Actually many ideas were 30-40 years ahead of their time and were
extremely difficult to implement on primitive hardware that existed at mid 60th. Multics began as a research project and was an important influence
on operating system development. It also pioneered the use of high-level
languages for writing operating systems. PL/1 was used as system programming
language, paradoxically it was also at least 20 years ahead of its time and only
simplified dialect was preserved in C -- many key ideas such as exceptions
handing, built-in strings, etc were "reinvented" on C++ and Java many years
after they were introduced in PL/1. It also was a pioneer in computer
security, being essentially an opposite of Unix. Many of security
innovations in Multix find its way in Unix only 30 years later. Here is
how Wikipedia described
the situation that existed in early 60th.
Ken Thompson was one of about 25 members of the Bell Labs Technical Staff who
worked on Multics in 1965-1969. He got many key ideas from the system as
well as the spirit of free development environment with free sharing of ideas
that was created by
Fernando Corbato and that later
became an important factor in Unix success. As Corbato noted in his Turing
lecture:
The system became a commercial product sold by
Honeywell to education, government, and industry. Bull canceled Multics
development after 21 years. Multics continued to be used by customers for an
additional 14 years. As high level language was used as system programming
language and source was made available to all customers Miltix was probably the
first industrial strength OS that was both developed and shipped in spirit of
open source model:
It did not have huge commercial success but it did enjoyed some success (approximately 100 site with
Multics were deployed) due to the
fact that hardware requirements dictated very high price of hardware.
Multics seems to be an under appreciated predecessor of Unix -- Bell Labs
folks were trained in Multics were they became addicted to timesharing
environment and borrowed a lot from this system including the
key idea of the hierarchical filesystem and key ideas for C language as a
simplified PL/1 with BCPL address arithmetic (see also links in
Multics Home
page). For example Thompson in
recent interview stated:
... ... ...
Bob: How did the Multix people feel about Unix?
Dan: Well people always called it, you can look
at the different spellings. It's one of what Multix was many of. And
it was Multix that was missing some features, important things, but
it was the fact that the people who went off and did it were major
people in the development of ideas in Multix.
Whenever I use Unix I just remember how we did it in Multix and
try it and sure enough it will often work. You know it had pipes.
You know ls was the way you got lists. It had drivers and stuff and
I used a higher-level language to program it. So, nowadays we feel
wonderful that that was the 386 architecture that came out of the
Multix architecture. So an awful lot of things.
Bob: I didn't know that.
Dan: Oh yeah, yeah. Now they decided to use it
in the - most people use it in the flat address space, not in the
segmented address space format that the Multix used. We missed that
because in Multix all you did is say, I want a file, open a file and
you got an address pointer to it. And you can just do memory
operations on it and it made input/output really easy. And you could
have any amount of memory because everything started at zero because
it was a new segment. But initially that was too expensive in the
hardware so a lot of people didn't do that and they went with the
flat memory system where they have to simulate as if they had
segments. But the rings of protection in the way and the whole idea,
the kernel in the media and that type of stuff, writing your
operating system in a higher level language and all those things,
using a shell do to stuff, a lot of them were pioneered in many ways
through Multix.
Security being extremely important, Multix was partially funded
by NSA and others in the government. So we were very security
conscious and we were always told stories about that would teach us
why you needed to be more secure. They would tell us about how -
what does it mean to be secure? Well, you know it's not just not
telling but not letting them know. So, for example we were told that
on the day of the Cuban Missile Crisis, so the story goes, the head
of NSA was seen rushing from his car. That was considered a security
breach. Okay? These are the war stories you are told.
Dan: I see. Spies have to saunter.
Bob: Yeah, well no. If they run, they have to
always run.
Dan: Oh, okay.
Bob: It's poker, your poker face. If you are
always nervous, that's fine but if only you can distinguish. People
looked at what type of attacks could you do to figure out what was
going on. Trust - Bob Frankston, my partner during VisiCalc, he did
his master's thesis on how do you in a computer market place, like
Multix was or the Internet, I guess the Arpanet or whatever it was
becoming at the time, and you have all these parties that don't
necessarily trust each other. How do you have micro payments? How do
you have payments? And his thesis was how you need a third party you
can trust and how you can automatically do stuff. It was very
interesting. This was back in the early 70s I think or the 70s he
did it. The mid 70s he did that.
So we were thinking about those types of things. And we had
people you know who's job it was to try and crack systems. The
government was paying them to try and figure out how to crack Multix.
And I sort of taught from one of them a year or two ago. It was
fascinating about what they did and the exploits that they used,
which of course all work today.
But the Multix, you didn't worry about stacked overflow. Because
of its architecture, it could handle that a lot better but its
stacks went down or up or whatever, they were very security
conscious. And even when they started doing the ArpaNet, they wanted
to do things in cryptic but they were told by the government that it
was a no supposedly.
Bob: Really?
Dan: NSA was, you'll have to talk to some people
about that.
Bob: Yeah, I will.
Dan: I can give you some names. So, there were
real war stories.
Bob: Literal war stories.
Dan: But I went from there to working in
newspapers, computerized type newspapers, which was the real world.
