WEBVTT

00:00:00.000 --> 00:00:02.520
When VisiCalc first came out, you could explain

00:00:02.520 --> 00:00:06.639
it or show it to people, okay? And if you showed

00:00:06.639 --> 00:00:09.400
it to a computer person, they look at it and

00:00:09.400 --> 00:00:11.820
say, what's so special here? I can write a program

00:00:11.820 --> 00:00:14.099
that does that and I can do better and stuff

00:00:14.099 --> 00:00:17.739
like that. You show it to a regular person, a

00:00:17.739 --> 00:00:19.420
normal person, they say, computers can do anything.

00:00:19.480 --> 00:00:21.399
They can forecast the weather. What's so special

00:00:21.399 --> 00:00:23.859
about that? But if you showed it to people who

00:00:23.859 --> 00:00:27.420
actually did that stuff, They would start shaking

00:00:27.420 --> 00:00:28.940
and say, here's my credit card. Please take it.

00:00:28.940 --> 00:00:34.579
Give me that. Give me that. How did the best

00:00:34.579 --> 00:00:36.560
machine learning practitioners get involved in

00:00:36.560 --> 00:00:40.479
the field? What challenges have they faced? What

00:00:40.479 --> 00:00:44.679
has helped them flourish? Let's ask them. Welcome

00:00:44.679 --> 00:00:47.560
to Learning from Machine Learning. I'm your host,

00:00:47.759 --> 00:00:51.039
Seth Levine. Hello and welcome to Learning from

00:00:51.039 --> 00:00:53.200
Machine Learning. On this episode, we have a

00:00:53.200 --> 00:00:56.539
very special guest. Dan Bricklin, the co -creator

00:00:56.539 --> 00:00:58.859
of VisiCalc, the first electronic spreadsheet

00:00:58.859 --> 00:01:01.740
and really the first killer app that helped launch

00:01:01.740 --> 00:01:05.280
the PC revolution. Dan has lived through every

00:01:05.280 --> 00:01:07.900
major computing platform shift from mainframes

00:01:07.900 --> 00:01:11.280
to today's AI. Few people have a better perspective

00:01:11.280 --> 00:01:13.599
on what separates transformative technology from

00:01:13.599 --> 00:01:17.200
hype. Dan, welcome to the show. I'm glad to be

00:01:17.200 --> 00:01:19.420
here. Thanks a lot, sir. It's such a pleasure

00:01:19.420 --> 00:01:23.140
to have you. Let's get right into it. So what

00:01:23.140 --> 00:01:26.340
initially attracted you to computer science,

00:01:26.640 --> 00:01:28.959
computers in general? Well, as a child, this

00:01:28.959 --> 00:01:33.359
is back in the 50s and 60s. I was, you know,

00:01:33.640 --> 00:01:36.459
I love technology and, you know, mechanical stuff

00:01:36.459 --> 00:01:38.739
and electrical, electronic stuff and whatever.

00:01:39.420 --> 00:01:42.120
And I would read Popular Electronics magazine.

00:01:42.239 --> 00:01:45.579
That was my one of my favorites. So a computer

00:01:45.579 --> 00:01:48.010
seemed like I didn't know what it was really,

00:01:48.030 --> 00:01:49.750
you know, how they worked or how to make them.

00:01:50.170 --> 00:01:53.129
But I knew that pretty cool stuff. And I knew

00:01:53.129 --> 00:01:56.189
some of the electronics of how to do whatever.

00:01:56.870 --> 00:01:59.730
But so I was interested in that. The first time

00:01:59.730 --> 00:02:03.349
I actually saw a computer is I was quite young.

00:02:03.430 --> 00:02:05.109
I was in early and grade school and I visited

00:02:05.109 --> 00:02:07.909
my grandfather, editor of a small newspaper,

00:02:07.909 --> 00:02:10.430
and I visited him at work and they had a punch

00:02:10.430 --> 00:02:12.719
card machine. IBM punch card machine they use

00:02:12.719 --> 00:02:15.479
it I guess for circulation for subscriptions

00:02:15.479 --> 00:02:18.460
and stuff and I was really taken by that how

00:02:18.460 --> 00:02:21.280
it could sort do the sorting and I actually built

00:02:21.280 --> 00:02:24.860
for a science fair project I built a rudimentary

00:02:24.860 --> 00:02:28.539
punch card sorter that you could Crank it and

00:02:28.539 --> 00:02:31.419
you know as the thing went through it It was

00:02:31.419 --> 00:02:33.840
able to tell and the magnet would open lift something

00:02:33.840 --> 00:02:36.659
that would go here or there depending on it So

00:02:36.659 --> 00:02:39.800
that that sort of got me wanting to get into

00:02:39.800 --> 00:02:43.889
it Um, and so that attracted me. Uh, I was always,

00:02:43.889 --> 00:02:47.650
uh, viewed as a techie into space and things

00:02:47.650 --> 00:02:51.169
like that. Very cool. So then, yeah, we'll fast

00:02:51.169 --> 00:02:53.930
forward. You will see you did, uh, you started

00:02:53.930 --> 00:02:56.229
at MIT and well, yeah, but before, but I got

00:02:56.229 --> 00:02:58.830
into computers early. I mean, I got into computers,

00:02:58.830 --> 00:03:03.090
uh, when I was like, uh, what, 15. My cousin

00:03:03.090 --> 00:03:05.710
came home from his school and they had just gotten

00:03:05.710 --> 00:03:08.370
access to a time sharing system. They were one

00:03:08.370 --> 00:03:11.069
of the few schools in the whole Philadelphia

00:03:11.069 --> 00:03:14.150
area who were able to do a version of Fortran

00:03:14.150 --> 00:03:16.849
called QuickTran using a remote terminal, type

00:03:16.849 --> 00:03:19.949
terminal. And I got to go over and learn. He

00:03:19.949 --> 00:03:21.889
brought the manual home and I studied the manual

00:03:21.889 --> 00:03:24.590
and learned Fortran, enough Fortran too, and

00:03:24.590 --> 00:03:27.370
started programming. Like one of the first programs.

00:03:27.789 --> 00:03:31.150
that I wrote was we had learned in school patterns

00:03:31.150 --> 00:03:33.789
of English. We learned about noun, verb, whatever

00:03:33.789 --> 00:03:36.210
and that a sentence had this. So I put in all

00:03:36.210 --> 00:03:38.349
these types of sentences and I put in a list

00:03:38.349 --> 00:03:40.800
of nouns and verbs and it would make up... It

00:03:40.800 --> 00:03:43.300
would make up sentences and stuff that were technically

00:03:43.300 --> 00:03:45.419
correct. I thought that was pretty cool. That's

00:03:45.419 --> 00:03:47.300
very cool. So that's one of the first things.

00:03:47.580 --> 00:03:49.240
And then I built a graphing program. I think

00:03:49.240 --> 00:03:52.439
the thing that did graphing, you know, on the

00:03:52.439 --> 00:03:55.020
printer, the typewriter output. Then I would

00:03:55.020 --> 00:03:57.319
beg, steal or borrow computer time to be able

00:03:57.319 --> 00:04:00.060
to improve my stuff wherever I could find access

00:04:00.060 --> 00:04:02.439
to computing. In those days, it was very hard

00:04:02.439 --> 00:04:05.240
for like a high school or anybody to get access

00:04:05.240 --> 00:04:08.060
to computers. But I learned how to be able to

00:04:08.060 --> 00:04:10.319
find where... There were ones that I could get

00:04:10.319 --> 00:04:12.240
access to. They're usually, you know, like the

00:04:12.240 --> 00:04:15.000
Bureau of Public Education in Philadelphia. Right.

00:04:15.360 --> 00:04:17.680
I was like the only one who would go convince

00:04:17.680 --> 00:04:22.160
them to let me go down and use their 1401 computers.

00:04:22.439 --> 00:04:24.839
Yeah. I mean, it's so cool because it was like

00:04:24.839 --> 00:04:28.620
right as it was really becoming a thing. So not

00:04:28.620 --> 00:04:31.060
very many people had access to it. So you were.

00:04:31.639 --> 00:04:34.019
Well, yeah, as these are all mainframes. Right.

00:04:34.220 --> 00:04:38.889
I mean, I in late. Let's see. In later high school,

00:04:39.209 --> 00:04:41.449
I took a course, a summer course at the University

00:04:41.449 --> 00:04:45.110
of Pennsylvania, where the National Science Foundation

00:04:45.110 --> 00:04:48.329
had a programming computer course. And they had

00:04:48.329 --> 00:04:52.170
this 1130 computer that was brand new. And I

00:04:52.170 --> 00:04:56.129
was able to program that and learn computer stuff

00:04:56.129 --> 00:04:59.089
at that course at the University of Pennsylvania,

00:04:59.189 --> 00:05:00.949
even though I was a high school student. And

00:05:00.949 --> 00:05:04.529
then I got a job at the person who ran that course.

00:05:05.270 --> 00:05:07.470
worked at the Wharton School at the University

00:05:07.470 --> 00:05:10.790
of Pennsylvania and worked in computational services.

00:05:10.810 --> 00:05:14.629
And I got an afternoon job in high school to

00:05:14.629 --> 00:05:17.310
be able to work there. And I had access, suddenly

00:05:17.310 --> 00:05:19.850
I had access to computers through the University

00:05:19.850 --> 00:05:22.350
of Pennsylvania. Very cool. As a high school

00:05:22.350 --> 00:05:24.970
kid. Nice. Yeah. So I was at the engineering

00:05:24.970 --> 00:05:27.329
school at University of Pennsylvania. So every

00:05:27.329 --> 00:05:29.230
day when I would go through one of the halls,

00:05:29.370 --> 00:05:32.470
I'd pass the ENIAC machine. Is that, is that?

00:05:32.709 --> 00:05:35.069
Yeah. So I guess, yeah. Well, whatever. It was

00:05:35.069 --> 00:05:36.949
down the hall from where I took my course. I

00:05:36.949 --> 00:05:39.509
knew that there was nothing to look at or anything.

00:05:40.389 --> 00:05:43.029
Yeah. Now it's just there behind glass and you

00:05:43.029 --> 00:05:45.089
can kind of look at it. So they built a shrine.

00:05:45.290 --> 00:05:46.889
So they didn't have a shrine. They didn't have

00:05:46.889 --> 00:05:48.529
the shrine. I don't remember there being a shrine.

00:05:48.649 --> 00:05:51.490
I didn't see it. I just knew that it was somewhere.

00:05:51.819 --> 00:05:54.220
I was at the Moore School, which is where it

00:05:54.220 --> 00:05:56.420
was done. Yeah, that's where I took a lot of

00:05:56.420 --> 00:05:58.139
my classes. It's kind of cool to be connected

00:05:58.139 --> 00:06:00.879
to the past. I like looking at the history of

00:06:00.879 --> 00:06:04.279
stuff. So it's nice to be connected to the past.

00:06:04.519 --> 00:06:07.019
Yeah, definitely. Tell us a little bit like a

00:06:07.019 --> 00:06:09.980
mainframe computer. Like, so does it have all

00:06:09.980 --> 00:06:12.899
of the same things? You know, how different?

00:06:13.180 --> 00:06:15.680
Those days? Well, in those days, I mean, they're

00:06:15.680 --> 00:06:18.639
very, very powerful for those days. Today they're,

00:06:18.720 --> 00:06:20.959
you know, my watch is more powerful. Right. Microwave

00:06:20.959 --> 00:06:26.180
oven is probably as powerful. They were in these

00:06:26.180 --> 00:06:29.040
big, you know, the cabinets that look like big

00:06:29.040 --> 00:06:32.600
refrigerators and multiple of them, and each

00:06:32.600 --> 00:06:35.649
one had different parts in that. Uh, now we might

00:06:35.649 --> 00:06:37.550
have, you know, half a chip or a fraction of

00:06:37.550 --> 00:06:39.970
a chip to be that whole thing. So there was the

00:06:39.970 --> 00:06:41.910
CPU, maybe a couple of them. There were tape

00:06:41.910 --> 00:06:45.170
drives, you know, that, uh, were used for storage

00:06:45.170 --> 00:06:47.329
for longer term storage. And there were disk

00:06:47.329 --> 00:06:50.189
drives, you know, and stuff like that. And it

00:06:50.189 --> 00:06:52.670
produced a lot of heat and they had lots of big

00:06:52.670 --> 00:06:56.509
cables for lots of power and stuff. So the floors

00:06:56.509 --> 00:06:59.990
were raised. They had, so you could get, you

00:06:59.990 --> 00:07:03.449
could pull a, a tile off and get underneath and.

00:07:03.769 --> 00:07:06.370
you know put the wires down there between them

00:07:06.370 --> 00:07:08.110
so you wouldn't trip but also they could take

00:07:08.110 --> 00:07:10.430
air conditioning through it to get the heat off

00:07:10.430 --> 00:07:13.189
and the air conditioners are running you know

00:07:13.189 --> 00:07:14.970
you hear the noise when you go in the there was

00:07:14.970 --> 00:07:16.889
a computer room and you didn't go in there because

00:07:16.889 --> 00:07:18.850
you don't want to get any dust in there right

00:07:18.850 --> 00:07:21.029
or anything like that that might mess up the

00:07:21.029 --> 00:07:24.629
card readers and stuff so you know you did things

00:07:24.629 --> 00:07:27.699
outside and Plus, you don't want to give normal

00:07:27.699 --> 00:07:29.620
people, regular people access to that room. They

00:07:29.620 --> 00:07:33.660
might mess something up. So that was what mainframes

00:07:33.660 --> 00:07:38.500
were like. They had CPUs, they had short -term

00:07:38.500 --> 00:07:40.980
storage and long -term storage and non -volatile

00:07:40.980 --> 00:07:43.480
storage and they did all the same. They had computer

00:07:43.480 --> 00:07:46.160
languages and you could connect terminals to

00:07:46.160 --> 00:07:49.300
them of various sorts to do visual stuff or printed

00:07:49.300 --> 00:07:52.779
output with paper printers that could print 10

00:07:52.779 --> 00:07:56.040
pages a minute or something like that. You know,

00:07:56.600 --> 00:07:59.399
and that's how checks were printed and and reports

00:07:59.399 --> 00:08:02.860
and stuff So it's a lot of what we have today,

00:08:02.939 --> 00:08:07.600
but just it was bigger Just if you take today

00:08:07.600 --> 00:08:10.220
literally within the last few days for most for

00:08:10.220 --> 00:08:12.620
any normal person getting it You have the new

00:08:12.620 --> 00:08:16.009
Apple has a new phone where all the electronics

00:08:16.009 --> 00:08:18.410
are basically are where the camera is a little

00:08:18.410 --> 00:08:20.589
strip where the camera is is the whole thing

00:08:20.589 --> 00:08:22.250
of the phone because they've managed to rise

00:08:22.250 --> 00:08:24.769
the whole phone down to the cameras and everything

00:08:24.769 --> 00:08:26.670
fit in a little strip and the rest is battery

00:08:26.670 --> 00:08:30.730
and screen but the basic idea of storage computation

00:08:30.730 --> 00:08:34.490
you know uh and display and input in and those

00:08:34.490 --> 00:08:37.049
are all different parts software that runs on

00:08:37.049 --> 00:08:40.330
them it's a stored program computer that that

00:08:40.330 --> 00:08:43.429
is you know, from an architectural viewpoint,

00:08:43.649 --> 00:08:47.710
it's very similar. Right. It's pretty amazing

00:08:47.710 --> 00:08:50.769
to have seen the entire evolution and being one

00:08:50.769 --> 00:08:53.529
of the people that was accessing those mainframes

00:08:53.529 --> 00:08:57.690
to now see how that has all evolved into today.

