter Lax. He was one of
100
00:08:23,096 --> 00:08:30,041
the greatest mathematicians, head of, one
of the top people in the [inaudible] Institute, a long time
101
00:08:30,041 --> 00:08:35,456
advocate of things computational, and I had
this long, long battle with Peter, because
102
00:08:35,456 --> 00:08:39,574
in the draft report of the Lax report, it
was not, it was going to say, well yeah,
103
00:08:39,574 --> 00:08:44,117
we oughta get one of these supercomputers
and make it available to the universities
104
00:08:44,117 --> 00:08:48,988
but you know, let's put it in Livermore or
put it somewhere that people know how to
105
00:08:48,988 --> 00:08:52,894
do this stuff. And a university was not
on the list of what the Lax report
106
00:08:52,894 --> 00:08:57,357
considered to be appropriate sites for a
supercomputer. So I had this long battle
107
00:08:57,357 --> 00:09:01,923
over the telephone, I remember, with Peter
Lax. It was sort of David and Goliath cuz
108
00:09:01,923 --> 00:09:06,868
I mean he was this giant of the field, I'm
nobody. And I, I finally convinced him not
109
00:09:06,868 --> 00:09:12,103
to exclude universities as a possibility.
Even though he felt that it was fairly
110
00:09:12,103 --> 00:09:17,965
unlikely that any of them would be able to
play with sharp instruments and not hurt
111
00:09:17,965 --> 00:09:23,177
themselves. So, I mean, you got to
understand the world was totally different
112
00:09:23,177 --> 00:09:28,917
when the supercomputer center's program
was coming into being. And people, it's so
113
00:09:28,917 --> 00:09:33,900
hard for people now on the web and
everything to go back to that time. There
114
00:09:33,900 --> 00:09:38,771
was no Internet, you know? There was,
there was these wonderful people Vince
115
00:09:38,771 --> 00:09:43,412
Cerf and Bob [inaudible] and all
these people. Bill Joy, who had developed
116
00:09:43,412 --> 00:09:47,773
TCP/IP, and, and, and embodied it in the
ARPANET. This was a few computer science
117
00:09:47,773 --> 00:09:52,556
departments and military, okay? Nobody in
the physics department or chemistry
118
00:09:52,556 --> 00:09:57,487
department ever heard of the ARPANET, much
less had any access to it. But it was
119
00:09:57,487 --> 00:10:02,666
obviously the right idea. And so once we
got the Congress to put through money for
120
00:10:02,666 --> 00:10:07,888
a national supercomputer program, then
there was a national competition, and
121
00:10:07,888 --> 00:10:12,954
so on, and the five centers were selected.
The first thing then that NSF realized.
122
00:10:12,954 --> 00:10:18,660
Well okay, now we put these in place to be
providing access to academic scientists,
123
00:10:18,660 --> 00:10:23,619
and yet, like, they have to fly to
Champaign-Urbana like I had to fly to
124
00:10:23,619 --> 00:10:28,690
Livermore, okay? This isn't right so there
were a lot of discussions then but the
125
00:10:28,690 --> 00:10:33,200
trouble was that the telecom lobbyists in
Washington would block any discussion of
126
00:10:33,200 --> 00:10:37,813
the Federal Government putting in the kind
of network we have today, which is what
127
00:10:37,813 --> 00:10:42,380
people wanted. I mean everybody knew that
they wanted to have a ubiquitous email
128
00:10:42,380 --> 00:10:46,805
person to person network. But as soon as
they'd start talking about that the
129
00:10:46,805 --> 00:10:51,268
telecom lobbyists would come up and say,
no way guys. That's private sector. Don't
130
00:10:51,268 --> 00:10:56,176
get the Federal Government involved in
that. Stay out of it. So what we learned
131
00:10:56,176 --> 00:11:00,179
early on was real interesting. It's like
if we had argued instead of the
132
00:11:00,179 --> 00:11:03,780
supercomputer program, let's get the
Federal Government to buy everybody a
133
00:11:03,780 --> 00:11:08,328
personal computer which was IBM personal
computer was two years old in 1985 and put
134
00:11:08,328 --> 00:11:13,172
on people's desk, okay? Again, this would
be an interference with the private
135
00:11:13,172 --> 00:11:18,501
sector. So but we said oh, we'll just take
a few of these esoteric supercomputers.
136
00:11:18,501 --> 00:11:23,081
And, is that okay? That's right, there's
no market there, that's okay, the Federal
137
00:11:23,081 --> 00:11:27,175
Government has a role. Well, it was the
same thing with networking. What we said
138
00:11:27,175 --> 00:11:31,661
is, we just want to put a high speed
backbone across the country to connect the
139
00:11:31,661 --> 00:11:35,979
five centers and, and the telecom people
said, okay, that's cool, you know 56
140
00:11:35,979 --> 00:11:40,110
kilobits was the national high speed
backbone. Less than ISDN today yeah.
141
00:11:40,110 --> 00:11:43,875
That's not a market, okay? And, and, and
so we got a few of these weird
142
00:11:43,875 --> 00:11:48,429
supercomputer types who are out in the
universities who want to hook into
143
00:11:48,429 --> 00:11:53,027
that. They say, yeah that's okay. That's
not a market. You can do that. Well, that
144
00:11:53,027 --> 00:11:56,862
was the NSF Net backbone. Then the
regionals got funded. And then the
145
00:11:56,862 --> 00:12:02,168
campuses were afraid if they didn't dig up
their quad and put in some fiber then the
146
00:12:02,168 --> 00:12:08,072
professors who want to get access to
supercomputers would go to a university
147
00:12:08,072 --> 00:12:15,929
that would do that, so gradually the whole
Internet emerged out of the, the sort of
148
00:12:15,929 --> 00:12:22,595
policy vice you get into Washington
where you can't do the right thing. You
149
00:12:22,595 --> 00:12:28,570
have to do something that seems
irrelevant, but has a logic to it that will
150
00:12:28,570 --> 00:12:33,871
gradually bring the market forces into
play. That will spin out, ultimately a
151
00:12:33,871 --> 00:12:35,080
whole industry.