Right after that I worked with microprocessors at a company that
made electronic cash registers for fast food places and word
processing before that in between. So, I got a range of regular
people using my products and what's it like. Like they didn't have
modems. They didn't have extra telephone lines for the fast food, so
you'd have to call up the, if you wanted to download what happened
that day, you'd call up a store, which was the pay phone and they
would throw a switch, which would connect it so they could dial up
to the modem so they could download it. Sometimes the person said I
can't talk to you now, we're being robbed. It was interesting
working for companies like that. But actually that was a small
company, a very small company and I found out that it was able to do
all the things that big companies could do. You know it made, wave
soldering made its equipment there. Was using the latest shift sets
and doing all this neat development and it was a little company. And
we were in decked as a billion dollar company in those days, which
was a lot of money.
Bob: Huge.
Dan: Huge in those days and I was there when it
through the billion I think. And this little company could do all
that same stuff. It just inspired me in becoming a little company
person.
Bob: You realized the potential of a little
company could be a big company or to just be a successful little
company?
Dan: That little companies aren't a bad thing to
work for. They can do just as good work as the big company. You
don't have to be a big company to do interesting work, as a techie
or as anybody. And since my dad, my grandfather was a small
businessperson, it just fit with me. But I did go off to Harvard
Business School to learn what business is about because I figured I
didn't know much about business. And what better way to learn than
to go to school. And I treated myself to two years of school there.
Well, I had been making money and saving it as a programmer and all
that and you know to suddenly become a student again, that is sort
of treating yourself. Spending your money to go back to school.
Bob: So, how did you fit in at Harvard Business
School?
Dan: The first evening, we showed up at a
reception and spouses would come up to me and look at me, with the
beard and a ponytail down almost to my waist, and said, "Oh my God,
you're here!" You know because they say I was afraid they would turn
into these capitalistic accomatants here but with somebody like you
- I had the third longest hair in my section of 90 people. Three of
the women had longer hair than me. And I did cut it the summer
between my two years when I was going cross-country, camping
cross-country. I had seen Easy Rider and decided it. And I talked to
people that actually had people with tire irons coming for them,
decided I would go. If I'm ever going to cut my hair, I should cut
it now. I cut it down. But I fit in well. The professors were you
know, they are not the big business whatever type. I mean they are
much more into ethics and stuff than a lot of the students
necessarily were. And I took some courses in entrepreneurship and
stuff and they brought in all these wonderful entrepreneurs of all
different sorts. Sprig who did Advent came in and the head of Analog
Devices and we met all these different people. For me it was great
because I got to see all sorts of business and how it was done, the
case method. You learn about all sorts of things. And it helped
inspire the spreadsheet. So, and I had one of my classmates, who
inspired me in certain ways was a real inventor. He had gone to RISD,
the Rhode Island School of Design and he worked with this guy by the
name of Ron. And Ron has a company called Ronco.
Bob: Oh sure.
Dan: And my friend has gone on to do the
rotisserie. His name is on the package, among other things, spray on
hair. He helped get that out the door. One of my classmates is
governor of New Hampshire right now.
Bob: Ron Popeil.
Dan: Yeah, his partner or whatever, I mean Ron's
the head, was one of our classmates in our section. We had two
Olympic athletes. I mean it's a really interesting, you know people
from England and one of my classmates was a member of OPEC. He was
from a Central American company that did oil and he actually sat
behind Sheikh Yamani, who was at OPEC, like a big guy at OPEC. He
had been, himself was kidnapped by Carlos. I mean it was very
interesting learning about what the world was filled with.
Bob: I was there the day that happened.
Dan: But you weren't kidnapped.
Bob: No I was sitting in OPEC headquarters in
Vienna. I was in the lobby when Carlos and the people came in.
Dan: Oh my God.
Bob: And they said, this is - I didn't know this
was Carlos. And there was this fat Viennese off duty policeman who
was the security guard and he said who are you? And Carlos said,
we're the Palestinian Delegation. And he waved them through.
Dan: Oh my gosh. You get interesting stories
about all sorts of stuff and you learn about all different ways,
about certain people so it was a really great experience to go to
school like that. I graduated Chairman of the Board with no salary
but I had this product that I had announced a few days before
graduation. The NCC in New York, which is also when I first met Bill
Gates and Ben Rosen and a lot of people.
Bob: And that was the spreadsheet?
Dan: That was the spreadsheet. The idea for
VisiCalc came to me sitting there in Harvard Business School. And
taking my word processing background from Digital, my interpreter
and other background from days before MIT and at MIT and stuff and
putting it all together and coming up with the idea of word
processing for numbers. The idea that it was, like words like
desktop publishing became but in those days it was for newspapers.
Layout - you could put things anywhere you want because we used the
paper as a two-dimensional grid where you put things anywhere you
felt like it. Not that it's column and rows that are named and
whatever but it's a general way you put things.
And the original idea was supposed to be a head up display
because I'm daydreaming in class. There are 90 people in a class,
you know and you're listening to professors. So, I imagine my
calculator had a mouse, ball at the bottom. I had actually seen a
mouse before so and you moved it and you had a head up display so
you could look at people. And you could pound out numbers and stuff
like that and I came up with the whole idea you could put things
where you want, just like a blackboard because I was thinking
blackboard and spreadsheet paper. I even found some of my notes
about it were on the back of a publication called The Spreadsheet
that we had, which was one of the rags we had business school.