00:08:58.330 --> 00:09:01.710
So some of the work that you did helped propel

00:09:01.710 --> 00:09:04.710
the PC, you know, the personal computer revolution.

00:09:04.990 --> 00:09:07.309
You worked on a word processor pretty early on,

00:09:07.429 --> 00:09:11.340
right? Yes. Yeah, very early on, I was at Digital

00:09:11.340 --> 00:09:14.759
Equipment Corporation after college. At the end

00:09:14.759 --> 00:09:17.960
of college, I worked on a system that used a

00:09:17.960 --> 00:09:21.419
language called APL that regular people, but

00:09:21.419 --> 00:09:24.620
mainly engineers and people like that, could

00:09:24.620 --> 00:09:27.799
type in a simple program and get the results

00:09:27.799 --> 00:09:30.879
out of it. But then when I went to Digital Equipment

00:09:30.879 --> 00:09:33.220
Corporation, which was a large mini computer

00:09:33.220 --> 00:09:36.059
manufacturer, I worked in computerized typesetting.

00:09:36.820 --> 00:09:41.759
and worked with editing for editing the articles

00:09:41.759 --> 00:09:45.220
that went through for connecting to wire service

00:09:45.220 --> 00:09:47.659
stuff. That's data that's coming in from outside.

00:09:47.840 --> 00:09:51.279
And then I ended up in their first word processor.

00:09:51.779 --> 00:09:55.220
It was a computerized word processor, a document

00:09:55.220 --> 00:09:57.899
oriented one. In those days, there were two types

00:09:57.899 --> 00:10:00.899
of electronic with screen based word processing.

00:10:01.860 --> 00:10:03.759
There were just a few companies that made any.

00:10:03.929 --> 00:10:07.049
of those and one of them was was page oriented

00:10:07.049 --> 00:10:10.909
kind of like page maker or um powerpoint or something

00:10:10.909 --> 00:10:13.789
you laid out one page and then you laid out the

00:10:13.789 --> 00:10:16.909
next page and you position things within that

00:10:16.909 --> 00:10:19.450
as you type document oriented is what we're is

00:10:19.450 --> 00:10:22.159
more like word is what we're familiar with today.

00:10:22.379 --> 00:10:25.440
Ours used just one long thing of text and it

00:10:25.440 --> 00:10:28.379
broke it up into pages and cared about layout

00:10:28.379 --> 00:10:31.100
horizontally with tabs and decimal tabs that

00:10:31.100 --> 00:10:33.779
line up just right and how do you do that and

00:10:33.779 --> 00:10:38.679
I worked on that which was to be used by typists.

00:10:38.970 --> 00:10:42.190
Then hopefully by managers and executives, but

00:10:42.190 --> 00:10:45.049
they didn't type in those days, but they would

00:10:45.049 --> 00:10:47.549
give things off to typists So that's I worked

00:10:47.549 --> 00:10:50.789
on that that software ended up on all sorts of

00:10:50.789 --> 00:10:53.190
different hardware It started out with a computer

00:10:53.190 --> 00:10:56.330
in the deck inside of a desk. The computer was

00:10:56.330 --> 00:10:59.809
a big and the screen Part of the desk eventually

00:10:59.809 --> 00:11:02.470
they sold it as a one -piece unit for about ten

00:11:02.470 --> 00:11:05.710
years as the deck mate Word processor that started

00:11:05.710 --> 00:11:10.009
as a deck WPS eight system. So and it looked

00:11:10.009 --> 00:11:14.190
like this. Here is an ad for it. This is a typewriter.

00:11:14.669 --> 00:11:19.809
This was our system. That was on top of the desk,

00:11:20.389 --> 00:11:25.129
you know, and but you can see the desk with the

00:11:25.129 --> 00:11:27.590
and the floppy disks, the eight inch floppy disks

00:11:27.590 --> 00:11:30.990
that went in it. So and it could do output like

00:11:30.990 --> 00:11:34.929
this. This is the functional specification for

00:11:34.929 --> 00:11:38.889
it. for this thing written in itself, in an early

00:11:38.889 --> 00:11:43.009
version of itself. So I got to, that was word

00:11:43.009 --> 00:11:45.610
wrapping, being able, and I wrote software that

00:11:45.610 --> 00:11:49.690
did some of that stuff. There was a team of about

00:11:49.690 --> 00:11:53.049
four of us doing the software for that word processor.

00:11:53.490 --> 00:11:57.549
Before that, I had worked on an editing system

00:11:57.549 --> 00:12:01.470
for newspapers. I was familiar with some newspaper

00:12:01.470 --> 00:12:04.940
systems that were specifically for editing. You

00:12:04.940 --> 00:12:07.620
know two -dimensional layouts for giving the

00:12:07.620 --> 00:12:10.620
commands to a type setting system for doing the

00:12:10.620 --> 00:12:14.419
ads in a newspaper I got to see those that people

00:12:14.419 --> 00:12:16.860
had there was an interactive system called the

00:12:16.860 --> 00:12:20.460
Harris 2200 that you can't find too much about

00:12:20.460 --> 00:12:23.159
but was a groundbreaking machine that was basically

00:12:23.159 --> 00:12:27.899
page maker and hardware and That I was quite

00:12:27.899 --> 00:12:31.539
taken by and that influenced visit calc in a

00:12:31.539 --> 00:12:34.799
way Yeah, so getting into VisiCalc. So, you know,

00:12:35.000 --> 00:12:39.019
VisiCalc, I mean, widely regarded as fueling,

00:12:39.019 --> 00:12:41.399
you know, the rapid growth of the personal computing

00:12:41.399 --> 00:12:43.559
industry. You know, Steve Jobs went as far as

00:12:43.559 --> 00:12:45.899
to say that VisiCalc is what propelled Apple

00:12:45.899 --> 00:12:48.799
II to the success it achieved. He says that in

00:12:48.799 --> 00:12:51.519
an interview in like the 1990s. Because it was

00:12:51.519 --> 00:12:54.659
only available for a year on the Apple II. Because

00:12:54.659 --> 00:12:57.090
software takes a long time. Yeah, where do you

00:12:57.090 --> 00:12:59.690
want to start with visit calc? It's it's referenced

00:12:59.690 --> 00:13:02.730
everywhere right like in prediction machines

00:13:02.730 --> 00:13:05.450
The book it talks about like that's like one

00:13:05.450 --> 00:13:08.009
of the first steps towards realizing the power

00:13:08.009 --> 00:13:10.830
of automation You know change the way that it

00:13:10.830 --> 00:13:14.269
was interactive computing people were not used

00:13:14.269 --> 00:13:17.330
to interacting with a computer other than to

00:13:17.330 --> 00:13:19.470
put their card in and get money out of a machine,

00:13:19.470 --> 00:13:22.590
right? That was the main way that people interacted

00:13:22.590 --> 00:13:25.350
with a computer. Here was a thing, a word processor,

00:13:25.370 --> 00:13:27.929
you know, maybe they had tried a word processor,

00:13:27.929 --> 00:13:31.789
but here is a thing that, you know, did things

00:13:31.789 --> 00:13:33.870
that they would have to do by hand they were

00:13:33.870 --> 00:13:36.529
supposed to, but it could do it so much faster

00:13:36.529 --> 00:13:43.439
and open up all sorts of worlds. So that That

00:13:43.439 --> 00:13:46.279
really affected things quite a bit. It let people

00:13:46.279 --> 00:13:48.220
say, oh, this is what an interactive computer

00:13:48.220 --> 00:13:51.279
is. It's not just games. They did have access.

00:13:51.419 --> 00:13:53.740
They were playing Pong and stuff like that, Atari

00:13:53.740 --> 00:13:59.799
stuff. And so, oh, it can be useful in things

00:13:59.799 --> 00:14:01.919
that are better than anything else. Even a mainframe

00:14:01.919 --> 00:14:05.000
doesn't do anything. This interactive, me give

00:14:05.000 --> 00:14:08.279
and taking with it and getting the output that

00:14:08.279 --> 00:14:11.610
I needed and it's custom to what I need. that

00:14:11.610 --> 00:14:14.090
just like a word processor. There were programs

00:14:14.090 --> 00:14:16.629
before word processors that would construct a

00:14:16.629 --> 00:14:18.309
will, would ask you questions and then write

00:14:18.309 --> 00:14:21.529
a will. But a word processor lets you write anything.

00:14:22.149 --> 00:14:24.309
You know, that's a general purpose tool. Well,

00:14:24.370 --> 00:14:25.850
a spreadsheet was a general purpose tool for

00:14:25.850 --> 00:14:27.590
numbers, like a word processor was a general

00:14:27.590 --> 00:14:30.929
purpose tool for paragraphs or prose. For numbers

00:14:30.929 --> 00:14:35.129
and for also text, people often used it only

00:14:35.129 --> 00:14:39.149
for text to keep lists. And since you could lay

00:14:39.149 --> 00:14:41.769
things out any way you wanted, it was not strict

00:14:41.769 --> 00:14:45.190
columns and rows that all the rows are the same,

00:14:45.370 --> 00:14:47.710
and all the columns have the same one. No, it

00:14:47.710 --> 00:14:49.509
wasn't like that. You could put anything anywhere,

00:14:49.529 --> 00:14:51.529
but you could take advantage of it. You could

00:14:51.529 --> 00:14:54.470
very quickly create things because it had just

00:14:54.470 --> 00:14:58.029
the right tools to let you do things very fast,

00:14:58.370 --> 00:15:01.429
because it was optimized for keystroke, minimum

00:15:01.429 --> 00:15:04.120
keystrokes, because I came from type setting

00:15:04.120 --> 00:15:06.360
where you're paid by the keystroke so I learned

00:15:06.360 --> 00:15:09.419
how to optimize for quick and I was competing

00:15:09.419 --> 00:15:14.179
against paper with a calculator and a pencil

00:15:14.179 --> 00:15:17.159
and paper and it had to be faster the first time

00:15:17.159 --> 00:15:19.820
you used it because otherwise you wouldn't use

00:15:19.820 --> 00:15:21.779
the recalculation if it was quicker you say oh

00:15:21.779 --> 00:15:23.980
I only need this once I'll do it by hand right

00:15:23.980 --> 00:15:28.909
so so that I think that said people wow Computers

00:15:28.909 --> 00:15:31.950
can be useful in so many ways. What other things

00:15:31.950 --> 00:15:34.629
can they do? This is interactivity. What you

00:15:34.629 --> 00:15:37.049
see is what you get. The thing is you could build

00:15:37.049 --> 00:15:38.889
something and build on it and build on it and

00:15:38.889 --> 00:15:43.710
give it to somebody else. And then all sorts

00:15:43.710 --> 00:15:46.269
of other software start coming out. Yeah. So

00:15:46.269 --> 00:15:49.820
you talk about it... You have like a some sort

00:15:49.820 --> 00:15:52.299
of formula about figuring out if something can

00:15:52.299 --> 00:15:54.019
be like a killer app. You were kind of referencing

00:15:54.019 --> 00:15:55.899
it before, like, so you're comparing, like, what

00:15:55.899 --> 00:15:57.419
was it before, right? Like people were trying

00:15:57.419 --> 00:15:59.539
to do spreadsheets existed, right? It just was

00:15:59.539 --> 00:16:02.340
the idea that you could use a spreadsheet on,

00:16:02.340 --> 00:16:04.740
you know, on your machine, you could change one

00:16:04.740 --> 00:16:06.320
number, and then that would lead to cascading

00:16:06.320 --> 00:16:08.179
of everything changing, you could put in custom

00:16:08.179 --> 00:16:10.850
formulas, you could do all this stuff. I guess,

00:16:10.950 --> 00:16:13.889
as you were making it, did you know, like, did

00:16:13.889 --> 00:16:16.090
you know the magnitude of what you were creating?

00:16:16.129 --> 00:16:17.950
Well, of course it should be. Whenever you do

00:16:17.950 --> 00:16:20.009
something, it's going to be great. You know,

00:16:20.129 --> 00:16:23.210
it usually isn't. Yes. I knew it would be very

00:16:23.210 --> 00:16:25.350
useful, but people were not adopting computers

00:16:25.350 --> 00:16:27.509
at the rate we expected. Word processing was

00:16:27.509 --> 00:16:30.250
not taking off at that point. Right. Back in

00:16:30.250 --> 00:16:33.730
1979, it was not going like that. Normal people

00:16:33.730 --> 00:16:36.789
did not get it. It wasn't until the inexpensive.

00:16:37.210 --> 00:16:41.730
computers like the Osborne and the Kaypro, you

00:16:41.730 --> 00:16:46.129
know, that a journalist could buy their own computer

00:16:46.129 --> 00:16:49.250
to be able to take home and be able with a word

00:16:49.250 --> 00:16:52.509
processor at all and write things. And suddenly

00:16:52.509 --> 00:16:55.330
people could do their novels on a word processor.

00:16:55.590 --> 00:16:58.370
That wasn't until a few years later. And that

00:16:58.370 --> 00:17:01.629
was because PCs became popular. PCs became popular

00:17:01.629 --> 00:17:04.589
because I think one of them is people involved

00:17:04.589 --> 00:17:08.039
in money like the spreadsheet. They said, oh,

00:17:08.059 --> 00:17:11.779
these PCs are useful. And then when they were

00:17:11.779 --> 00:17:14.380
good for word processing, the journalist said,

00:17:14.380 --> 00:17:20.359
oh, PCs can be useful. And then people, their

00:17:20.359 --> 00:17:22.859
kids went away to college, and the only way to

00:17:22.859 --> 00:17:24.900
communicate with them was by instant messaging

00:17:24.900 --> 00:17:29.240
and email. Oh, computers can be useful when connected

00:17:29.240 --> 00:17:33.380
to a dial -in remote stuff. And then the internet.

00:17:35.470 --> 00:17:38.670
Every time there is some new capability from

00:17:38.670 --> 00:17:40.710
a hardware viewpoint, or hardware and software

00:17:40.710 --> 00:17:45.450
combination, that is much, much, at least 100

00:17:45.450 --> 00:17:48.410
times better than what was before. Like, think

00:17:48.410 --> 00:17:51.069
the mouse is so much better than arrow keys,

00:17:51.289 --> 00:17:54.569
right? In terms of precise positioning quickly.

00:17:55.049 --> 00:17:57.829
Every time the laser printer compared to the

00:17:57.829 --> 00:18:01.690
typewriter output, you know. It was a ball, remember,

00:18:02.089 --> 00:18:04.089
the Selectrix. I don't know if people know that.

