10 Questions On Innovation to Alan Kay
April 2006 – Alan Kay is one of the earliest pioneers of object-oriented programming, personal computing, and graphical user interfaces. He invented or co-invented object-oriented programming; Smalltalk; the 1968 FLEX Machine, a desktop computer with graphical user interface and object-oriented operating system; the Dynabook, a laptop computer for children; Alto, the first networked PC; and participated in the design of the ARPAnet.
1. Who are your favorite innovators?
Hundreds if not thousands of important innovators have flourished over the 400 years that we’ve had real science, and antiquity shows that even though it was much harder to innovate in the deep past, there have been deep seminal thinkers operating in most generations. My interest in reading all sorts of things makes it difficult to set down a short list here. (Who doesn’t love Archimedes, Eratosthenes, Aristarchus, etc.?)
In our own time and my own field, I was particularly struck by the ideas of Licklider (man-computer symbiosis), Ivan Sutherland (invention of interactive computer graphics and object oriented modeling), Dave Evans (the father of continuous tone 3D graphics), Nygaard and Dahl (Simula), Bob Barton (the B5000 computer), John McCarthy (Lisp), Seymour Papert (Logo and constructivist education), Marvin Minsky, Doug Engelbart, Bob Taylor, etc.
In my own generation I’ve been tremendously impressed by the work of Butler Lampson, Chuck Thacker, Dan Ingalls, Adele Goldberg, Vint Cerf, Bob Kahn, Nicholas Negroponte, Carl Hewitt, Paul MacCready, and many more from the 60s and 70s communities funded by ARPA and Xerox PARC.
In the generation(s) after me, I have been very impressed with Dave Reed’s distributed object system, Bill Atkinson’s Hypercard, Mitchel Resnick’s starLOGO, Brewster Kahle’s file systems, Doug Lenat’s CYC research, Takeo Igarashi’s UI work, etc.
The Draper Prize of the NAE is aimed not just at engineering feats but at those that really got out there and benefited humanity:
integrated circuit, jet engine, GPS, Internet, Personal Computing, etc.
One of my favorites from the early 19th century is the guy who invented the hydraulic ram pump, a truly elegant simple outstanding machine.
2. What do you consider are the most promising innovations of the last 3 years?
I can’t think of any in computing (we are in a very incremental and almost moribund period in our field) but this could simply be “old-fogeyism” on my part. If the content of the $100 Laptop is done well and we can figure out how to help the helpers of the children who will get them, then this should qualify as a grand innovation (perhaps in 2008?).
But if we widen the scope, there are some very promising ideas that grapple with some of the largest human problems.
a. For example, about 70% of the world does not have drinking water of high enough quality. The conversion of low quality water to high quality water requires both new technologies (e.g. membranes for reverse osmosis) and energy (for forcing water backwards through the osmotic membranes, or distillation, etc.). Some very good things are now being done here, including a very inexpensive and efficient tidal-powered high quality water producer, etc.
b. I forget when the modifications to bacteria to produce fuel from garbage were done. The insulin modification was done longer ago, but I think I recall seeing that some good fuel production recombinant modifications were done more recently. I think we could claim that the progress in this general area are such to be a promising innovation.
c. Another innovation that has gotten particularly important over the last 3 years is the joining of a number of fields into a real brain science, that has led to a much better understanding of brain chemistry, process, psychology, and pharmacology.
3. What helped you to become a successful innovator?
No one has benefited more from their community than I have (see: “The Power of the Context” – PDF). Beyond that, my interest in helping children learn how to think (catalyzed by Papert and my dismay with adult behavior) and wide reading got me hooked into the deep implications of computing as the next 500 year “big deal” since the printing press. This made it much easier to not get sucked into the whirlpool of current technical problems and to focus on “what should the technology be doing regardless of what it is doing now?”. Also, “artists” are people who have to do art regardless of all else, and fundamental research and engineering is an art form whose best practitioners are kind of compulsive about making things happen. Xerox PARC was a concentration of technological artists with enlightened funding and organization provided by Bob Taylor. It was the perfect atmosphere for people with artistic visions.
4. Did you pay a price to be an innovator? Which one?
Interesting question. Beyond some of the compulsive behaviors (which can get a little too total), I would say “no” with regard to the 60s and most of the 70s. In the last 25 years there have been far fewer funding sources that have the enlightened model of ARPA and Xerox PARC. I don’t have any interest in “computing as a job” but only in trying to round off the first phase of the invention of personal computing, especially forchildren. This hasn’t really happened yet, and it is disturbing to see the poor subset of personal computing that is vended and generally accepted today.
So the price today is much higher than when mainframe people were calling us crazy for trying to do personal computing. That was nothing. What’s harder today is that people think they’ve got personal computing and they don’t really. The analogy here is between the jump from non-literacy to the printing press (which might be fought but the chasm that is to be crossed is clear) vs. the much trickier one of trying to explain to a culture that has the printing press and universities that it is not very well educated and thinks pretty poorly (for example, the US is in this general state — it has the trappings but not the perspective to see how poorly educated most “educated people” actually are).
5. What are the rewards to be an innovator?
I think this differs with the personalities of the innovators. I felt most rewarded when an idea really worked out as imagined (this is a biggie if you have to leave present realities to do the imagining), and also it is a big deal to me that I was funded by ARPA and Xerox PARC (in fact I got better funding as a post-doc in constant dollars and environments than I do today). That is quite a statement to make, but it’s quite true.