Bob: Sure, sure.
Dan: But decided to run, to found a company to
publish, to make this thing. It was published by another business
school graduate that I met through the business school. I had
professors advising me. One professor hooked me up with my
publisher, with Dan Fylstra and Peter Jennings and their company and
two MBAs, one from Harvard and one elsewhere. And when we wanted
advice on things, I went to Glauber at one point, who ended up as a
head of Nasdaq, the company that owned Nasdaq, was Assistant
Treasury Director later on, Michael Porter of Competitive Analysis
in Oil gave us advice. We went to _____ Miller. I mean this was
really cool what's available from the business school.
Bob: So had you gone to Cleveland State Business
School, it wouldn't have been the same?
Dan: It would have been different. I would have
ended up with the same product because I had the same background.
The first professor who said it might be a good idea was an
Operations Manager/Production Manager but the first computer person
that I told about it was my professor, Professor Cash, Jim Cash, who
actually had a Masters in Computer Science. And he told me user
interface. I mean he couldn't visualize what I was talking about
because without actually seeing the product, it's hard to visualize
it. But he said you are talking about the input/output, the user
interface. That is the important thing, Dan. You should do this. And
Jim went on to head the MBA program. He's now retired. He's a
director of Microsoft and Migrator and General Electric and stuff
like that. He's head of the Compliance Committee on the Microsoft
Board. Jim's a really good guy.
Bob: There's a job.
Dan: I feel really bad for Jim to have it but if
anybody's going to have it, he's a good guy to have it. Jim's a
really good, he decided to become a business school professor when
he could have gone on to industry. But he had the chance to be the
first tenured black professor at Harvard Business School in Wharton.
And that helped a lot of people of color to get into business
school. So there was good stuff. And he also was the one who helped
spear head getting email and a lot of those other things into the
curriculum and stuff and teaching managers who came in for the
advanced programs to learn to use computers more and stuff.
So I had professors like that but he taught me cost accounting.
Bob: Exactly. Were you further ahead in the use
of computers in business school then industry was in a comparable
level?
Dan: Well let's see. They used a timesharing, a
depth timesharing system and they used it for simulation and that
was about it. We used our calculators. Some people like my friend,
Allen who went to be with Ron with Ronco, Allen had a programmable
calculator. He had more money than I did. He had one of the HP
programmable and he would always program the things in the
calculator and there I would run down and use the basic terminal and
he would always beat me _____. I learned that quick and dirty can
win when we were study partners.
And so we didn't use computers that much at the time. Though I
did do some prototyping for VisiCalc on their machines. The actual
time when I came up with that I'll use the A,B,C/1,2,3 type grid and
that you would only see the results and then you would have a little
window where you would be able to see the formula and stuff, that
was actually that idea came to me while prototyping on one of their
terminals in Basic.
So, but no, they weren't really into using it. Like you didn't do
email. We didn't do any of that stuff. That, Jim and the others
started bringing it in, I think first to the - well when VisiCalc
came in, after me, there were professors, Eric Rosenfeld, who was a
professor there, who was one of the partners in Personal Software,
early Lotus or pre-Lotus with Visiplot, Visitrend and stuff like
that. He bought an Apple and would bring it in and use it. And I
think used it to help mark tests and stuff. The professors actually
started using it and of course they noticed that certain students
and then of course they brought in using a laptop, which was an IBM
portable I think at the time. And they started adding it. But
actually in the development, while I was prototyping, I had the
early versions of it; I brought professors in under nondisclosure to
see it. And one of them was Barbara Jackson who had been my
professor where I had learned about financial forecasting systems at
the school. We actually got to play with some of the systems.
And she was on the calculator committee because remember they had
gone from slide rules. First they had slide rules in the old days,
then they had to go to calculators and redo all the - they had to
make calculations that were easy to do on slide rules and then had
to redo all the cases to be able to have numbers and stuff for
calculators. And so she actually was very helpful. She explained to
me she had done a lot of work with computers with senior executives
and was telling me how my competition was not any of these other
tools. It was the back of an envelope. It was paper/pencil to
calculator. That if it wasn't easier the first time, they won't do
it. And it has to be, she said, easier, easier, easier.
She pushed me to that stuff. My classmate, John Reese, the other
MIT grad in my section, said well Dan, you know, you have it this
way. Can't you remove that because it's obvious that one plus and
then pushing an arrow is unambiguous? That you want to point, put
the thing. You don't have to put a key. You know I said, oh well,
John that's a great idea. We're having a reunion tonight and he will
be there. I get to see him about that. I got people who gave me
ideas along those lines. So they started and eventually they
switched over and then now they use the spreadsheet, which is kinda
cool to affect teaching at your old institution.
That was one of the big things at every institution.
Bob: At every institution.
Dan: But especially at, you know how you like to
have your parents be happy with what you do and your family happy
with what you do. You know having your teachers who you look up to
and who taught you things to say you did good stuff is a nice thing
to have. And to bring back something to help you to add something to
it is even better. Especially in my case, to make it easier for
people who are going through what I went through to go through that
stuff.
Bob: How did Bob Frankston come into the
picture?