00:18:04.349 --> 00:18:07.849
But in the line printers, the laser printer could

00:18:07.849 --> 00:18:12.289
put 300 dots per inch each direction and do pictures

00:18:12.289 --> 00:18:14.950
and everything. It was so much better than what

00:18:14.950 --> 00:18:17.430
came before. That opens up new capabilities.

00:18:17.690 --> 00:18:20.410
So you then say, with this new capability that

00:18:20.410 --> 00:18:23.410
is two orders of magnitude, 100 times or more

00:18:23.410 --> 00:18:26.369
than what I had before, what does this open up

00:18:26.369 --> 00:18:29.029
and what is it good for? This is good for something

00:18:29.029 --> 00:18:32.230
I used to do or I needed to do, but didn't do.

00:18:32.650 --> 00:18:36.630
And that then lets us move forward. Portability

00:18:36.630 --> 00:18:39.150
makes a big difference. It's a big difference

00:18:39.150 --> 00:18:41.710
to be able to find, you know, to find messages

00:18:41.710 --> 00:18:45.230
in a fraction of a second compared to have to

00:18:45.230 --> 00:18:49.910
dial up, you know, to AOL or something and log

00:18:49.910 --> 00:18:53.089
in to read your emails and stuff like that. Or,

00:18:53.089 --> 00:18:55.069
or, but that was so much better than waiting

00:18:55.069 --> 00:18:57.569
for the mail to come, you know, the snail mail.

00:18:57.869 --> 00:19:02.150
and each of those things, the fax machine, it

00:19:02.150 --> 00:19:04.750
opened up so many things, right? Because you

00:19:04.750 --> 00:19:07.490
could send a document, anything drawn or typed

00:19:07.490 --> 00:19:10.970
or whatever, you could send it immediately to

00:19:10.970 --> 00:19:14.829
other people. Now, what was interesting is about

00:19:14.829 --> 00:19:18.009
putting up with things that aren't so good. We

00:19:18.009 --> 00:19:20.829
worked really hard in word processing to be letter

00:19:20.829 --> 00:19:23.329
perfect, they called it. The output, you wanted

00:19:23.329 --> 00:19:26.470
it to look as good as a good typewriter. Okay,

00:19:27.069 --> 00:19:30.069
really really good. Then we had the laser printer

00:19:30.069 --> 00:19:33.309
that could produce fonts of all sorts, perfectly

00:19:33.309 --> 00:19:35.890
sharp, indistinguishable from the old stuff.

00:19:36.269 --> 00:19:38.450
Then what do we do? We said, oh, we'll stick

00:19:38.450 --> 00:19:40.609
it in the fax machine that turns it down to 150

00:19:40.609 --> 00:19:43.269
dots per inch and you know, whatever and it's

00:19:43.269 --> 00:19:47.150
only black and white and no no gray and You know,

00:19:47.450 --> 00:19:49.309
we'll send it that way We'll take our litter

00:19:49.309 --> 00:19:51.670
perfect thing that we work so hard and turn it

00:19:51.670 --> 00:19:53.849
into this barely readable thing on paper That

00:19:53.849 --> 00:19:56.670
was kind of wet and you know faded whatever but

00:19:56.670 --> 00:19:59.869
we could get it immediately right and you know

00:19:59.869 --> 00:20:02.970
And we put up with that when we first had cell

00:20:02.970 --> 00:20:05.140
phones Can you hear me now? Can you hear me now?

00:20:05.380 --> 00:20:08.539
It was, you know, was, but. And they were enormous.

00:20:08.660 --> 00:20:10.519
They were huge. They were huge. But at least

00:20:10.519 --> 00:20:14.119
you could carry it with you. Right. And so that's,

00:20:14.119 --> 00:20:17.740
you know, you look at when you have a new capability,

00:20:17.980 --> 00:20:20.460
it opens up some things and you're willing to

00:20:20.460 --> 00:20:23.220
put up with it not being as good. A lot of this

00:20:23.220 --> 00:20:26.539
is described by Clay Christensen. May he rest

00:20:26.539 --> 00:20:31.349
in peace. And the innovator's dilemma. Yes. So

00:20:31.349 --> 00:20:34.670
about that whole thing about when you have something

00:20:34.670 --> 00:20:37.930
that disruptive technology. So we can look at

00:20:37.930 --> 00:20:41.049
that for any new technology we have today. And

00:20:41.049 --> 00:20:43.490
how is it better or not from what we had before?

00:20:43.549 --> 00:20:45.789
And what is it good for? What is it not good

00:20:45.789 --> 00:20:49.049
for? And what doesn't it matter? Right, right.

00:20:49.769 --> 00:20:52.630
This sort of trade off when there's something

00:20:52.630 --> 00:20:55.349
like some new capability, but then there's this

00:20:55.349 --> 00:20:58.920
trade off between quality and then port, like

00:20:58.920 --> 00:21:01.680
portability. I think about music also, right?

00:21:01.779 --> 00:21:05.339
It's like the quality of music by changing it

00:21:05.339 --> 00:21:08.559
and having it be into an mp3, you're losing a

00:21:08.559 --> 00:21:10.680
lot of quality. For those that can hear that,

00:21:10.859 --> 00:21:14.000
not all of us. The thing is that the portability

00:21:14.000 --> 00:21:18.140
and the fact that you could carry in your pocket

00:21:18.140 --> 00:21:21.980
hundreds of songs, you know, to listen anywhere.

00:21:23.000 --> 00:21:25.640
But the quality was pretty you know compared

00:21:25.640 --> 00:21:28.579
to a transistor radio You know listening on the

00:21:28.579 --> 00:21:30.359
radio waiting to hear whatever happened to be

00:21:30.359 --> 00:21:32.440
able this tinny little speaker when you had those

00:21:32.440 --> 00:21:34.920
really good earphones that The Walkman when I

00:21:34.920 --> 00:21:36.680
when Sony did the Walkman they did these really

00:21:36.680 --> 00:21:38.960
good earphones with these special magnets and

00:21:38.960 --> 00:21:42.410
whatever So it actually sounded better then it

00:21:42.410 --> 00:21:44.269
probably would be if you used your expensive

00:21:44.269 --> 00:21:47.230
stereo in a noisy room so and it doesn't matter

00:21:47.230 --> 00:21:49.029
because you're you're portable you can do it

00:21:49.029 --> 00:21:51.930
when walking or running or when you're commuting

00:21:51.930 --> 00:21:56.750
so yeah this this trade -off of uh and then eventually

00:21:56.750 --> 00:22:00.789
they get better where the sound quality you have

00:22:00.789 --> 00:22:03.769
with the latest little earphones wherever running

00:22:03.769 --> 00:22:06.009
off of you know directly off your watch or your

00:22:06.009 --> 00:22:09.750
phone or whatever are like concert quality compared

00:22:09.750 --> 00:22:12.769
to our old speakers that we had you know in our

00:22:12.769 --> 00:22:14.670
dorm rooms or something like that years before

00:22:14.670 --> 00:22:17.950
for an expensive stereo. So it does catch up

00:22:17.950 --> 00:22:20.470
that's the innovator that's exactly Clay Christensen's

00:22:20.470 --> 00:22:25.329
about how eventually the toy becomes fixes the

00:22:25.329 --> 00:22:27.130
differences between it and eventually becomes

00:22:27.130 --> 00:22:29.509
better because of the volume and stuff like that.

00:22:29.680 --> 00:22:33.740
So going into VisiCalc and creating something

00:22:33.740 --> 00:22:38.339
like that, what were, I guess, a couple of questions.

00:22:38.660 --> 00:22:40.680
Like, so when you're creating this general purpose

00:22:40.680 --> 00:22:44.440
tool, first, it's hard to think of all of the

00:22:44.440 --> 00:22:46.140
possibilities that you could, you know, that

00:22:46.140 --> 00:22:49.059
you could hit, right? But I guess could you talk

00:22:49.059 --> 00:22:51.799
about... I had no clue. I mean, I knew about

00:22:51.799 --> 00:22:54.720
a lot because I was at business school learning

00:22:54.720 --> 00:22:57.019
about all different aspects of business. So I

00:22:57.019 --> 00:23:00.250
was exposed to many uses. Right. but it was a

00:23:00.250 --> 00:23:02.289
fraction of the uses that it actually became

00:23:02.289 --> 00:23:05.450
used for, even in the first year. Right, right.

00:23:05.690 --> 00:23:10.309
Yeah, your story about... So you have a great

00:23:10.309 --> 00:23:12.609
TED Talk that I encourage all listeners to listen

00:23:12.609 --> 00:23:15.789
to about the history of VisiCalc, but you mentioned

00:23:15.789 --> 00:23:19.130
there that you used the spreadsheet basically

00:23:19.130 --> 00:23:21.869
to help you with your case studies and to do

00:23:21.869 --> 00:23:24.539
projections and to do that... those sort of things,

00:23:24.720 --> 00:23:27.039
which personally was, is kind of cool for me

00:23:27.039 --> 00:23:29.799
because one of my first case studies was, was

00:23:29.799 --> 00:23:32.960
yours, was your case. I mean, they're just, it's,

00:23:32.960 --> 00:23:36.099
there's so much echoing of the past. Yeah. So

00:23:36.099 --> 00:23:38.299
it's, it's, it's, it's a pleasure to be. And

00:23:38.299 --> 00:23:41.519
it was highly interact. VisiCalc is highly interactive.

00:23:42.140 --> 00:23:44.180
I worked on the word processor with somebody

00:23:44.180 --> 00:23:46.339
in the name of Jack Gilmore, may he rest in peace,

00:23:46.660 --> 00:23:51.380
who was on the word processor. He developed what

00:23:51.380 --> 00:23:54.099
is arguably one of the first video games. Well,

00:23:54.099 --> 00:23:56.319
it is one of the first, but it really could be

00:23:56.319 --> 00:24:00.480
the first video games in 1951. Oh, wow. You know,

00:24:00.660 --> 00:24:03.740
and, you know, on the whirlwind computer. So,

00:24:03.779 --> 00:24:08.039
you know, this moving along and from person to

00:24:08.039 --> 00:24:10.740
person to person is kind of cool. Yeah, it's

00:24:10.740 --> 00:24:15.319
it. Yeah, it's it's it's really cool. So while

00:24:15.319 --> 00:24:18.839
you were developing it at like, what were the

00:24:18.839 --> 00:24:22.180
challenges that you were facing? Were there like?

00:24:22.480 --> 00:24:26.200
I mean, I guess the technical challenges of trying

00:24:26.200 --> 00:24:28.240
to fit it. I know you mentioned that there was

00:24:28.240 --> 00:24:32.140
a particular upgrade in the Apple II that made

00:24:32.140 --> 00:24:35.579
it possible, I think, or there was some additional

00:24:35.579 --> 00:24:38.380
memory or something that allowed it. Oh, well,

00:24:38.480 --> 00:24:40.819
I mean, there was stuff. The thing is that, well,

00:24:40.819 --> 00:24:43.660
the Apple II was good enough, but the challenges

00:24:43.660 --> 00:24:48.539
were the Apple II had a very small screen in

00:24:48.539 --> 00:24:52.700
terms of characters. 25 rows of 40 characters

00:24:52.700 --> 00:24:56.160
wide on the screen for character based. And it

00:24:56.160 --> 00:24:59.559
had two arrow keys. That's it. It had paddles

00:24:59.559 --> 00:25:01.960
you could play a game with, but I was gonna use

00:25:01.960 --> 00:25:04.059
that like a mouse, but I gave up on that very

00:25:04.059 --> 00:25:06.859
early on. Like the... No, no, no. The type you

00:25:06.859 --> 00:25:08.940
turn. Oh, okay. The turning one, yeah. I should

00:25:08.940 --> 00:25:12.099
go get it. I have one. An old Apple II with it.

00:25:12.500 --> 00:25:15.279
It was for playing punk. Right. You know, that

00:25:15.279 --> 00:25:17.779
was that type of stuff. Or Space Invaders, which

00:25:17.779 --> 00:25:19.940
is what everybody wanted to play, which was a

00:25:19.940 --> 00:25:24.839
game on the Apple II. So the challenge was you're

00:25:24.839 --> 00:25:27.599
working with data that probably... You know,

00:25:27.680 --> 00:25:29.960
you're going to want a lot of it. So luckily

00:25:29.960 --> 00:25:32.140
from my word processing background, I was used

00:25:32.140 --> 00:25:34.799
to scrolling and dealing with scrolling, but

00:25:34.799 --> 00:25:36.619
how do you deal with scrolling on such a small

00:25:36.619 --> 00:25:39.720
screen? So we had to do things like what we now

00:25:39.720 --> 00:25:43.039
call locked pains. We call them titles where

00:25:43.039 --> 00:25:46.579
you would have, you know, the name, some kind

00:25:46.579 --> 00:25:49.420
of maybe a row names or something. And as you

00:25:49.420 --> 00:25:51.660
scroll, you know, they would stay on the screen.

00:25:51.700 --> 00:25:55.150
Right. and I'll let you have two windows so you

00:25:55.150 --> 00:25:56.910
could look at two places at once. You could do

00:25:56.910 --> 00:25:59.150
this and see the results of a sum somewhere else.

00:25:59.490 --> 00:26:02.049
And if you did two scrolling, they would be synchronized.

00:26:02.309 --> 00:26:04.990
So we decided to put that software in. Help,

00:26:05.089 --> 00:26:07.049
we wanted to have a help system, but there was

00:26:07.049 --> 00:26:10.630
no space to put that much information. The only

00:26:10.630 --> 00:26:14.789
help we had is you typed a slash and it showed

00:26:14.789 --> 00:26:18.309
you what characters you could type for the commands

00:26:18.309 --> 00:26:23.380
to put in. And slash IR is insert row, which

00:26:23.380 --> 00:26:26.099
I think still if you type slash IR into some

00:26:26.099 --> 00:26:28.680
versions of Excel, it still inserts a row. Those

00:26:28.680 --> 00:26:30.559
letters that we showed on the screen were the

00:26:30.559 --> 00:26:32.519
same letters that we were comparing what you

00:26:32.519 --> 00:26:35.799
type to, to see if, you know, what to do next.

00:26:35.859 --> 00:26:39.200
I mean, to save memory. So memory, how do you

00:26:39.200 --> 00:26:41.710
store the numbers? First of all, we had to make

00:26:41.710 --> 00:26:44.109
it that if you saved it and bring it back again,

00:26:44.109 --> 00:26:47.089
it would still fit. So if you used up every byte

00:26:47.089 --> 00:26:49.549
of memory available on your computer, there was

00:26:49.549 --> 00:26:52.390
no putting it out to disk or any of that paging

00:26:52.390 --> 00:26:56.069
stuff that we knew from our other day. So you

00:26:56.069 --> 00:26:58.329
had to build it so that it would be repeatable.

00:26:58.690 --> 00:27:01.309
And therefore you needed certain things to be

00:27:01.309 --> 00:27:04.769
very quickly garbage collected using fixed size,

00:27:04.769 --> 00:27:08.130
whatever. So what was the largest number that

00:27:08.130 --> 00:27:09.970
we would store? Because we needed to know that.