A terrific byproduct is that I’ve been able to meet with many of my heros, and work with quite a few of them. That has been very exciting.
My main feeling about the much later awards and medals is that (in our case at least) they miss the actual community-synergistic process. I think this is generally true about work that has a lot of engineering and building of stuff as part of it. As many important things come out of the building as in the original visions, yet the awards tend to celebrate the visions more than the larger process.
So actually getting to do the work with others of like mind has been the big thrill and reward.
6. What are in your opinion the top 3 criteria for successful innovation?
A lot depends on the time scale of requisite change. I’ve been more of an inventor than an innovator, in that I think innovation also has something with getting stuff into the larger world. Most of my ideas that have gone out into the world have been taken there by others.
Example of short time scal
e: the Google folks did not invent the Internet or search engines. They had a particular idea about what was important and concentrated on that, hung in there, and have been successful. I think most innovation is like that: limited invention and ability to couple to existing structures. The ideas that require a general learning curve are more difficult to innovate.
For example, if you look at the grand dreams of the ARPA community, most required large changes both with technology and society. At PARC we decided (rightly) that we would have to build all of our HW and all of our SW to avoid artificial blocks from current practice. The first order theory is: this is crazy! But the second order theory says: if you can do it, then you must do it! A few years before, when we were all in ARPA research (and one of the joint projects was to do the ARPAnet), the IMP “routers” were made to order in one of the projects because no vendor equipment would do.
One way to answer the question is to note that the personality types and motivations of the inventors and innovators I’ve known have been rather different. A lot of different types can be successful. However, the ability to compulsively focus on a goal (short or long term) seems to be quite common across types. The more artistic and grand the goals, the more deep self criticism (without degenerating into the immobilization of depression) is required. This is not easy, because it has to be combined with an almost boundless confidence (but one that is not stupid about current goodness of idea). Since most ideas (even by talented people who have lots of ideas)are mediocre down to bad, the combination of criticism and optimism is a tricky important dance.
You’ve got to have lots of ideas, you’ve got to get rid of most of them, you’ve got to think that the visions are doable.
7. What would you recommend to young people who wish to pursue innovation?
Again, the best strategies depend on personality and motivational types. My type is “lone wolf” but I realized in grad school that what I wanted to do could not be done by myself. This led to quite a bit of conflict before I was able to start to adjust to the idea that I would have to set up a research group of folks whose talents were complementary to mine and somehow try to guide it. I wasn’t terribly good at this, but we managed to get some things done.
So the simple advice here is to
(a) try to understand both one’s strengths and weaknesses. The best plans require strategies for both.
(b) It’s pretty hard to get by without gaining a fair amount of knowledge. Some innovators concentrate in their field (or a sub-field), and others draw on many fields for inspirations and analogies. Many technologists today are quite ignorant of history (even in their own field), and this leads to some really horrible blind attempts to reinvent. Awareness of history and anthropology in the large can relate technologists to the larger human condition and the actual end-users of their ideas. The needs that can be gleaned from history, etc., can also be a great source for ideas.
(c) Avoid taking your identity from your “product” or “process” or “possessions” or “peers” — instead try to identify with your “potential”. I.e. don’t become a “brand”. This has always been a bit of a problem in computing, but seems to be much bigger today (and getting still bigger) for individuals, groups, companies, and countries. Potential allows looking ahead and changing in a way that the other “p’s” don’t.
(d) As Jim Watson likes to say: if you are the smartest person in the room then you are in the wrong room. The genius of Bob Taylor at PARC was to get very good people in a designed environment so that no one was the smartest person in the room. This was very powerful.
(e) Learn how to hang in there.
8. In your opinion how can we create a culture that supports and sustains innovation?
Well, we had one until a few decades ago (and if you put more weight on incrementalism, we still have one). Awareness, scope, and multiple perspectives are all things that real education is supposed to magnify. The lack of real education (and really educated people) especially in the US is a disaster on many fronts. Once a culture gets to the point where it can’t tell whether it is educated or not, it is in real trouble. One of the main goals of education early on is to give a sense of what there is to be known, and the thresholds required. One way to look at the US is that a little over 1% of US adults are scientists, engineers, mathematicians and physicians. That plus the industrial revolution to magnify the results of this small group gives the US the illusion of being a modern civilization. But if you take the industrial revolution away, it starts to look more like a 3rd world country.
A simpler way of looking at this is that any culture that requires war and threat of war to spend on research is missing the self-awareness to climb into the next stage of civilization.
9. What do you think stops/slows down innovation?
People who would rather be in control of mediocre to moribund processes rather than feel out of control with possibly great processes. This has been the big change in funding over the last 30 years.
10.Do you think becoming an innovator can be taught?
Well, with reference to my debts to my own community, I would say “yes, up to a certain extent”. To return to my soapbox on education, real education requires the following to happen to a mind: to create “a new field/idea/technology/path (either radically or incrementally) which had a profound (positive) impact on a particular subject area/field” (which is pretty much your definition of innovation). In other words, we have to do invention/innovation in our own heads to get educated and to learn how to think, and this is not easy for anyone, and it is quite difficult for many, but doable.
Another way to look at these questions from a wider point of view is that the ultimate aim of education is really to learn how minds work, especially human minds, and most especially our mind.