Dan: I met Bob at MIT. After the first semester,
I figured MIT is going to be so hard. I'm not going to do any
computer stuff because I'll flunk out if I do that. So my first
semester I took a computer course, which I did really well in, I
just going into the final I was number one in the class I think. It
was pass/fail I think.
Bob: First among equals.
Dan: Do what?
Bob: First among equals.
Dan: Yeah, among equals. Well, no not everybody
was a freshman. Only freshman were pass/fail.
Bob: Oh, okay.
Dan: And then worse yet, the second half of the
year was Kent State happened and the student strike and they made
everybody pass/fail for the last of the semester. But it didn't help
for us freshman because we were pass/fail anyway to cut pressure
down. So we wouldn't have as many suicides. And I decided to look
for a job for programming. So I started walking around the labs and
I poked my head into different offices and I poked my head in
Professor Corbido's office, who was the head of the Multix project.
And he said oh, you want a job. I'll give you a job. We'll give you
a job on the Multix project. We usually do it for credit but we'll
pay you some money. And I joined that project in late December,
January of must be January of 70.
And my first thing to do was to take Bob Frankston's bachelor
thesis project, which was a limited service system to give access to
computers in a controlled environment to students and add some
features to it in order to make fit into Multix better. So I met Bob
through that and those of us who worked in that lab together got to
know each other real well. Bob had a car because he had been working
as a programmer since sometime in the early sixties. And he worked
for WhiteWell, which later became Interactive Data Corporation. And
so he was paid really well. He didn't get into MIT his first try.
Went to Stoneybrook for a year. And then applied again. So he had to
live off campus. So he had a car but he had a computer terminal in
his room you know because he was paid. It was great.
So he would take us out to dinner, those of us who worked in
Multix or out to breakfast, you know because we worked through the
night. So we had a real social thing going on among us computer
nerds. So Bob and I got close. We figured one of these days we'll
make a business together. Both of us have parents that were small
business people. His in electronics and mine in printing. And we
stayed friends since.
Bob: Bob is a great guy.
Dan: Oh he really is. He lives just a few blocks
from here. So, he's off at a conference or something right now. I
talked to him last night for a second to tell him that I got to get
permission from him to use my new product so, you know, before I do
certain things, I want to make sure it's okay with him.
Bob: Because of the name?
Dan: Well because of the name. Before I would
use the name, Visicalc, I would at least ask Bob. I asked a couple
others but Bob particularly before I hit the buy button on the
domain name and send the application in. And also just to tell him
what I was doing. You know because I want to do something like that
I know that he would peripherally be mentioned periodically. So, he
should at least get an okay. Though he's not interested, he's never
really been interested in spreadsheets. He's interested in
communications much more.
Bob: Really?
Dan: Oh yeah. Ubiquitous communications is his
life. He just did VisiCalc as some fun on the side.
Bob: Well he probably made a couple bucks too.
Dan: Not as much as he did doing other things.
He went to Lotus, after Software Arts and stayed there and that did
pretty well for him. And got them into some things, which they
should have followed up on like Lotus Express, which was this email
thing, where they had a tax on every email sent over MCI mail. That
was a product he got started. But Lotus didn't follow up on that
email or that product. Gates just asked me, why did they give up on
that? Don't they breathe? Don't they breathe air, you know? They
gave up. Microsoft doesn't give up. Lotus gave up and then they had
to buy CC Mail later.
Bob did join me at Slate to do at hand a spreadsheet for propend
computing. He did something that was a kin to VBA, I think before
VBA a Basic like language. Like when I first met him, he had written
Basic for Multix. He had a Basic like language for spreadsheet that
had cells built into the Basic language. After Slate, he went on and
went to Microsoft and had a very good five years, four and half
years, whatever. And he commuted from here.
Bob: I remember.
Dan: He spent half the time but when he was
there, he got them into home networking. He said it really had to
just work. That we should be able to just plug a computer in and to
a home network and it should just work. DHCP and whatever and stuff
like that and you can use things like you know, the wired LANS that
work through the phone wires was pushing on that. He was pushing
them to the Internet. I remember him showing me. I have a video of
him somewhere of him showing me Mosaic running the browser or
running on hardware from Microsoft through Microsoft in connection
to the Internet the month that Netscape was incorporated. I think he
was at that meeting. "The Meeting" where Microsoft decided to go
into the Internet. But basically he views his main thing is he
really pushed them for home network and his home was the original
model for doing that stuff and the prototype.
So he did his good for the world there and he did well by it. So,
but that's all about connectivity. Now he's doing a whole lot of
other things that has to do with connectivity. Trying to push to
people to make freedom of speech on the Web. And freedom of speech
being the program can do what they want to communicate to other
programs and not to have people patrol ways in the way.
Bob: Interesting.
Dan: That everything can connect to everything
so that there'd be ubiquitous connectivity of every light bulb. He
said, "why should a light switch have anything to do with the light
bulb other than to tell it to turn on and off?" You should have a
little thing in it sort of almost like you can slap it on the wall
anywhere you want. Why do you have to have the wires come through
the light that goes to the you know? Why does it have to even be in
the same room? Why can't I, if I want to, be able to check the
status of the switch from elsewhere in the world securely?