00:27:10.579 --> 00:27:12.759
and Bob and I talked about it, Bob Frankston,

00:27:12.839 --> 00:27:15.160
who wrote most of the code, and we'd said, okay,

00:27:15.480 --> 00:27:19.000
let's imagine that people would use this to calculate

00:27:19.000 --> 00:27:21.240
the budget of the United States of America in

00:27:21.240 --> 00:27:25.240
dollars. That's enough significant digits before

00:27:25.240 --> 00:27:27.579
we go to scientific notation, where it's powers,

00:27:27.859 --> 00:27:31.690
you know. And sure enough, It ended up being

00:27:31.690 --> 00:27:34.730
used for that a few years later. But we had to

00:27:34.730 --> 00:27:37.589
make decisions like that. How many columns, how

00:27:37.589 --> 00:27:41.289
many rows, max, because, and when do we allocate

00:27:41.289 --> 00:27:43.950
memory to be fast versus how do we do it so that

00:27:43.950 --> 00:27:46.109
you could actually have a spreadsheet that could

00:27:46.109 --> 00:27:48.009
be this wide or could be this high, but it couldn't

00:27:48.009 --> 00:27:49.549
be both at the same time because there wasn't

00:27:49.549 --> 00:27:52.910
enough memory for it. We only had, you know,

00:27:52.910 --> 00:27:57.109
I don't know, like 30, you know, 30K thousand

00:27:57.109 --> 00:28:00.000
bytes of memory for your whole spreadsheet. I

00:28:00.000 --> 00:28:05.180
mean, we started on a system that was 32K bytes

00:28:05.180 --> 00:28:10.480
for the whole system. That was the program, the

00:28:10.480 --> 00:28:13.539
operating system, the screen buffer of what was

00:28:13.539 --> 00:28:17.079
going to be on the screen, and your data. We

00:28:17.079 --> 00:28:20.559
could fit a few K of data for that. Most people

00:28:20.559 --> 00:28:23.980
had 48K, which was the biggest version then that

00:28:23.980 --> 00:28:27.299
we could use if you... bought a special card

00:28:27.299 --> 00:28:29.660
to run some other program that sold this card,

00:28:29.740 --> 00:28:32.180
you got another 16K of memory and we could use

00:28:32.180 --> 00:28:34.700
that card to give you more memory and people

00:28:34.700 --> 00:28:36.619
would spend hundreds of dollars on this card

00:28:36.619 --> 00:28:38.480
and throw away the software that came with it

00:28:38.480 --> 00:28:42.319
in order to get the extra memory for bigger spreadsheets.

00:28:42.640 --> 00:28:46.019
So we were very constrained. We had a program,

00:28:46.220 --> 00:28:48.779
Bob had a program really, really well to be able

00:28:48.779 --> 00:28:53.980
to scroll and, you know, display. turn it all

00:28:53.980 --> 00:28:55.779
into the screen. You had to turn it into the

00:28:55.779 --> 00:28:58.079
screen dynamically as you scrolled. That's a

00:28:58.079 --> 00:29:00.240
lot of programming. It has to be fast enough

00:29:00.240 --> 00:29:05.140
for the repeat key. What we didn't know is that

00:29:05.140 --> 00:29:07.420
the repeat key was done with hardware that isn't

00:29:07.420 --> 00:29:10.160
based on time, but was based on an RC circuit,

00:29:10.380 --> 00:29:12.539
which means it varied from computer to computer.

00:29:13.079 --> 00:29:15.480
Our computer, you could scroll and stop, and

00:29:15.480 --> 00:29:17.779
it would stop. But some, if you stopped, it still

00:29:17.779 --> 00:29:20.180
had a few extra keys and kept going. But it had

00:29:20.180 --> 00:29:23.059
to be that fast. That was a real challenge to

00:29:23.059 --> 00:29:26.259
make it fit in memory, do all the things that

00:29:26.259 --> 00:29:29.400
we wanted it to do. And how do we decide what

00:29:29.400 --> 00:29:31.740
not to do? We wanted a great help system. We

00:29:31.740 --> 00:29:34.660
wanted more functions. We wanted, you know, I

00:29:34.660 --> 00:29:37.400
mean, you're just, you know, it's deciding what

00:29:37.400 --> 00:29:40.339
not to do is a very tough thing. And we had to

00:29:40.339 --> 00:29:43.400
make it easy to understand, you know, to learn.

00:29:44.039 --> 00:29:47.920
And people learn by using a reference card and

00:29:47.920 --> 00:29:50.480
a manual. They either read the manual or they

00:29:50.480 --> 00:29:52.740
looked at the reference card or both that tell

00:29:52.740 --> 00:29:54.980
them, if you push this key, it does that. And

00:29:54.980 --> 00:29:57.779
we had to make that as simple as possible and

00:29:57.779 --> 00:30:02.500
we succeeded. Yeah. And useful. Yeah, for sure.

00:30:02.619 --> 00:30:06.799
I mean, it's still the most used, one of the

00:30:06.799 --> 00:30:08.940
most used software. So VisiCalc was the precursor

00:30:08.940 --> 00:30:14.200
to Lotus 123, precursor to Excel. And then Google

00:30:14.200 --> 00:30:18.299
Docs and dozens and dozens and dozens of other

00:30:18.299 --> 00:30:20.440
spreadsheets that have been built. Right. Linux

00:30:20.440 --> 00:30:24.109
has their own open source version as well. You

00:30:24.109 --> 00:30:27.630
know, what's funny is, you know, designing software,

00:30:28.009 --> 00:30:32.609
often the competitor is the status quo. So you

00:30:32.609 --> 00:30:35.230
created something that's like the status quo,

00:30:35.309 --> 00:30:38.670
like any... No, the status quo was paper and

00:30:38.670 --> 00:30:41.529
pencil and a calculator. I'm saying that it became

00:30:41.529 --> 00:30:43.750
now that that has... No, it was a status quo.

00:30:43.950 --> 00:30:47.349
It was pretty easy to beat us out if you were

00:30:47.349 --> 00:30:49.130
on a more powerful computer because you had more

00:30:49.130 --> 00:30:51.829
columns. You had graphing ability on the screen.

00:30:52.150 --> 00:30:55.450
You had lots more memory. And if you happen to

00:30:55.450 --> 00:30:57.930
know what people wanted, things we left out like

00:30:57.930 --> 00:31:01.369
commas in numbers or the thousand separators,

00:31:01.710 --> 00:31:04.690
the columns being different widths and some other

00:31:04.690 --> 00:31:06.809
things. And there were certain things we left

00:31:06.809 --> 00:31:09.269
out because they were hard to program or we weren't

00:31:09.269 --> 00:31:13.630
willing to make certain compromises that other

00:31:13.630 --> 00:31:16.690
products like 123 did do and did very well because

00:31:16.690 --> 00:31:19.309
they had those. And then when there was new hardware,

00:31:20.049 --> 00:31:23.000
so Excel could take advantage of fonts. Just

00:31:23.000 --> 00:31:27.640
think what multi -plan from Microsoft put in

00:31:27.640 --> 00:31:29.940
and then Excel used, which was the ability to

00:31:29.940 --> 00:31:34.720
have a single line between each cell, to have

00:31:34.720 --> 00:31:39.259
a grid visible. That's a huge advance for many

00:31:39.259 --> 00:31:42.119
applications. And they were able to do that because

00:31:42.119 --> 00:31:45.599
the hardware, initially of the Macintosh, the

00:31:45.599 --> 00:31:49.700
early Macintosh that they did, let you do that.

00:31:49.799 --> 00:31:53.240
So there are all these advances that you could

00:31:53.240 --> 00:31:57.400
put in new products. With Excel, you could run

00:31:57.400 --> 00:31:59.599
it under Windows. And it came bundled with your

00:31:59.599 --> 00:32:03.400
word processor with Office. And so whenever you

00:32:03.400 --> 00:32:06.359
bought your computer, it came with it. So that

00:32:06.359 --> 00:32:10.220
was a reason that people would go with that.

00:32:11.579 --> 00:32:15.339
So 123 was just much better for the IBM PC. It

00:32:15.339 --> 00:32:20.720
was tuned so much for it. was so tuned to the

00:32:20.720 --> 00:32:23.339
IBM PC, which had just come out, was now available

00:32:23.339 --> 00:32:25.619
from a major manufacturer. People now knew PCs

00:32:25.619 --> 00:32:27.940
were useful. So they were finally all buying

00:32:27.940 --> 00:32:30.940
them. So Jeff Moore's, you know, inside the tornado

00:32:30.940 --> 00:32:35.799
and about the early adopters now going to the

00:32:35.799 --> 00:32:40.079
early majority where there's a huge uptick. And

00:32:40.079 --> 00:32:43.640
IBM PC was the computer of the time. And the

00:32:43.640 --> 00:32:46.940
spreadsheet of the time was one, two, three.

00:32:47.099 --> 00:32:50.220
And it came even with a piece of plastic, molded

00:32:50.220 --> 00:32:52.960
plastic that went over the keyboard's function

00:32:52.960 --> 00:32:56.059
keys to label them. So you knew where help was

00:32:56.059 --> 00:32:58.500
and you knew where this was and all that because

00:32:58.500 --> 00:33:01.660
you had physical labels on this piece of plastic

00:33:01.660 --> 00:33:06.019
tied to the hardware so much. So why not buy

00:33:06.019 --> 00:33:10.250
that? And so what did 123 cost like? for $500,

00:33:10.890 --> 00:33:13.269
the computer with the printer and all that cost

00:33:13.269 --> 00:33:15.549
you several thousand, like three, four, five

00:33:15.549 --> 00:33:18.230
thousand dollars perhaps for your computer with

00:33:18.230 --> 00:33:20.650
a printer and all the things that you, the screens

00:33:20.650 --> 00:33:22.970
and the stuff that you wanted. So the software,

00:33:23.049 --> 00:33:24.930
who cares? But you bought it for the software,

00:33:25.190 --> 00:33:28.549
which is what it means to be a killer app. That

00:33:28.549 --> 00:33:31.829
it's worth buying the hardware for that software.

00:33:32.369 --> 00:33:34.599
Yeah. There you go. But thanks for tying that

00:33:34.599 --> 00:33:37.619
back in. Yeah. So the idea of having that killer

00:33:37.619 --> 00:33:39.920
app. The matter is when you already have the

00:33:39.920 --> 00:33:44.319
hardware. Right. So like when Netscape came in

00:33:44.319 --> 00:33:47.859
with a nice browser, the idea of that type of

00:33:47.859 --> 00:33:49.819
a browser, there had been browsers before, but

00:33:49.819 --> 00:33:52.819
they weren't integrated with screen, with pictures

00:33:52.819 --> 00:33:55.099
and stuff the way that it was. There were just

00:33:55.099 --> 00:33:57.799
a few and it was available for free. All you

00:33:57.799 --> 00:33:59.960
needed was a personal computer running Windows.

00:34:00.680 --> 00:34:04.039
and you needed to have that computer on a desk

00:34:04.039 --> 00:34:07.000
and you wanted it connected to the local area

00:34:07.000 --> 00:34:10.820
network to be able or somehow connected to what

00:34:10.820 --> 00:34:13.719
could be the internet or whatever. You needed

00:34:13.719 --> 00:34:16.360
that hardware all in place. And companies had

00:34:16.360 --> 00:34:19.139
that because of Lotus Notes, a product that Lotus

00:34:19.139 --> 00:34:22.159
did that was really good for email and workflow

00:34:22.159 --> 00:34:24.480
and all sorts of stuff. And it was one of the

00:34:24.480 --> 00:34:27.519
premier applications for Windows to help Windows

00:34:27.519 --> 00:34:30.449
being developed. And so the hardware was already

00:34:30.449 --> 00:34:33.289
there. You didn't need a killer app. All you

00:34:33.289 --> 00:34:36.130
needed was the good app. You don't have to buy

00:34:36.130 --> 00:34:39.750
the hardware if it was already there. And then,

00:34:39.750 --> 00:34:42.530
um, when Napster came around, you know, you don't

00:34:42.530 --> 00:34:46.170
need to buy anything. Right. When, now when you

00:34:46.170 --> 00:34:48.329
needed a music player is when the MP3 stuff,

00:34:48.429 --> 00:34:50.170
you know, and Napster and whatever came in, that

00:34:50.170 --> 00:34:53.150
was, you sort of needed that. Uh, if you wanted

00:34:53.150 --> 00:34:57.949
portable music. But when, you know, when Facebook

00:34:57.949 --> 00:35:01.429
came in, they already, students already had computers.

00:35:01.630 --> 00:35:05.170
They already were connected to whatever. So they,

00:35:05.309 --> 00:35:07.409
they didn't have to be a killer app that way.

00:35:07.809 --> 00:35:11.949
They just needed to be a useful app. Right. Almost

00:35:11.949 --> 00:35:15.659
like use the system that's in place. to then

00:35:15.659 --> 00:35:17.320
kind of to move things forward. Once you have

00:35:17.320 --> 00:35:19.860
a fertile environment, if you want a fertile

00:35:19.860 --> 00:35:22.380
environment, meaning it's a thing where it's

00:35:22.380 --> 00:35:25.539
a good place to do different types of applications

00:35:25.539 --> 00:35:28.480
and stuff that lends itself to innovating in

00:35:28.480 --> 00:35:32.679
different areas. So the PC was a fertile platform

00:35:32.679 --> 00:35:36.039
of sorts. And once you have that, you can do

00:35:36.039 --> 00:35:38.559
things. for that. And we're seeing that today

00:35:38.559 --> 00:35:41.300
with AI, where people are using AI systems that

00:35:41.300 --> 00:35:43.900
other build as the fertile platform in which

00:35:43.900 --> 00:35:45.719
to build something else, now that I have something

00:35:45.719 --> 00:35:50.000
that can take inputs and then do outputs in English.

00:35:50.079 --> 00:35:53.420
In terms of win something, it's hard for adoption

00:35:53.420 --> 00:35:56.239
versus easy for adoption. Right. Now we have

00:35:56.239 --> 00:36:00.460
adoption in... Just a thing I saw that David

00:36:00.460 --> 00:36:05.480
Smith wrote. David Smith wrote a product called

00:36:05.480 --> 00:36:10.590
Widget Smith. which is on the Apple iPhone that

00:36:10.590 --> 00:36:12.710
lets you put widgets, customize them. And he

00:36:12.710 --> 00:36:15.809
writes about it, how he was building it originally

00:36:15.809 --> 00:36:18.150
for the watch and then he was into customization

00:36:18.150 --> 00:36:21.869
and Apple made it possible to do that. And then

00:36:21.869 --> 00:36:23.750
he came out with it and it started selling like

00:36:23.750 --> 00:36:27.369
normal. Then somebody put up something on TikTok

00:36:27.369 --> 00:36:31.530
where they said, boy, you can customize exactly

00:36:31.530 --> 00:36:35.130
what I want. Customize my screen to be exactly

00:36:35.130 --> 00:36:38.519
what I want. Within hours he became the number

00:36:38.519 --> 00:36:41.559
one app for a week or for two weeks on the Apple

00:36:41.559 --> 00:36:45.440
iPhone Right. He sold millions and millions.