Bob: Exactly.
Dan: He writes about a lot of that stuff in his
weblog.
Bob: Well we'll have to get to Bob.
Dan: You really should get to Bob.
Bob: We definitely will.
Dan: But you can't today.
Bob: No, he's at a conference. So, how did it
come that Bob hired on for VisiCalc?
Dan: For VisiCalc? I was in business school and
let's see. So, I had sort of come up with what it should be. I had a
prototype of it written in Basic that run on this computer I used
over at Dan Fylstra's house, which was the offices of Personal
Software, one of the largest publishing of software in those days.
Run out of almost a living room or something. And he needed a
programmer to convert the bridge program from I think TRS80 to the
Apple.
So, I said why not Bob. Bob's a consultant who does computer
programming. He knows the 6502 and that stuff. So, Bob went to do
some work for him and he went over there. He got a listing of the
basic program. The way he did it was he brought in a X70 Camera and
took pictures of the screen as he went through it because they
didn't have a printer on that machine. It was very basic then.
So when I came up with the idea and then made a deal with
Personal Software to publish it, Bob said he'll program it. He had
access to some tools on the Multix system that were 6502 assembler
stuff. He had written 6502 stuff for a previous company that he was
a consultant to - ECD Corporation, competitor to Apple. And I had
done a little bit of consulting for ECD too, so Bob said it will
only take a few weeks to write. So why don't he'll write it so I
said sure. You can write. We'll be partners. We've always wanted to
be in business together and we'll use this to fund our computer
business or whatever we do, my software business. So, Bob ended up
doing most of the programming. I did the user interface design,
mainly in the Basic prototype. I played with that. Wrote the spec
and early manual, the reference card.
So we had been trying to go into business, thinking of different
areas before and hadn't been able to come up with any idea that was
worthwhile. So this is sort of a way to start a business. And he
came up with a name, Software Art, which was a good name for the
company and he helped fund it because he had some more money than I
did. I borrowed money from my family and he borrowed some money from
his family and money he had. We borrowed money from a bank and
bought a computer from Prime, which was a follow on to Multix. It
was very much like it. And he wrote tools and I wrote an editor. And
he wrote an assembler and a linker and we did VisiCalc. It took
about six months until it was basically in final data. It took about
six to nine months. It shipped at ten months or so, ten a half.
Bob: And what was their early reaction?
Dan: Some people went gaga over it. I mean Ben
Rosen when he wrote that famous, "The Tail that Wags", the software
The Tale that Wags, personal computer, hardware dog about it.
Hewlett Packard, one of my classmates, ______ Howe worked in the
personal computing division. The one that did the HP 75, I think it
was. They licensed it from us based on that. So there were people
who were interested in it. Apple was not as interested. Well, maybe
they thought they could it themselves. I remember there was politics
there but I didn't get to see a lot of that.
But among non-computer people, see if you're not the actual
audience for a product, you don't understand it. Most people say
computers can do anything. What's special about a spreadsheet?
Programmers look at it and say you wrote a basic program. But only
the people who really needed it, looked at it and only of those,
those that understood the idea of a general purpose tool, the people
that know what a general purpose tool can do and they can visualize
it. That's a very small segment of the population. They can say I
can use it to do this for my area. Other people needed somebody else
to have done it like what they do and look it and you can use it for
that?
That's exactly what I do. Like I showed a person, this guy Joe
Suldnick, who was The Computer Store, he was their financial person
and he started shaking. And he said, that's what I do all week. You
can do it in 15 minutes what I take days every week to do, showing
me all the spreadsheets he did by hand. Those people got real
excited. But I only sold about a thousand units a month for the
first year. Then we came out with the Radio Shack version, Basic
version, Commodore Version 2. Radio Shack started advertising. HP
started advertising their version, which was a recoded basic the
program or something rather in their system. They took our source
code and recoded from it.
And so then it started taking off. And you know it did more and
more and I sold more and more personal software. Later Visicorp,
they told me, there VP of Marketing - Ed Esber at the time, told me
that the more they raise price the more they sold. No matter what
they did they wanted more. That's what happened.
Bob: I'm not sure that cause and effect was
actually there.
Dan: It wasn't cause and effect. It didn't
matter because they had to go out and buy a computer for it and then
you have to buy a good daisy wheel printer for it. It's like five
thousand dollars anyway. It was replacing maybe timesharing or
something that's five thousand dollars a month. So who cared you
know? It was like the two orders of magnitude improvement, two-week
payback. Like desktop publishing, it pays for itself in two weeks.
So what the hell? You know it's heads I win, tails I don't lose
much. One of my classmates is a business school professor now who
wrote a book - he's at Columbia, Anwar Betty and he wrote a book
about new small businesses where he interviewed all these Inc. 100
companies and he said the thing about the small business person
isn't that they take risks other won't take. They don't take dumb
risks. They just are willing to take risks but only good risks. But
they can't investigate things to death like big people do so they'll
say things like heads I win, tails I don't lose much. That's good
enough.
Bob: Exactly.
Dan: Now, so it became from a thousand units a
month at some point it exploded into this huge, eventually sold a
few hundred thousand, maybe 700,000 copies, tying Lotus and shutting
it down, which in those days was a lot. Basically we knew how many
Apples were sold every month because we tracked it pretty well. And
IBM came to us and basically when they shipped the ads showed what
was supposed to VisiCalc, it was a Basic mock up. They didn't trust
us or whatever. We could tell. It was just a character missing. But
we shipped within a week of when the IBM PC came out and sold an
awful lot of copies on there.
So, it did really well.
Bob: Well things went sour. What happened?
Dan: Well we were in an author/publisher
relationship. In those days that was the model some of the
businesses had. It was the model Dan Filestro at Personal Software
came out of the publishing business, who's a co-founder, editor
something like that of Byte Magazine. Carl Helmers was a good friend
of his. And so that model of publishing and even the lawyers I think
were involved in publishing ______ used. So, they started with a
publishing business and it turned out that the software business has
hits. In the hit business like that with the structure, that's not a
good model and they also went for venture capital.
And which needed different returns and it just messed things up
and the relationship between them and a lot of the authors soured.
But we were the biggest author. Kapor, who was one of their authors
at the time and also an employee, just left and they bought him out.
He got a couple million, two million dollars out of that. We were
selling ten times that much. We would have taken ten million bucks
but we couldn't come to an agreement with them.
So things soured with that and we ended up in a lawsuit with
them. They sued us. It was bad stuff. It was really bad. We were
about to sell our company. We were days from selling our company for
cash and we got sued. It was bad. They bet everything on VisiOn,
which was something kind of like the Lisa and then the Macintosh
except it was character based but it was bitmapped. It had all the
disadvantages of bitmap with all the disadvantages of character
base. It was an early and it goaded Bill Gates into doing Windows.
But Bill Gates had a real cash cow. In our case, they sued our cash
cow. So we had a real problem. And they expected I guess to ship a
lot and make money but they didn't ship enough and they went under.
We sold out to Lotus. They saved us from going under. Mitch was a
real hero there for us and bought our assets.
Bob: How was Mitch a hero?
Dan: For me? He kept my company from going
bankrupt.
Bob: Okay.
Dan: And he paid me more than what somebody else
was about to pay me. And he did it real quick and easy. And gave me
no handcuffs. It was great. I mean I ran into him one morning on the
way to Sofcon, a software conference, at the airport and he asked me
how things were going and I said lousy. And he invited me out to sit
next to him on the plane, first class. I was sitting next to another
suitor who was trying to buy our company. I had been flying around.
I went to visit Esborough, Ashton Tate. And Mitch said oh you have
an engineering, scientific product called TK Software. We have an
engineering, scientific product division. Maybe we can do something.
That was Monday. Friday we had a check. And Friday night was the
first Passover Seder. Liberation, you know, of the slates.
And that afternoon we signed the deal.
Bob: So he property value was TK Software.
Dan: That's the way they saw it, of course it
turned out that the VisiCalc copyrights were more valuable to them
in the long run with the look at field cases that came in. But
didn't know that. And TK Software, they resold for what they paid.
It didn't make much difference. PR value was good. They brought
Dave, they brought Bob in which they got Lotus Express out of it and
Lotus Metro out of it. David Reed joined them who had been with us.
He was one of the authors of the End-to-End argument paper or the
basis's of the Internet, as we know it. David led some of the
projects for them and went off later to _________ and other places
and he's now working at the Media Lab and I think it's Hewlett
Packard.
And so they got a whole lot of other good people out of it. It
was good for them. And PR wise it was good. For Ed it was the right
thing to do. Oh and for me it was good because I was part of the
deal. What we worked out was that I had this product that I was
working on which was a demonstration program, when I was at Software
Arts and written in our internal language and Mitch tried using it.
We used it to prototype some of our stuff and of course it had bugs
and kept crashing. So, I said I want to write a real version of this
and see as a program. And they said fine. We'll make a deal. You can
write the program and then just give us a license, a company wide
license, to copy the source or something for our own use. And I said
sure. And Dan Berkman's Demo Program came out of it.
And I made more money off the demo than I did off of VisiCalc.
Bob: Really?
Dan: Yeah. Well VisiCalc basically, I made my
salary, which was a good salary but I frequently took no salary. At
the end I got something from Lotus's money. I got a little bit of
back salary. And we had a building that unfortunately we had put
effort into making it a nice building and when we sold the building,
I made a little bit of money. Paid for my house and stuff, for a
small house. But Demo paid for a bigger house. And I sold that off
to Peter Norton for a little bit and that was better than -
Bob: Now Dan Berkman's Demo Program was the way
of making a demo version of your program before you would write the
real version.
Dan: It would look like, you could paint the
screen to look the way you wanted and then the next screen and then
the next screen. So you could make it look the way wanted and it
would respond to keystrokes. You could program it. And it was such
that a marketing person could use it. People would use it for
training. It was used for sales, for showing off things. And it was
used for prototyping.
And in fact, a whole lot of the programs you see when you see a
character based program, there's a good chance that Demo was used in
the creation of the look with all the boxes around things for those
programs. Thousands and thousands of copies were sold, 75 bucks
apiece list. And it was used for head up display design on fighter
planes, I'm told, cash registers, what was it, at Prodigy they used
it for designing screens. I think it was used by UPS. I mean it was
used. All sorts of people used it to demo. Lotus How was sold to
Lotus by a demo that was done in Demo apparently.