00:36:45.440 --> 00:36:47.280
I think what do you say a hundred million copies

00:36:47.280 --> 00:36:50.260
or something didn't sell them all But he you

00:36:50.260 --> 00:36:51.659
know, they're free. It was for free, but you

00:36:51.659 --> 00:36:56.219
can buy the the upgrades You know, I mean within

00:36:56.219 --> 00:37:00.679
within days or within hours today is quite different

00:37:00.679 --> 00:37:03.199
Right how fast things how fast you can go up

00:37:03.199 --> 00:37:05.900
and how fast you can go down. Yeah That's for

00:37:05.900 --> 00:37:10.239
sure. I wonder, are there any new criteria that

00:37:10.239 --> 00:37:14.559
you add in to making a killer app today? I haven't

00:37:14.559 --> 00:37:16.579
thought about it much, but I think the whole

00:37:16.579 --> 00:37:19.480
thing of whenever there's a new capability, look

00:37:19.480 --> 00:37:22.739
at what is it good for and better than anything

00:37:22.739 --> 00:37:27.820
before by not a little bit, but by a lot, 100

00:37:27.820 --> 00:37:30.400
times better type of thing. Right. Now that you

00:37:30.400 --> 00:37:34.960
have that, what does that enable You know, so

00:37:34.960 --> 00:37:37.320
let's take machine learning. Yeah, okay the old

00:37:37.320 --> 00:37:40.300
machine learning It was used for visual stuff

00:37:40.300 --> 00:37:43.260
to recognize things and really kind of cool on

00:37:43.260 --> 00:37:45.639
the production line You know, you could see is

00:37:45.639 --> 00:37:47.599
this a good one or a bad one and you could then

00:37:47.599 --> 00:37:50.280
automatically You know do some quality control

00:37:50.280 --> 00:37:53.760
and stuff Well, what happens if you can put it

00:37:53.760 --> 00:37:57.679
on a small enough ship to be able to detect people

00:37:57.679 --> 00:38:01.239
by eyes? Being able to say that's a face All

00:38:01.239 --> 00:38:03.019
you have to do is figure. Well, that's a really

00:38:03.019 --> 00:38:05.760
tough thing to find faces. OK, but we figured

00:38:05.760 --> 00:38:07.460
out machine learning and stuff. They figured

00:38:07.460 --> 00:38:09.480
out how to do that. And they put it into cameras.

00:38:10.000 --> 00:38:12.300
And we take this for granted. The fact that a

00:38:12.300 --> 00:38:14.519
camera can figure out it's a face and then do

00:38:14.519 --> 00:38:17.239
the autofocus there, not just the dot in the

00:38:17.239 --> 00:38:19.320
middle, wherever you point it used to be wherever

00:38:19.320 --> 00:38:22.599
you pointed it. That was the cool thing. Right.

00:38:22.760 --> 00:38:24.739
Where you pointed it was where it focused this.

00:38:24.900 --> 00:38:28.360
It's able to say, oh, I see a face. I see two

00:38:28.360 --> 00:38:32.989
faces. Let's try to figure that out. What was

00:38:32.989 --> 00:38:35.730
it good for? It was good for that. It was good

00:38:35.730 --> 00:38:38.469
for lots of other things. So what is it good

00:38:38.469 --> 00:38:41.510
for and that we could do that we couldn't do

00:38:41.510 --> 00:38:45.949
before that's worth it, that pays for itself

00:38:45.949 --> 00:38:49.510
almost immediately? I say within two weeks type

00:38:49.510 --> 00:38:52.750
of thing. When you got desktop publishing, a

00:38:52.750 --> 00:38:56.389
laser printer, something like PageMaker or later

00:38:56.389 --> 00:38:59.590
Word and stuff like that, being able to... get

00:38:59.590 --> 00:39:01.909
your output and, you know, do a newsletter or

00:39:01.909 --> 00:39:05.010
whatever. The old days, you had to type it up,

00:39:05.449 --> 00:39:07.949
send it out to a typeset, you know, to a person

00:39:07.949 --> 00:39:09.429
who would typeset it and then print it to the

00:39:09.429 --> 00:39:12.090
printer and come back. It paid for itself the

00:39:12.090 --> 00:39:14.030
first time you used it. There was a two -week

00:39:14.030 --> 00:39:17.010
payback for a lot of these things. VisiCalc paid

00:39:17.010 --> 00:39:20.250
for the Apple II. You know, there were certain...

00:39:20.110 --> 00:39:22.710
Programs you could use to do financial forecasting

00:39:22.710 --> 00:39:25.429
not as easy to use and whatever but for your

00:39:25.429 --> 00:39:28.010
particular application You could use a time -sharing

00:39:28.010 --> 00:39:30.289
system six hundred six thousand dollars you a

00:39:30.289 --> 00:39:33.730
month. Well for six thousand dollars That's much

00:39:33.730 --> 00:39:37.329
more than I to get the top end. You'd get an

00:39:37.329 --> 00:39:41.929
Apple to a nice monitor Sony's Trinitron probably

00:39:41.929 --> 00:39:45.030
or something that I could use to watch TV and

00:39:45.030 --> 00:39:48.420
a printer the best printer that I could get at

00:39:48.420 --> 00:39:50.199
the time, you know, a daisy wheel printer or

00:39:50.199 --> 00:39:52.500
something. So it pays for itself the first time

00:39:52.500 --> 00:39:55.280
I use it in a couple of weeks. No -brainer. I

00:39:55.280 --> 00:39:57.559
mean, it's obvious. None of this one -year payback,

00:39:57.659 --> 00:40:02.119
two -year payback. So, you know, cell phones,

00:40:02.820 --> 00:40:05.860
the first time you're stuck somewhere and can't,

00:40:05.940 --> 00:40:09.260
with a flat tire that you can't change, in the

00:40:09.260 --> 00:40:11.429
middle of nowhere, and you find out you could

00:40:11.429 --> 00:40:13.429
have had a cell phone you know when your mother

00:40:13.429 --> 00:40:15.070
does that you buy her a cell phone right away

00:40:15.070 --> 00:40:17.489
i did that you know she had a flat tire and you

00:40:17.489 --> 00:40:18.809
know whatever i said mom you're getting a cell

00:40:18.809 --> 00:40:22.170
phone right this is many many years ago um when

00:40:22.170 --> 00:40:25.849
but or late enough for cell phones so that's

00:40:25.849 --> 00:40:30.170
um you know when is it obvious that it pays for

00:40:30.170 --> 00:40:33.230
itself the first time you really need it or something

00:40:33.230 --> 00:40:38.170
so those that sort of gets into today's stuff

00:40:38.170 --> 00:40:43.559
um and separately is user interface. Why was

00:40:43.559 --> 00:40:47.820
my spreadsheet implementation versus other calculating

00:40:47.820 --> 00:40:51.960
things that came out? Why was mine so successful?

00:40:52.440 --> 00:40:54.980
Was it just the marketing? Was it just the cool

00:40:54.980 --> 00:40:58.800
name? Was it our implementation? Was it, you

00:40:58.800 --> 00:41:00.500
know, I mean, there were other, you know, was

00:41:00.500 --> 00:41:03.219
it because it was on the Apple II? I mean, why,

00:41:03.440 --> 00:41:06.960
what particularly, what was the things that made

00:41:06.960 --> 00:41:11.289
it, you know, but a lot of it, is user interface

00:41:11.289 --> 00:41:16.969
and stuff, and what were they good for, and is

00:41:16.969 --> 00:41:22.610
it fun? In the early days, when I first did computerized

00:41:22.610 --> 00:41:25.610
typesetting, in our office we had a typesetter,

00:41:26.010 --> 00:41:28.389
thing used chemicals and stuff, whatever that

00:41:28.389 --> 00:41:31.610
you could use as an output, like a printer. And

00:41:31.610 --> 00:41:33.730
out would come this paper, little shiny paper,

00:41:33.929 --> 00:41:38.750
that you could do fonts. you know, Times Roman

00:41:38.750 --> 00:41:41.650
and then Italic and all that stuff and different

00:41:41.650 --> 00:41:44.909
point sizes. And the first things we did were

00:41:44.909 --> 00:41:46.989
look like ransom notes. These are our status

00:41:46.989 --> 00:41:49.210
reports, you know, where we could have just made

00:41:49.210 --> 00:41:50.989
it look like a typewriter, but no, we wanted

00:41:50.989 --> 00:41:53.809
to do all that. And sure enough, when the first

00:41:53.809 --> 00:41:56.489
word processors came out that did fonts with

00:41:56.489 --> 00:41:58.449
laser printers, what did everybody do? They went

00:41:58.449 --> 00:42:00.190
overboard. They use it for everything, you know,

00:42:00.190 --> 00:42:02.289
and they made fonts like, you know, and it was

00:42:02.289 --> 00:42:04.150
the most garish thing. And now we've sort of

00:42:04.150 --> 00:42:07.869
settled down. Right. The same thing. It's the

00:42:07.869 --> 00:42:09.190
hammer. Once you have a hammer, everything looks

00:42:09.190 --> 00:42:11.150
like a nail. I'll use I love this new thing.

00:42:11.210 --> 00:42:12.750
Let's use it for this. Let's use it. I don't

00:42:12.750 --> 00:42:14.010
know what it's really good for, but I'll use

00:42:14.010 --> 00:42:16.670
it for everything. Right. With today's stuff,

00:42:17.269 --> 00:42:20.239
what's it good for? You know, I can ask it a

00:42:20.239 --> 00:42:21.599
question, it gives me an answer. That's good.

00:42:21.699 --> 00:42:23.340
Well, it turns out it's not very good for certain

00:42:23.340 --> 00:42:25.579
answers. Right. And it's really good for certain

00:42:25.579 --> 00:42:27.820
other answers, you know, and certain type of

00:42:27.820 --> 00:42:32.900
questions. And understanding that is important.

00:42:33.579 --> 00:42:37.820
Yeah, 100%. And I think the challenge right now,

00:42:37.820 --> 00:42:41.320
I mean, you've seen all of the cycles, you know...

00:42:41.320 --> 00:42:43.820
Lots of cycles. It goes back, you know, back

00:42:43.820 --> 00:42:47.380
to the Stone Age. Right. The AI winters now...

00:42:47.380 --> 00:42:52.980
Look at my wheel. Now we're in this, people argue

00:42:52.980 --> 00:42:56.300
where we are in terms of the, you know, the hype

00:42:56.300 --> 00:42:59.659
cycle for AI right now. I think we're really

00:42:59.659 --> 00:43:01.960
high. I think some people are getting into that

00:43:01.960 --> 00:43:04.340
trough of disillusionment also a little bit,

00:43:04.539 --> 00:43:07.840
but I wonder for you seeing it from your perspective,

00:43:07.840 --> 00:43:11.659
how do you separate kind of the hype of AI and

00:43:11.659 --> 00:43:14.000
the reality of it? Well, a lot of people hyping

00:43:14.000 --> 00:43:16.780
it don't know what it is or what its limitations

00:43:16.780 --> 00:43:20.239
are. Right. What I used to say, let's see, when

00:43:20.239 --> 00:43:23.679
this account first came out, you could explain

00:43:23.679 --> 00:43:27.780
it or show it to people, okay? And if you showed

00:43:27.780 --> 00:43:30.539
it to a computer person, they look at it and

00:43:30.539 --> 00:43:32.980
say, what's so special here? I can write a program

00:43:32.980 --> 00:43:35.280
that does that, you know, and I can do better

00:43:35.280 --> 00:43:38.679
and stuff like that. You show it to a regular

00:43:38.679 --> 00:43:40.900
person, a normal person, they say, computers

00:43:40.900 --> 00:43:42.500
can do anything. They can forecast the weather.

00:43:42.679 --> 00:43:44.679
They can tell you whatever they do. What's so

00:43:44.679 --> 00:43:47.159
special about that? But if you showed it to people

00:43:47.159 --> 00:43:50.679
who actually did that stuff, they would start

00:43:50.679 --> 00:43:52.219
shaking and say, here's my credit card, please

00:43:52.219 --> 00:43:53.579
take it, give me that, give me that, give me

00:43:53.579 --> 00:43:57.280
that. So the same thing's happening with a lot

00:43:57.280 --> 00:44:00.760
of stuff with like AI today, is people look at

00:44:00.760 --> 00:44:03.800
it and they say, they don't really understand

00:44:03.800 --> 00:44:05.840
its limitations or what's what, but it looks

00:44:05.840 --> 00:44:07.639
like it's real smart, like a person, it can do

00:44:07.639 --> 00:44:09.880
things I can't, and I can ask a question because

00:44:09.880 --> 00:44:13.059
of an answer. And it sounds like a person doing

00:44:13.059 --> 00:44:16.260
that who is that confident. must be right and

00:44:16.260 --> 00:44:18.719
it's a computer you know and it's trained one

00:44:18.719 --> 00:44:20.940
on the whole web and the web must have everything

00:44:20.940 --> 00:44:22.880
there must be everything on the web well we know

00:44:22.880 --> 00:44:24.840
that the not everything is on the web and not

00:44:24.840 --> 00:44:27.900
everybody puts stuff on the web so it's not trained

00:44:27.900 --> 00:44:29.900
on everything and the people who are training

00:44:29.900 --> 00:44:33.500
it to check it don't always know the right answer

00:44:33.500 --> 00:44:38.719
right um etc for me what i need in my application

00:44:38.719 --> 00:44:40.760
stuff i just saw with today or something that

00:44:40.760 --> 00:44:44.019
a lot of training was on certain populations

00:44:44.329 --> 00:44:47.130
of medical stuff is in the literature because

00:44:47.130 --> 00:44:49.449
they only did the studies on certain people they

00:44:49.449 --> 00:44:52.989
could get to. Right. But that's normal people.

00:44:53.130 --> 00:44:54.809
I mean, you should understand the, you know,

00:44:54.849 --> 00:44:56.949
well, maybe from my experience, you know, it's

00:44:56.949 --> 00:45:00.309
different that the training now is based on that

00:45:00.309 --> 00:45:05.349
as that's what it is. There's an old joke. about

00:45:05.349 --> 00:45:08.570
the policeman sees this drunk and he's looking

00:45:08.570 --> 00:45:11.289
around on the ground and he says, what's the

00:45:11.289 --> 00:45:14.429
matter? And he said, well, I lost my wallet at

00:45:14.429 --> 00:45:18.929
23rd and Walnut. He said, well, oh, well, we're

00:45:18.929 --> 00:45:21.449
at 19th and Walnut. Why are you looking here?

00:45:21.510 --> 00:45:24.889
He said, well, the light's better. And so that

00:45:24.889 --> 00:45:28.550
is that when we have the data, we use that, whatever

00:45:28.550 --> 00:45:32.409
we can get. Whatever is easy to measure is the

00:45:32.409 --> 00:45:34.829
thing that we measure even though that may not

00:45:34.829 --> 00:45:37.969
be a good measure but there are people who don't

00:45:37.969 --> 00:45:41.510
know that who just assume that it must be all

00:45:41.510 --> 00:45:46.989
-knowing and It's better than me. So therefore

00:45:46.989 --> 00:45:50.480
it's good. So so we learn people who really use

00:45:50.480 --> 00:45:52.559
it learn to use the tool, like those that learn

00:45:52.559 --> 00:45:54.800
to use the spreadsheet, those that learn to use

00:45:54.800 --> 00:45:56.880
the word processor, those that learn how to drive

00:45:56.880 --> 00:45:58.780
a car versus, if you don't know how to drive

00:45:58.780 --> 00:46:00.260
a car, you're gonna crash and you might kill

00:46:00.260 --> 00:46:02.780
yourself. Once you've learned how to drive a

00:46:02.780 --> 00:46:05.860
car and stuff like that, then you can do also,

00:46:05.920 --> 00:46:07.539
and if you're really good, you might be a race

00:46:07.539 --> 00:46:09.719
car driver or something and can do it, but you

00:46:09.719 --> 00:46:13.039
have to learn to do that. dancing, you know,

00:46:13.159 --> 00:46:14.780
learning how to dance. You know, some people

00:46:14.780 --> 00:46:16.400
are naturals and she goes, look at one other

00:46:16.400 --> 00:46:18.900
person. The same thing about these things. We

00:46:18.900 --> 00:46:20.780
have to learn how to use them and what they're

00:46:20.780 --> 00:46:24.039
good for. Unfortunately, they keep changing.

00:46:24.920 --> 00:46:29.860
So I become a prompt master. I'm really good

00:46:29.860 --> 00:46:34.139
at prompting for GPT -4 for three. And I know

00:46:34.139 --> 00:46:36.079
all of what is good and I know the way to do

00:46:36.079 --> 00:46:38.539
it and how to, you know, and then of course it's

00:46:38.539 --> 00:46:41.980
changed. and suddenly my skills may not be as

00:46:41.980 --> 00:46:45.260
useful, things that I worked really, and maybe

00:46:45.260 --> 00:46:47.400
there are different ways to convince it for this,

00:46:47.719 --> 00:46:50.880
and that's just with one type of input, with

00:46:50.880 --> 00:46:54.219
one type of output, and it's changing so fast.

00:46:56.420 --> 00:47:01.369
That's not a recipe to depend upon it. it's great

00:47:01.369 --> 00:47:04.349
for advancing. Are you going to depend, let your

00:47:04.349 --> 00:47:06.730
life depend upon it? Reporters would ask lots

00:47:06.730 --> 00:47:08.809
of questions and learn lots of stuff about an

00:47:08.809 --> 00:47:10.969
air, it's something and report about it. But

00:47:10.969 --> 00:47:12.750
if you knew that particular thing, you were at

00:47:12.750 --> 00:47:14.590
the event and you knew whatever, you knew what

00:47:14.590 --> 00:47:18.110
they wrote isn't exactly right. Close, but not

00:47:18.110 --> 00:47:20.530
necessarily exactly right. Especially if they

00:47:20.530 --> 00:47:23.409
weren't, you know, had been in that area and

00:47:23.409 --> 00:47:25.630
whatever. There's the same type of stuff here

00:47:25.630 --> 00:47:28.369
that we have something that, yes, it's pretty

00:47:28.369 --> 00:47:32.099
good. Is it perfect? Well, right now we're learning

00:47:32.099 --> 00:47:34.860
like encoding. We're doing vibe coding, whatever.

00:47:34.900 --> 00:47:37.099
You have to check what it does. Yeah. Well, if

00:47:37.099 --> 00:47:39.760
you don't know how to check, you know, if you

00:47:39.760 --> 00:47:41.699
don't know the type of failures that you might

00:47:41.699 --> 00:47:43.940
run, and we don't even know the type of mistakes

00:47:43.940 --> 00:47:47.099
that it might be prone to, that we have to look

00:47:47.099 --> 00:47:50.400
for what type of failures and what uses will

00:47:50.400 --> 00:47:53.599
be used for. And it keeps changing. So what it

00:47:53.599 --> 00:47:55.460
was good for, how do we know if it's still good

00:47:55.460 --> 00:47:57.980
at the thing that I want to do? It's going to

00:47:57.980 --> 00:48:02.159
take time. to settle these things out. Is it

00:48:02.159 --> 00:48:06.559
still better than not using it? For some things,

00:48:06.619 --> 00:48:09.940
it is. And some people who know how to use it,

00:48:10.079 --> 00:48:13.320
like any tool, can get great advantage. Anybody

00:48:13.320 --> 00:48:16.960
who had a spreadsheet, suddenly, in the early

00:48:16.960 --> 00:48:19.679
days of the Apple II, and nobody else knew they

00:48:19.679 --> 00:48:22.570
had it or what it was, thought they were... these

00:48:22.570 --> 00:48:25.809
geniuses you know they unbelieve you did you

00:48:25.809 --> 00:48:29.130
know what people would do is they would get a

00:48:29.130 --> 00:48:32.130
job quote it for this many hours of money go

00:48:32.130 --> 00:48:33.809
away and then eventually when they needed to

00:48:33.809 --> 00:48:35.469
do it spend the 15 minutes use the spreadsheet

00:48:35.469 --> 00:48:38.369
to do the answer and then come up with it and

00:48:38.369 --> 00:48:41.829
wow whatever i got my money's worth um you know

00:48:41.829 --> 00:48:44.329
we're we're in that that world for certain things

00:48:44.329 --> 00:48:49.170
right so we'll see um and what was the best way

00:48:49.170 --> 00:48:54.840
my idea of uh using the paddles was not a good

00:48:54.840 --> 00:48:59.679
idea on the Apple II. The mouse is good for certain

00:48:59.679 --> 00:49:02.400
things on the spreadsheet, but the arrow keys

00:49:02.400 --> 00:49:05.380
are good for others. Using the keyboard is better

00:49:05.380 --> 00:49:11.320
than reading into voice input for a spreadsheet

00:49:11.320 --> 00:49:15.320
with numbers and words and all that. You can

00:49:15.320 --> 00:49:18.000
tell by anybody's texting who texts you voice.

00:49:18.170 --> 00:49:20.550
You know whether, do you want to depend on your

00:49:20.550 --> 00:49:24.030
doctor using that type of stuff where the difference

00:49:24.030 --> 00:49:29.309
of voice recognition is life and death? You may

00:49:29.309 --> 00:49:31.650
want different type of inputs with the right

00:49:31.650 --> 00:49:35.409
type of feedback. The feedback that we get from

00:49:35.409 --> 00:49:38.110
today's systems sometimes is, well, I can read

00:49:38.110 --> 00:49:41.449
it and I can tell where it's good or bad, or

00:49:41.449 --> 00:49:43.010
something that doesn't matter because anything

00:49:43.010 --> 00:49:45.210
that gets me in the ballpark. In other words,

00:49:45.210 --> 00:49:47.800
I'm trying to get I'm trying to get within the

00:49:47.800 --> 00:49:49.659
ballpark. I don't have to get to the seat. I

00:49:49.659 --> 00:49:52.420
need to be able to get to the ballpark. So therefore,

00:49:52.559 --> 00:49:54.559
anything that gets me to the ballpark or nearby

00:49:54.559 --> 00:49:57.760
most of the time is a help. And you have to understand

00:49:57.760 --> 00:50:01.460
that. I mean, can you imagine if with today's

00:50:01.460 --> 00:50:04.559
systems at the error rate in terms of even small

00:50:04.559 --> 00:50:08.079
or big, you're using it to get to your seat in

00:50:08.079 --> 00:50:10.880
a ballpark or anything else. You think a lot

00:50:10.880 --> 00:50:12.880
of us would end up in the same seat by mistake

00:50:12.880 --> 00:50:15.099
and not in our others with, you know, I mean,

00:50:15.280 --> 00:50:17.659
we're talking about in a stadium, if you have

00:50:17.659 --> 00:50:20.380
0 .0001 mistakes, you're still going to have

00:50:20.380 --> 00:50:24.539
some people who are in the wrong seats. So you

00:50:24.539 --> 00:50:27.260
have to have the error correction and stuff.

00:50:27.639 --> 00:50:30.019
So I don't know how we're going to do that. I

00:50:30.019 --> 00:50:33.079
know that there's a real challenge with the interfaces

00:50:33.079 --> 00:50:37.300
to these different models that are using the

00:50:37.300 --> 00:50:42.000
technology that the LLMs use. and the type of

00:50:42.000 --> 00:50:45.159
training and whatever, what is the best use of

00:50:45.159 --> 00:50:49.019
this technology for what type of things? I'm

00:50:49.019 --> 00:50:51.900
really excited by new hardware that's coming

00:50:51.900 --> 00:50:55.880
out that will be able to do it apparently. Some

00:50:55.880 --> 00:50:58.780
of the new phones and stuff, the particular chips

00:50:58.780 --> 00:51:00.860
in them are powerful enough and coming with the

00:51:00.860 --> 00:51:03.619
right software to be able to do certain things

00:51:03.619 --> 00:51:07.960
even on... in your hand and what's that gonna

00:51:07.960 --> 00:51:11.179
open up? We're not gonna just be talking. You're

00:51:11.179 --> 00:51:12.780
not gonna wanna talk to it. You're in a noisy

00:51:12.780 --> 00:51:15.400
environment. You're gonna wanna be able to do

00:51:15.400 --> 00:51:19.800
other ways of interfacing. We're learning. Remember

00:51:19.800 --> 00:51:26.059
with a lot of the visual stuff, it uses hand

00:51:26.059 --> 00:51:27.519
motion. Well, is that really the best way or

00:51:27.519 --> 00:51:29.739
is it better to have a controller? We now move

00:51:29.739 --> 00:51:34.070
back to controllers. Remember the iPad? Right,

00:51:34.389 --> 00:51:37.449
the touch screen. Yeah, with no pen. Well, now

00:51:37.449 --> 00:51:39.630
you sort of must have a pen to be able to use

00:51:39.630 --> 00:51:41.969
some of the stuff on the iPad for some of the

00:51:41.969 --> 00:51:46.989
applications. But we wouldn't have a pen when

00:51:46.989 --> 00:51:53.050
we can do touch. But it turns out that the interface

00:51:53.050 --> 00:51:58.289
is better when you have the sharp point. and

00:51:58.289 --> 00:52:00.449
the ability that people have, but you also need

00:52:00.449 --> 00:52:02.849
others for those that can't use their hands that

00:52:02.849 --> 00:52:06.769
way. There's accessibility. There's where I'm

00:52:06.769 --> 00:52:11.449
using it, noisy environments. Where I can't,

00:52:11.449 --> 00:52:14.590
how do I present things to you so you know what

00:52:14.590 --> 00:52:19.309
to check? Write me a 10 ,000 word treatise on

00:52:19.309 --> 00:52:21.449
such and such. Well, how am I gonna check if

00:52:21.449 --> 00:52:24.340
it's right? How do I know where to check? I mean,

00:52:24.559 --> 00:52:27.320
it's the teacher will just use an LLM to tell

00:52:27.320 --> 00:52:30.219
you if it's right or not. No, I'm just, you know,

00:52:30.300 --> 00:52:33.059
and what's that trained on, you know, what, you

00:52:33.059 --> 00:52:36.280
know, what class is it then? And do you trust

00:52:36.280 --> 00:52:38.159
the person who trained the train, you know, to

00:52:38.159 --> 00:52:40.480
train those systems? There are people you can

00:52:40.480 --> 00:52:43.059
read about the people. They get some pretty good

00:52:43.059 --> 00:52:46.150
people to train these systems, you know. to say

00:52:46.150 --> 00:52:48.349
no, you know, no, LLM, you got it wrong, you

00:52:48.349 --> 00:52:50.650
know, whatever, and it fixes it and stuff. They

00:52:50.650 --> 00:52:52.829
still do weird stuff, you know, weird stuff.

00:52:53.309 --> 00:52:55.349
I think it's that's one of the flaws, actually,

00:52:55.530 --> 00:52:58.469
that we're using human preferences. So it's.

00:52:59.179 --> 00:53:02.519
Trained to output things that humans prefer and

00:53:02.519 --> 00:53:04.639
that's sometimes just off of like what you see

00:53:04.639 --> 00:53:06.619
and what you look like and sometimes the formatting

00:53:06.619 --> 00:53:09.280
of an answer can cause you to like it more than

00:53:09.280 --> 00:53:12.900
some other answer and the fact how Factual it

00:53:12.900 --> 00:53:15.519
really is isn't always checked. There's so many

00:53:15.519 --> 00:53:17.960
there's so many challenges for it We're so early,

00:53:17.960 --> 00:53:20.860
but it's interesting to see the way the pendulum

00:53:20.860 --> 00:53:23.199
swings It's interesting to see hat what the new

00:53:23.199 --> 00:53:26.059
interfaces are gonna are gonna be like for it.

00:53:26.059 --> 00:53:28.699
Yeah Um, you raised so many good, do we need

00:53:28.699 --> 00:53:30.659
new hardware? Do we knew? Well, first of all,

00:53:30.659 --> 00:53:33.599
we knew, we knew we needed new hardware to do

00:53:33.599 --> 00:53:35.820
the calculation. So therefore they've been putting

00:53:35.820 --> 00:53:37.760
the neural engines and all that, and all these

00:53:37.760 --> 00:53:40.480
hardware. That's okay. So we've done that, that

00:53:40.480 --> 00:53:43.099
at least, but is there other hardware? Like we

00:53:43.099 --> 00:53:46.260
needed bitmap displays, meaning you could, you

00:53:46.260 --> 00:53:48.659
could address every dot on the screen in order

00:53:48.659 --> 00:53:52.059
to be able to do, you know, uh, mouse and stuff

00:53:52.059 --> 00:53:56.210
like that. What other hardware, what interfaces,

00:53:56.829 --> 00:54:01.510
input -output devices, do we need for the particular

00:54:01.510 --> 00:54:03.969
applications for which these things are useful

00:54:03.969 --> 00:54:06.809
for? Like if you don't have an autofocus motor,

00:54:07.869 --> 00:54:10.710
what good is being able to recognize where the

00:54:10.710 --> 00:54:14.789
person isn't focusing, right? It'll be interesting

00:54:14.789 --> 00:54:19.110
to see how those co -evolve with each other as

00:54:19.110 --> 00:54:22.429
these new capabilities, you know, arise and emerge

00:54:22.429 --> 00:54:24.550
and then there's going to be new hardware. I

00:54:24.550 --> 00:54:27.170
don't think we know yet what the right interface

00:54:27.170 --> 00:54:29.289
is exactly to take advantage of all of these

00:54:29.289 --> 00:54:31.909
things. But it is interesting for each of these

00:54:31.909 --> 00:54:35.650
things, which interfaces for each of these things.

00:54:35.750 --> 00:54:38.969
Right. Is it going to be right for we still know

00:54:38.969 --> 00:54:41.750
people still buy laptops even though they have

00:54:41.750 --> 00:54:45.210
just as powerful phones. Why? Well, there must

00:54:45.210 --> 00:54:48.409
be something about that laptop that is better

00:54:48.409 --> 00:54:50.090
for them that they're willing to spend a thousand

00:54:50.090 --> 00:54:54.420
bucks. or a few hundred to 2000 bucks for that

00:54:54.420 --> 00:54:57.719
laptop versus their phone that they're also spending

00:54:57.719 --> 00:54:59.820
hundreds of dollars or a thousand dollars for.