Bob: Really?
Dan: I think that's what Bill Grosse had told
me. All sorts of things were done in Demo. I learned there about
different licenses. The product had a free runtime but it had a
warning notice on it and some people didn't like the warning notice.
So, you can get a runtime that didn't have a warning notice. But you
had to identify me and pay me a thousand bucks. And companies asked
for it. IBM needed to do a demo and they didn't want to have that on
the demo they were showing off and they paid me. I had to get a fax
machine because it was always being used immediately.
So, I found out that with special licensing deals, you can make a
lot more money. In fact it's the difference between having a really
good vacation and a lot of other things one year and that was the
difference because I had these dual licenses that you could get by
with that. It was kind of, learned about that. The day of learning
about Shareware and things like that. Learned about that because I
was on the committee that decided who got the Andrew Flugelman award
when it first came out and we gave one of the first ones to, I
forget. It was this bulletin board system and it was freeware. And
we also gave it another one to the development of Freeware. No
Flugelman did the freeware so it was one of these bulletin boards.
So I learned about that. And of course I was using freeware
stuff, editors and that, the shareware type of stuff. So I was
learning about that type of world, which was in the legal world. I
had between the lawsuits I was involved with and running a business
and all the contracts involved in financing and all that and courses
I took, I had gotten a whole education. I could have gone to law
school many times over with the money that was spent doing those
things by somebody but it's not really a law school education. It's
knowing a lawyer. But it's been helpful in understanding a lot of
the other things, which is helping me in my understanding of open
source.
Bob: And that's where you are now, right?
Dan: Yeah.
Bob: In the open source world?
Dan: I'm in the open source; well I'm in the
business models world among other things. Of which, open source is
an important part of it. So, when I was at Trellix, we were doing
web authoring stuff and we were doing hosting stuff and I found that
a lot of what we were building in-house was based on Linux because
it was cheaper and better for stuff and Corel and we used open
source this and that and we needed to have a certain type of product
and we were able to get an open source version of that product to
build our thing on but we had to separate it appropriately so we
could because of the GPL and I started learning about all of these
things.
I knew Richard Stallman from folk dancing and stuff from years
ago and from you know, MIT. I didn't know him well but I knew. To me
he was that really good folk dancer at the time. He could do the
Balkan dances and I couldn't. But you know I've been learning about
that and when I decided to go off on my own and program because I
wanted to program again, I'd been releasing my things on a different
type of open source-like, eventually GPL licenses. So I had been
experimenting with business models, learning about it. And I did
consulting. And companies want to know what you should do. What you
should open source, what shouldn't you open source. I'm on an
advisory board to the state here, which is now this whole thing
about open document format and open source, open standards, what the
difference between the two and what should we do.
So I had sort of gotten thrust into all of that. I released a
product in open source last year, List Guard and it did pretty well
in its area. It was a very narrow area, about hand RSS feed, when
you don't have a blogging tool creating it for you. And that did
really well and the dynamics of the open source part really they
made a difference. I mean people treated it differently. The
marketing costs were so much lower. Like they were zero basically.
People don't expect the same support. And they are willing to pay
for it. People give you donations if you ask for it. And plus people
now think you are an expert. I mean there are all these things. And
that was one month of programming.
So when I decided to do something follow on and did _______, I
decided to go the same way because I see that the market likes a
product that people seem to say - and it's GPL, it's open source, it
means to them it's kind of like it's organic or it's kosher. Some
people think that something kosher means it's better. No it isn't.
It means it follows certain laws, certain rules. But it turns out
that because it's been carefully watched at some end, because it
costs so much to do it anyway, a lot of things are of higher
quality. And of course, Hebrew National tries to make it look that
way so there is this image that because it's that, it must be
better. Well there is some image about open source products of
certain ilk that is positive and people are more likely to recommend
it because it's that. They are not going to - they feel that, I
guess it's theirs. I don't know why. Maybe it's part of the culture,
it's the right thing.
So you get certain viral stuff that doesn't pass on of your
product. People telling people about it that you wouldn't get
otherwise. That to me, open source pays - some of the business model
is your costs go down.
Bob: But where's the revenue?
Dan: Well it depends on what you need as
revenue. Well where was the revenue for some of my other products
where the company lost money. I raised tens of millions of dollars
of venture capital and the company in the end, didn't turn a profit.
So, that doesn't - what is my goal? If my goal is to run a
consulting business or to establish a reputation in the area where I
do have a product, it's a help. If I just want to do good for the
world and I need it myself, so I'm gonna spend the money anyway, why
not do the little extra to make it a product. I know with open
source, if you have the volume, there is money. That's the way it
is. If you keep your cost structure. You have to build your business
differently. That's a real problem with venture capital. Because
venture capital is set up for big businesses. They have to be at
least spending millions of dollars to fit their model. When you can
start some businesses in the open source world, you don't have to
spend on the marketing and sales people the same way initially. You
don't have to spend on some of the development. You don't have to
spend on those things. So your costs are much lower but they can't
invest in those chunks. So you are in that netherworld where you
either do it as a labor of love or you have some money to spend of
your own or you are just a fast coder and you do something simple
enough for whatever or you luck out.