00:55:00.579 --> 00:55:05.300
And so we have to sort of understand that whole

00:55:05.300 --> 00:55:09.380
thing. Yeah, yeah. You raise so many interesting

00:55:09.380 --> 00:55:12.619
points, so many things to think about. It's hard

00:55:12.619 --> 00:55:14.659
to be an engineer because there's so many questions.

00:55:14.679 --> 00:55:17.920
And how do you decide, now's the time for me

00:55:17.920 --> 00:55:21.590
to bet on this? This combination, is it too early

00:55:21.590 --> 00:55:24.670
or is it too late? Am I choosing the wrong combinations,

00:55:24.789 --> 00:55:27.349
the wrong application? And then there's the old

00:55:27.349 --> 00:55:31.210
market fit, do people care? And a lot we don't

00:55:31.210 --> 00:55:34.769
know. That's why it was interesting to see David

00:55:34.769 --> 00:55:38.590
Smith, what he wrote about how he... He didn't

00:55:38.590 --> 00:55:41.409
know that there were I mean he cared about fonts

00:55:41.409 --> 00:55:44.329
particular you choosing this font and this color

00:55:44.329 --> 00:55:47.510
and this type of little layout whatever the style

00:55:47.510 --> 00:55:51.250
for a clock or the showing the weather because

00:55:51.250 --> 00:55:53.769
you cared about He cared about that, but he didn't

00:55:53.769 --> 00:55:55.690
know that there were a lot of people that did

00:55:55.690 --> 00:55:57.349
care He thought most people didn't because on

00:55:57.349 --> 00:55:59.550
other products they didn't care But it turns

00:55:59.550 --> 00:56:02.010
out on your home screen. There are people who

00:56:02.010 --> 00:56:03.630
view the home screen like they view the case,

00:56:03.630 --> 00:56:07.119
right? and they view their clothes or something

00:56:07.119 --> 00:56:09.579
like that, that that does matter to them. And

00:56:09.579 --> 00:56:11.719
that's how they express themselves. He happened

00:56:11.719 --> 00:56:14.480
to hit the right combination that others did

00:56:14.480 --> 00:56:16.920
not. They also had some stuff, but they didn't

00:56:16.920 --> 00:56:19.639
allow you to do the chartreuse that you wanted

00:56:19.639 --> 00:56:22.780
or something like that that made all the difference

00:56:22.780 --> 00:56:25.139
in the world to some people. And that was good

00:56:25.139 --> 00:56:27.880
for them. That's true of a lot of you don't know

00:56:27.880 --> 00:56:31.869
the space. You're not that type of surgeon. You

00:56:31.869 --> 00:56:35.010
may not know the needs for that type of surgery.

00:56:35.829 --> 00:56:38.929
If you're not that type of business person, you

00:56:38.929 --> 00:56:43.409
may not know the needs of that. Oh, the person

00:56:43.409 --> 00:56:46.250
using it works in this environment, physical

00:56:46.250 --> 00:56:49.429
environment. What must that hardware have? They

00:56:49.429 --> 00:56:51.949
only have one hand to do because the other hand

00:56:51.949 --> 00:56:54.630
is busy. So they have to do one -handed operation.

00:56:55.769 --> 00:56:59.369
There are just so many things. That you and you

00:56:59.369 --> 00:57:01.030
don't always know it and they have to go right

00:57:01.030 --> 00:57:02.969
for you to be successful. Yeah, it's a good point

00:57:02.969 --> 00:57:06.710
I mean, yeah taking into account When you're

00:57:06.710 --> 00:57:08.849
trying to design anything right, you know who

00:57:08.849 --> 00:57:10.949
your end user is what what environment they're

00:57:10.949 --> 00:57:14.590
gonna be using you're Offering in the context

00:57:14.590 --> 00:57:16.570
that they're in the experience that they've had

00:57:16.570 --> 00:57:19.170
in trying to create something that can kind of

00:57:19.170 --> 00:57:22.090
address You know all of those things. That's

00:57:22.090 --> 00:57:25.190
why I think that Robots, you know humanoidish

00:57:25.190 --> 00:57:28.619
robots might be interesting in that So much of

00:57:28.619 --> 00:57:32.159
our society in the world around us is built around

00:57:32.159 --> 00:57:35.340
a person within a certain range with one or two

00:57:35.340 --> 00:57:37.039
hands, you know, with this many fingers and whatever,

00:57:37.440 --> 00:57:39.920
and with this ability to move, et cetera, and

00:57:39.920 --> 00:57:43.840
this strength, you know, and speed that therefore...

00:57:43.659 --> 00:57:47.159
maybe those input -output devices, which is the

00:57:47.159 --> 00:57:51.000
humanoid robot, may be appropriate to fit in

00:57:51.000 --> 00:57:54.360
situations that we couldn't automate without

00:57:54.360 --> 00:57:57.340
specialized machines. It's a real challenge,

00:57:57.800 --> 00:58:01.219
but what are the physical configurations that

00:58:01.219 --> 00:58:04.860
are appropriate for our physical world? You know,

00:58:05.000 --> 00:58:08.119
the fact that a cell phone fits in your pocket

00:58:08.119 --> 00:58:12.179
matters when the first scientific, really good

00:58:12.179 --> 00:58:14.800
scientific calculator came out, the Hewlett -Packard

00:58:14.800 --> 00:58:20.280
calculator, HP 35. The story is that Packard,

00:58:20.360 --> 00:58:22.800
Hewlett -Packard came to his engineers and said,

00:58:23.179 --> 00:58:26.239
I want something that fits in my pocket, my shirt

00:58:26.239 --> 00:58:29.860
pocket. They took out their measurement and they

00:58:29.860 --> 00:58:32.420
measured his pocket. And sure enough, it could

00:58:32.420 --> 00:58:35.179
fit. It was, you know, a certain thing that he

00:58:35.179 --> 00:58:37.559
could fit in his pocket because he needed to

00:58:37.559 --> 00:58:41.369
be able to carry it around, you know. Um, before

00:58:41.369 --> 00:58:43.769
that we had slide rules and we had special strap,

00:58:44.210 --> 00:58:47.389
leather straps that fit them on our, on our belts.

00:58:47.969 --> 00:58:50.369
And that wasn't for regular people. This you

00:58:50.369 --> 00:58:53.849
wanted to, and that size calculator became dominant.

00:58:54.409 --> 00:58:57.289
Right. And same thing with the phone. It fits

00:58:57.289 --> 00:59:00.670
in the pocket ones that were too big, weren't

00:59:00.670 --> 00:59:03.750
right at first. Right. Once they, once the, um,

00:59:04.050 --> 00:59:06.929
like the Nokia, uh, small cell phones came out

00:59:06.929 --> 00:59:09.519
that you could fit in a purse. and you can fit

00:59:09.519 --> 00:59:12.119
in your pocket. That fitting in a purse and fitting

00:59:12.119 --> 00:59:14.840
in a pocket make a difference. What's really

00:59:14.840 --> 00:59:18.739
nice about the new iPhone Air, it's really, really

00:59:18.739 --> 00:59:22.360
thin and fits in your purse a lot better, among

00:59:22.360 --> 00:59:25.280
other things, which is packed. You put your phone

00:59:25.280 --> 00:59:27.960
right. Right. I mean, going back to what we were

00:59:27.960 --> 00:59:29.539
talking about earlier in terms of like how the

00:59:29.539 --> 00:59:31.679
pendulum swings back and forth. Yeah, like cell

00:59:31.679 --> 00:59:34.820
phones were smaller at one point in some ways,

00:59:34.840 --> 00:59:37.079
and they now have gotten bigger and now they're

00:59:37.079 --> 00:59:39.179
going back to being a little bit smaller and

00:59:39.179 --> 00:59:41.360
thinner. And then there's other tradeoffs. Now

00:59:41.360 --> 00:59:43.440
we're doing folding together. Right now there's

00:59:43.440 --> 00:59:45.559
the back to the flip. And then there was a time

00:59:45.559 --> 00:59:48.340
when who would ever have a full touch screen.

00:59:49.219 --> 00:59:52.099
And then there's other tradeoffs in terms of

00:59:52.099 --> 00:59:55.260
when you make something so small yet so interactive.

00:59:55.280 --> 00:59:57.139
right, there's trade -offs in terms of battery

00:59:57.139 --> 00:59:59.340
life, right, and things like that, and then you

00:59:59.340 --> 01:00:01.960
have to weigh, you know, those trade -offs, so

01:00:01.960 --> 01:00:03.659
that goes into... What's the sweet spot? We don't

01:00:03.659 --> 01:00:06.340
know what the sweet spot's going to be, and luckily,

01:00:06.579 --> 01:00:09.320
lots of people are trying, and we're seeing what

01:00:09.320 --> 01:00:12.280
sticks. Right, right. Which I guess is just the

01:00:12.280 --> 01:00:15.099
exciting thing about technology and evolution.

01:00:15.440 --> 01:00:16.980
Always been. Yeah. you know, look what happened

01:00:16.980 --> 01:00:19.320
with automobiles, you know, and stuff, you know,

01:00:19.480 --> 01:00:21.860
automatic transmission, you know, and stuff like

01:00:21.860 --> 01:00:24.239
that, you know, well, anti -lock brakes and whatever.

01:00:24.619 --> 01:00:27.159
And then something's caught on and then it became

01:00:27.159 --> 01:00:30.300
standard. You know, I don't know how, how to

01:00:30.300 --> 01:00:32.460
drive a, you know, an older car that doesn't

01:00:32.460 --> 01:00:34.639
have the backup checking and a backup camera

01:00:34.639 --> 01:00:36.739
and whatever, you know. And then we become so

01:00:36.739 --> 01:00:38.659
used to it, you know, then you start to become

01:00:38.659 --> 01:00:41.280
so used to it. Yeah. Which, which is. That's

01:00:41.280 --> 01:00:43.039
why I'm really scared in my car that doesn't

01:00:43.039 --> 01:00:45.599
have those. You know, it has a backup camera,

01:00:45.679 --> 01:00:47.980
but one of my cars doesn't have the sensors.

01:00:48.420 --> 01:00:51.280
Right. And now we're so used to hearing that

01:00:51.280 --> 01:00:53.500
sensor when you turn the blinker on that, you

01:00:53.500 --> 01:00:55.559
know, that if someone's in your thing, I, you

01:00:55.559 --> 01:00:57.780
just always need to check, especially as you

01:00:57.780 --> 01:01:02.840
get your neck back and all. So coming towards

01:01:02.840 --> 01:01:05.119
the end, I have two questions that I like to

01:01:05.119 --> 01:01:07.440
ask. What advice would you give your younger

01:01:07.440 --> 01:01:11.269
self? Maybe when you were kind of just say starting

01:01:11.269 --> 01:01:13.969
your career. Oh the career advice as opposed

01:01:13.969 --> 01:01:15.929
to personal advice Well, you could do both but

01:01:15.929 --> 01:01:20.269
personal advice is yes. She's the right one Okay,

01:01:20.670 --> 01:01:23.570
so that's you know, okay that's very important

01:01:23.570 --> 01:01:25.610
finding the right that's very important very

01:01:25.610 --> 01:01:28.389
important a finding a partner is is You know

01:01:28.389 --> 01:01:30.690
important and so what advice what would I want

01:01:30.690 --> 01:01:33.230
to do the different? I mean because I I really

01:01:33.230 --> 01:01:35.630
like how things turned out I mean, things could

01:01:35.630 --> 01:01:36.909
have turned out different where I could have

01:01:36.909 --> 01:01:38.630
made lots and lots of money, which I did not

01:01:38.630 --> 01:01:41.070
because of various business questions and stuff

01:01:41.070 --> 01:01:43.289
like that. On the other hand, I like my life

01:01:43.289 --> 01:01:46.030
and it's better than I expected it to be. So

01:01:46.030 --> 01:01:48.250
I don't know what advice that would change what

01:01:48.250 --> 01:01:51.809
I would have done except that, let's see, I mean,

01:01:52.150 --> 01:01:54.269
choosing the right, you know, right horses to

01:01:54.269 --> 01:01:57.090
get on. I mean, that's the, this, which technology

01:01:57.090 --> 01:01:59.690
one to ride, when, how to make that decision.

01:02:00.110 --> 01:02:01.769
But I still don't know. I mean, all the things

01:02:01.769 --> 01:02:04.500
I've told you. Some of it comes from what I've

01:02:04.500 --> 01:02:06.800
said way. I was saying back then that I learned

01:02:06.800 --> 01:02:09.199
then that's kind of that's kind of tough You

01:02:09.199 --> 01:02:12.980
know family matters enjoy Enjoy the good times

01:02:12.980 --> 01:02:15.260
because they may be the good times There's movie

01:02:15.260 --> 01:02:17.340
as good as it gets there are some things that

01:02:17.340 --> 01:02:19.239
are as good as it gets enjoy them when you have

01:02:19.239 --> 01:02:21.519
them When you have the really good time enjoy

01:02:21.519 --> 01:02:24.159
that don't feel bad that it could be or whatever.

01:02:24.179 --> 01:02:27.420
I've had some really Special things happen in

01:02:27.420 --> 01:02:31.159
my life, you know and been at unusual situations

01:02:31.159 --> 01:02:34.219
and no longer have that So the more that hopefully

01:02:34.219 --> 01:02:36.480
I enjoyed them at the time that I could enjoy

01:02:36.480 --> 01:02:40.480
them, that's good. Right. There's like a great

01:02:40.480 --> 01:02:43.980
quote that you reminded me of, of it's like about

01:02:43.980 --> 01:02:46.579
the good old days. Right. And sometimes you don't

01:02:46.579 --> 01:02:48.639
know, like I wish there was a way to know you're

01:02:48.639 --> 01:02:51.119
in the good old days before you've actually left

01:02:51.119 --> 01:02:54.019
them. It's from the office. It's a good quote

01:02:54.019 --> 01:02:57.039
that I recently came across again. Now, many,

01:02:57.099 --> 01:03:00.630
many years ago, early in the days of Apple, Steve

01:03:00.630 --> 01:03:03.289
Jobs came to speak at the Harvard Business School

01:03:03.289 --> 01:03:06.250
to an entrepreneurship club or something. And

01:03:06.250 --> 01:03:08.070
I went to go visit. I was a graduate at the time.

01:03:08.150 --> 01:03:11.969
I had graduated at the time. So I went back and

01:03:11.969 --> 01:03:14.949
I raised my hand and said, hey Steve, you know,

01:03:15.010 --> 01:03:18.309
where are you, what are you expecting for the

01:03:18.309 --> 01:03:20.650
future? Where are you trying to go? And Steve

01:03:20.650 --> 01:03:23.809
said, it's the journey. He was not looking for

01:03:23.809 --> 01:03:26.510
where he was. It was the journey. And that, you

01:03:26.510 --> 01:03:28.460
know, kind of, oh. You know, I had asked the

01:03:28.460 --> 01:03:30.599
same thing Gates at some point and, you know,

01:03:30.719 --> 01:03:32.719
Gates had, I'd asked him about some stuff and

01:03:32.719 --> 01:03:35.280
he talked about how he was going to give money

01:03:35.280 --> 01:03:39.019
away. And he sure did, you know, that, you know,

01:03:39.119 --> 01:03:42.119
this was his success. Look, we'll look in the

01:03:42.119 --> 01:03:44.780
future and if he's successful, he'll, uh, you

01:03:44.780 --> 01:03:47.380
know, he'll, he'll get back that way. So that

01:03:47.380 --> 01:03:50.960
was the thing about enjoy the journey. I always,

01:03:51.000 --> 01:03:54.489
when I did accompany What I did was when I was

01:03:54.489 --> 01:03:56.710
paying people and sending stuff I said I want

01:03:56.710 --> 01:03:59.730
to do it that when you look back if we if we

01:03:59.730 --> 01:04:04.949
fail I want you not to Feel bad about that. You

01:04:04.949 --> 01:04:07.230
wasted your time here. I wanted you to say, you

01:04:07.230 --> 01:04:10.190
know, I wish I hadn't You know, I want to make

01:04:10.190 --> 01:04:14.289
sure that I paid you enough that it isn't costing

01:04:14.289 --> 01:04:16.449
you to to become poor or something like that

01:04:16.449 --> 01:04:18.869
I wouldn't that was now that may not be the best

01:04:18.869 --> 01:04:23.380
business But I was a feeling I had that I wanted

01:04:23.380 --> 01:04:26.400
you want to do it so that you won't regret That

01:04:26.400 --> 01:04:29.719
you were doing that stuff because of what doesn't

01:04:29.719 --> 01:04:32.260
if it doesn't pan out from that viewpoint so

01:04:32.260 --> 01:04:34.320
do things that's the thing of doing it what you

01:04:34.320 --> 01:04:37.659
love and You can't always follow that. You don't

01:04:37.659 --> 01:04:40.019
always have that option. But if you can that's

01:04:40.019 --> 01:04:45.090
pretty good Yeah, it reminds me of your earlier

01:04:45.090 --> 01:04:47.989
remark. Life's a journey, not a destination.

01:04:48.949 --> 01:04:51.550
We might be seeking all of these things, but

01:04:51.550 --> 01:04:53.730
it's really about the journey that gets us to

01:04:53.730 --> 01:04:56.730
where we are. I'm afraid that this last question

01:04:56.730 --> 01:04:59.469
is too close to that last one, but I always ask

01:04:59.469 --> 01:05:04.510
it, so I have to. What has a career in tech taught

01:05:04.510 --> 01:05:10.610
you about life? I love tech, and I love that.

01:05:10.730 --> 01:05:13.969
And I'm happy that I did it. So I'm happy I chose

01:05:13.969 --> 01:05:17.010
that and that I'm still doing it. I'm still programming.

01:05:17.750 --> 01:05:19.730
You know, I'm still writing and I like, I'm still

01:05:19.730 --> 01:05:22.050
programming a lot by hand. You know, cause I

01:05:22.050 --> 01:05:25.349
like, it's like people who like to paint. They're

01:05:25.349 --> 01:05:28.090
not Picasso. They're not Rembrandt, but they

01:05:28.090 --> 01:05:30.289
like to paint. Even though they're not, you know,

01:05:30.510 --> 01:05:33.000
they're people like to play basketball. They're

01:05:33.000 --> 01:05:36.719
not LeBron, but they still like to play. You

01:05:36.719 --> 01:05:39.079
don't have to be the best. So that's, that's

01:05:39.079 --> 01:05:41.280
one of the things about, you know, about your

01:05:41.280 --> 01:05:43.159
career. But you also should think about what

01:05:43.159 --> 01:05:46.960
you're good at and what you like. You know, there's

01:05:46.960 --> 01:05:49.199
these retrospective I saw last night about Robert

01:05:49.199 --> 01:05:52.840
Redford and about what he went through figuring

01:05:52.840 --> 01:05:56.679
out what his calling was in life, where, you

01:05:56.679 --> 01:05:59.820
know, he realized he was in arts and he wanted,

01:05:59.940 --> 01:06:02.570
he thought he was going to be a painter. to france

01:06:02.570 --> 01:06:03.929
and stuff and he thought he was going to be a

01:06:03.929 --> 01:06:07.210
painter but then he ended up you know in school

01:06:07.210 --> 01:06:09.929
for for drama and stuff like that and a teacher

01:06:09.929 --> 01:06:12.349
who told him you're you're good to stay in that

01:06:12.349 --> 01:06:15.389
you know type of stuff and then he did he ended

01:06:15.389 --> 01:06:18.329
up on stage and then he ended up in one film

01:06:18.329 --> 01:06:20.690
and then he ended up behind the camera you know

01:06:20.690 --> 01:06:25.610
as a director and figuring out what and then

01:06:25.610 --> 01:06:27.550
then of course Sundance in terms of giving it

01:06:27.550 --> 01:06:32.389
back in terms of his love of The outdoor of of

01:06:32.389 --> 01:06:34.530
outdoors and stuff like that. That's an interesting

01:06:34.530 --> 01:06:37.650
thing to think about Look at people's paths how

01:06:37.650 --> 01:06:40.170
they got where they got especially ones that

01:06:40.170 --> 01:06:44.590
seem happy with the path that they took and Why

01:06:44.590 --> 01:06:48.329
and so that's a good reason to read biographies

01:06:48.329 --> 01:06:51.989
of various people that have done stuff to talk

01:06:51.989 --> 01:06:54.730
to people Who would not have written a biography

01:06:54.730 --> 01:06:58.230
or there being a biography of them? To find out

01:06:58.230 --> 01:07:01.010
and about that somebody contacted me that I knew

01:07:01.010 --> 01:07:03.989
from, from high school and said, Dan, I hear

01:07:03.989 --> 01:07:06.489
that you took pictures back in the old days and

01:07:06.489 --> 01:07:09.750
you have pictures of the play I was in. And so

01:07:09.750 --> 01:07:12.869
that's, you know, yeah. Um, and you still have

01:07:12.869 --> 01:07:14.210
them. I said, yeah, I scanned them for one of

01:07:14.210 --> 01:07:16.849
the other people who was his kids, didn't believe

01:07:16.849 --> 01:07:18.889
that he was ever in a play and I was the lead.

01:07:19.250 --> 01:07:23.110
And so sure enough, so I had the pictures and

01:07:23.110 --> 01:07:25.750
this turned out to be a major memory in her life.

01:07:26.250 --> 01:07:29.090
I was into photography. And I had pictures, I

01:07:29.090 --> 01:07:31.730
had great pictures of her. And I found out something

01:07:31.730 --> 01:07:34.250
I didn't know. She was a shy person and she had

01:07:34.250 --> 01:07:37.650
a crush on a guy. And she got Maria and he was

01:07:37.650 --> 01:07:43.190
Tony. And they, boy, I have pictures. You got

01:07:43.190 --> 01:07:45.849
to live out a fantasy. You get to live out all

01:07:45.849 --> 01:07:48.409
sorts of fantasies in life sometimes that you

01:07:48.409 --> 01:07:50.110
don't expect. I mean, they're not married and

01:07:50.110 --> 01:07:51.329
they haven't, you know, they're friends, but

01:07:51.329 --> 01:07:53.010
they, you know, they're different parts of the

01:07:53.010 --> 01:07:55.590
world, whatever. But looking at that piece of

01:07:55.590 --> 01:07:58.500
her life. And that turned out to be a major,

01:07:58.860 --> 01:08:02.019
in hindsight for the two of them, that play turned

01:08:02.019 --> 01:08:04.980
out to be, in high school, turned out to be a

01:08:04.980 --> 01:08:09.519
major good point along the way. I'm sure in terms

01:08:09.519 --> 01:08:12.579
of other things in their lives, they've had some

01:08:12.579 --> 01:08:14.539
other high points. I mean, I'm a grandparent

01:08:14.539 --> 01:08:17.260
and there are just such incredible high points

01:08:17.260 --> 01:08:22.500
you can get as a grandparent, and stuff like

01:08:22.500 --> 01:08:25.399
that. And being a parent and the same thing,

01:08:26.530 --> 01:08:30.689
So enjoy those and they will happen and don't

01:08:30.689 --> 01:08:32.989
belittle them when you have them. So that's,

01:08:32.989 --> 01:08:35.170
that's advice that I have, including in your

01:08:35.170 --> 01:08:37.649
career. You know, I mean, even if you're going

01:08:37.649 --> 01:08:40.210
to get fired at some point, you may have, you

01:08:40.210 --> 01:08:42.630
got to speak in front of a huge crowd and they

01:08:42.630 --> 01:08:44.930
clapped because they liked what you said and

01:08:44.930 --> 01:08:47.550
you liked that feeling. Right. Maybe, I don't

01:08:47.550 --> 01:08:49.970
know. Or you have a product, you built a product

01:08:49.970 --> 01:08:52.050
and people actually use it. And then you get

01:08:52.489 --> 01:08:55.609
I did a product where you draw on the screen,

01:08:55.689 --> 01:08:57.930
but you can draw big and it shrinks it down.

01:08:58.149 --> 01:09:02.930
You could mark up PDFs, okay? So on an Apple,

01:09:03.069 --> 01:09:06.930
on an iPad, even though you didn't have to write

01:09:06.930 --> 01:09:09.609
real small. So I hear from this mother who said,

01:09:09.750 --> 01:09:12.649
thank God, I really love your product. My daughter

01:09:12.649 --> 01:09:14.829
has cerebral palsy and now she can do her homework

01:09:14.829 --> 01:09:20.100
herself because she's able to draw big. on the

01:09:20.100 --> 01:09:22.039
homework and she doesn't have to tell me what

01:09:22.039 --> 01:09:27.720
to do. Thank you. I mean, that was like, I still

01:09:27.720 --> 01:09:29.819
remember, it was years ago. I still remember

01:09:29.819 --> 01:09:33.460
that moment. So you never know. There's a line

01:09:33.460 --> 01:09:36.859
in the play rent, you know, about helping other

01:09:36.859 --> 01:09:39.079
people and stuff like, I mean, and you know,

01:09:39.079 --> 01:09:41.460
this thing about the relationship with other

01:09:41.460 --> 01:09:46.239
people, you know, is part of what it's all about.

01:09:46.899 --> 01:09:50.859
So enjoy those. If you do, there may be other

01:09:50.859 --> 01:09:53.800
things you enjoy more, enjoy them. Ice cream.

01:09:55.439 --> 01:09:59.159
I think it's really good advice and it reminds

01:09:59.159 --> 01:10:03.220
people basically, like, you never know what the

01:10:03.220 --> 01:10:04.899
consequences of your actions are, but if you're

01:10:04.899 --> 01:10:06.920
kind of doing it and you have the right intentions

01:10:06.920 --> 01:10:08.539
and you're doing things for the right reasons,

01:10:09.100 --> 01:10:12.979
you could make an unbelievable, you know, you

01:10:12.979 --> 01:10:16.180
can have an unbelievable impact on other people.

01:10:17.230 --> 01:10:18.789
I think that, yeah, that's really beautiful.

01:10:19.090 --> 01:10:22.369
The speech that I give that the TED Talk is a

01:10:22.369 --> 01:10:24.529
piece of, I've had people years later say that

01:10:24.529 --> 01:10:26.609
I was inspired by that and I went and did this

01:10:26.609 --> 01:10:28.189
thing totally different and whatever, and this

01:10:28.189 --> 01:10:29.930
product that we all know about is there because,

01:10:30.210 --> 01:10:34.970
thank you. Yeah, it's amazing. Yeah, that's incredible.

01:10:35.409 --> 01:10:37.909
Or, thank you for the spreadsheet. I was helping

01:10:37.909 --> 01:10:40.210
this other person learn the spreadsheet because

01:10:40.210 --> 01:10:42.550
they needed it, and now they're my husband. Thank

01:10:42.550 --> 01:10:45.859
you very much. Yeah, and you never know. You

01:10:45.859 --> 01:10:48.600
never know. You never know what could happen

01:10:48.600 --> 01:10:51.479
because, yeah, because doing one particular thing

01:10:51.479 --> 01:10:56.479
could then lead to unbelievable impacts and unbelievable

01:10:56.479 --> 01:11:00.020
consequences. For others or for you. Yeah. Yeah.

01:11:00.359 --> 01:11:03.140
I think that's a really good place to wrap up.

01:11:03.399 --> 01:11:06.260
Dan, this was such a pleasure. Same here. I appreciate

01:11:06.260 --> 01:11:08.979
it. I love your questions, the areas where you're

01:11:08.979 --> 01:11:11.680
going after and stuff like that, and knowing

01:11:11.680 --> 01:11:14.779
your background makes it even more so in terms

01:11:14.779 --> 01:11:17.420
of your technical background and stuff. It was

01:11:17.420 --> 01:11:19.619
really, really a pleasure to have you on. Thank

01:11:19.619 --> 01:11:21.479
you for all of the incredible work that you've

01:11:21.479 --> 01:11:24.460
done. Yeah, your TED Talk is inspiring. It's

01:11:24.460 --> 01:11:28.880
inspired me. Your case study was one of those

01:11:28.880 --> 01:11:32.020
first ones. Reading that and, you know, connecting

01:11:32.020 --> 01:11:34.840
with you is a real honor and a real pleasure.

01:11:35.000 --> 01:11:37.319
I really appreciate you taking the time to chat.

01:11:37.359 --> 01:11:40.779
Let me pick in your mind. My pleasure, too. Thank

01:11:40.779 --> 01:11:54.529
you, Dan. the first electronic spreadsheet, and

01:11:54.529 --> 01:11:57.069
the killer app that helped launch the PC revolution.

01:11:57.989 --> 01:12:00.390
We explored what five decades of platform shifts

01:12:00.390 --> 01:12:03.810
teach us about today's AI moment. Dan's framework

01:12:03.810 --> 01:12:06.890
is simple but powerful. Breakthrough innovations

01:12:06.890 --> 01:12:09.930
must be a hundred times better, not incrementally

01:12:09.930 --> 01:12:13.489
better. The same questions he asks about spreadsheets

01:12:13.489 --> 01:12:17.270
and word processors apply to AI today. What is

01:12:17.270 --> 01:12:19.710
this genuinely better at? What does it enable?

01:12:20.140 --> 01:12:23.399
What trade -offs will people accept? Does it

01:12:23.399 --> 01:12:27.279
pay for itself immediately? Most importantly,

01:12:28.000 --> 01:12:30.319
Dan reminded us that we never fully know the

01:12:30.319 --> 01:12:33.000
impact of what we build. Whether it's a mother

01:12:33.000 --> 01:12:35.140
whose daughter with cerebral palsy can finally

01:12:35.140 --> 01:12:38.500
do her own homework, or a couple who met learning

01:12:38.500 --> 01:12:42.600
spreadsheets. The moments worth remembering aren't

01:12:42.600 --> 01:12:45.600
the product launches or exits. They're the unexpected

01:12:45.600 --> 01:12:48.430
times. when your work changes someone's life

01:12:48.430 --> 01:12:52.189
in ways you never imagined. Thank you for listening.

01:12:52.930 --> 01:12:54.670
Be sure to subscribe and share with a friend

01:12:54.670 --> 01:12:58.909
or colleague. Until next time, keep on learning.