And we see a variety of those from things in a dorm room and we
have our Face Book and things like that that people are able to
build businesses out of open source and all in today's world pretty
cheaply. So I'm learning about that and this is to be able to speak
about it, just like anything else, I like to actually have gotten my
hands dirty in the area. And this is a way to get my hands dirty in
the area. Hopefully people will like the product or what it becomes.
I mean it's only an alpha right now. But I assume more people will
try it then try List Guard. So, instead of a few thousand users -
Bob: There is a broader appeal.
Dan: There's a broader appeal. I'm using it to
create websites right now. I mean it creates its own website about
it because it can use it to create Wiki like design. It can create
blog-type looks. You don't have to know html, just like you do with
the Wiki. It handles the multiple editing a little differently where
I can be on my PC and it sucks down the source but then leaves a
little bread crumb up there so somebody else knows not to suck down
the source. It takes care of that automatically. And then when I
post it, then it goes away so somebody else can. When I do the
server version of it, which is the same code with a little different
front end, it'll do similar type stuff and you won't need a client
site. It's always through a browser. So, it looks like it might
bring in, you know that together with consulting -
Bob: What does the future hold?
Dan: For?
Bob: You, us, technology, the world?
Dan: Well, we know accept technology as being
part of the world. I mean the fact that the cell phone is now the
thing that everybody has and people don't even - they carry that
instead of their wallet. And the first new thing since the watch and
wallet, that everybody carries with them. Technology, the Ipod, the
success of the Ipod - we're seeing that it's become part of culture.
I do a lot of reading of stuff about sociology and stuff. Things
like the Death and Life of Great American Cities, you know that
great book and stuff like that about what real people do and how
communities interact and how the Web came about and we're now seeing
technology is so much more part of life. We have to develop things
along those ways.
I wrote an essay called Software in the Last 200 Years. As I
learned about it's part of our infrastructure of society. And we
can't keep redoing everything. We have to build things that sort of
run till who knows when somebody is going to replace it. And things
have to interact and be able to morph and move and be resilient and
non-brittle. They have to be non-brittle when things happen.
Society, you got to look at terrorists and other things. We have to
be able to deal with accidents and you know, at natural situations,
like Katrina and stuff. We look at what worked at Katrina, what
worked in 9/11, what didn't. The president on 9/11 ended up having
to use a cell phone because the secure system didn't at all work.
The vice president found out about it from his assistant who said
come watch it on CNN. You know and that's where they were finding
out. The planes were gotten down by adhoc people figuring out on
their own. You know, Blackberries work. Nextel worked. Some of those
things, the internet continued to work. With other things bogged
down, point to point stuff - they weren't able to rewire New York
that easily. But Voice over IP, look what happened with Katrina. The
only way you could get telephone calls to city hall in Katrina was
over Vonage, Voice over IP, by somebody who by hook or crook was
able to keep their ISP up, we can listen to podcasts on today. And
we were listening in almost real time how they were keeping that ISP
up, little realizing that the telephone call.
So we are learning about this is now a part of it. We have to
build things that are in different way than we used to build them.
Microsoft's change that they are going through with Ray, who worked
for me at one point way back when - great guy, it's going to be
interesting. They are making some changes. I think they have to
change more than they are saying about what the world is and what
the world wants but I think that this technology that I've always
loved and many of us geeky types, nerdy types from back then, people
you know didn't know what it was we were doing. They would laugh
about computers have now become just part of the fabric of society
and we always knew it was there and it's now happening.
I remember for me the big "aha" moment was when I picked up the
Wall Street Journal, I think and it had an editorial that mentioned
VisiCalc, as if you knew what it was. And that just got it for me.
And in comics you had to know what a spreadsheet was. And then when
you look at kids' textbooks today, you know and they teach
spreadsheets. People said did they have spreadsheets when you were
in school and I say no. Remember when I - oh yeah. So, that's kind
of cool. And all of these people that I knew when I was younger,
have helped build a lot of this stuff. We were all building and now
the youngsters today, the people who are the age that I was when I
did VisiCalc, are coming up with things that are changing the world
today. It's unbelievable.
So, we don't know where it's going to go. I mean we had no idea
what computing was going to be like back - we did. We did. Bob doing
in the seventies about micro cameras on the ArpaNet and about how to
trust third party and the next thing we know, we have a Pay Pal or
something like that.
Bob: Next thing you know, 25 years later.
Dan: Twenty-five years later but you know, I
remember first seeing the mouse some many years ago and then in 84,
finally, many years later, it got advertisements about it on the
Super bowl and then you know maybe ten years later, it was a normal
thing and people could laugh if somebody talked into a mouse on a
movie. We're moving a little faster than that now but ten years, 20
years is not a long time in society because you know some of those
people, I mean Andy Hertzfeld and stuff, and all of these people you
have, ____ Olsen and Winers and all these people have invented the
things that make our society, improving on our lives, all of our
lives and our social lives even. And we need to be able to continue
doing that.
Bob: Thank you.
Dan: Yeah you are running out of bandwidth here.
Bandwidth will hopefully get cheap.
Bob: I hope so. Thank you, Dan.