(2019-06-19) Kay Immense Challenges

Alan Kay: When “What Will It Take?” Seems Beyond Possible, We Need To Study How Immense Challenges (Grand Challenges) Have Been Successfully Dealt With In The Past. Just as one person can’t make an automobile from ore, but 1000 can, a well organized community of top people can be qualitatively and exponentially more powerful than simple top-down hierarchical tactics/strategies and general voting.

This has been done successfully a number of times in the past--mostly when “normal” feels under great enough desperation to allow very different approaches

We will look at some of these methods using seven historical examples; two of them in some detail.

We’ll finish up with a short survey of barriers to successful efforts, mostly caused by many aspects of “human nature”

Even things that seem to be completely stable are actually systems that are “only somewhat dynamically stable

A good way to think about most systems, no matter how stable they seem to be, is to compare a push on a bottle, and then upside down, and to see that the energy needed to topple can be many times less than the energy needed to restore

Many of the “Immense Challenges” are Intertwined Systems (Intertwingled)

Normal thinking” for us is “remembering and recalling” for short term concerns, often in the form of proverbs, stories, and simple rules and rituals

One of the most recently invented perspectives is whole-systems thinking

The complexities of systems were starting to be realized but were difficult to model and predict until the invention of the computer--itself a complex system-but also a kind of “language machine” that can represent complex systems and move the models over time--especially into the future--to make a new kind of imagination amplifier

In other words, we humans may have enough power to topple our critical systems but may not be able to restore them. (In short: don’t topple them!)

Many of “the systems we live in, and the systems we are” have only partial dynamic stability, so we have to be very careful about nudging them.

Very Small Changes Can Topple

Independent Small Actions That Combine To Topple

Anyone can build a doghouse from almost anything

Almost no one can build one just 100 times larger

Internet

The aim was to connect everyone on our planet with a system that would be “too large to control” and “could not be taken down to fix or grow”. Thus it was a qualitatively differently scaled “immense challenge” than the much smaller and more fragile switched telephone networks.

Societies

Perhaps the ultimate examples of complex systems are humans and human societies (human system)

This is because they carry around their own mental universes that are not always in accord with what is actually going on (mental model)

To deal with scalings we have to move from tinkering to learning engineering, maths, science and systems.

Historically, we moved from the “tinkering instincts” we share with other animals to “more principled making”--Engineering--to powerful “symbolic engineering”--Mathematics--to the deeply powerful ways of thinking that make up Science, and finally to unite all through systems relationships

IQ is not effective without Knowledge, and creating useful Knowledge requires Contexts containing powerful perspectives and world views.

"Context is worth 80 IQ points!” (PointOfView is worth 80 IQ points)

Besides being able to extend our thinking with the qualitatively new methods, one of the greatest benefits of this kind of training is internalizing a much deeper sense of how limited our thinking actually is -this helps promote “anti-fooling” when real thinking needs to be done

It’s important to note that one of the main reasons that the highly unorthodox methods used in the examples were tolerated was that all had a considerable sense of urgency connected with them

The projects we will glean to extract methods and principles are all large to very large group efforts that addressed problems previously thought impractical or impossible, were accomplished surprisingly quickly, involved a wide range of top talents with unfettered choices how to find and solve the problems, and had somewhat random funding support.
 The Empire State Building (‘30s) Radar (WWII) Code-Breaking (WWII) Manhattan Project (WWII) SAGE Air Defense System (‘50s) ARPA computing research (‘60s) Xerox Palo Alto Research Center (PARC) (‘70s)

Engineering: The Empire State Building

has the least amount of added science to its first class engineering

Maths + Engineering: Bletchley Park Code-breaking

This is one of the best examples of “try everything with every kind of talent, no matter how nutty or unlikely”, and despite quite a few attempts by “reasonable management” to limit the scopes of the attempts.

electro-mechanical “bombe” machines (both adapted and newly invented from already existing examples). These worked but were still slow, and led to a number of much more controversial proposals for mostly electronic machines using valves (vacuum tubes) to do computational and logical operations

Flowers ignored the official fetter, and instead almost single-handedly created in less than a year the remarkable Colossus computer using thousands of valves, which worked perfectly to defeat the German Lorenz cipher machine. This amounted to almost inventing and building a digital computer from scratch, and Colossus predated the American ENIAC by several years.

Science + Maths + Engineering: Manhattan Project

Groves saw that while the science part of the project was critical, the overall effort would have to be a massive undertaking to set up at giant scale all the possible ways to refine fissionable material. Over the few years the US was in the war, he spent 1% of the war budget, to

WWII

build entire functioning cities and had over 600,000 people involved in a project that might not work (or get done in time).

Science + Maths + Engineering: Radar

Henry Tizard (UK) and Vannevar Bush (US) were two scientists/politians who had considerable government influence, and were key to the sharing of the magnetron and also atomic research results.

Cold War

Science + Maths + Engineering: Whirlwind and SAGE

“Whirlwind II” in the mid-50s morphed into the SAGE air defense early warning system,

The image of SAGE and the PDP-1 prompted several far-thinkers to imagine “a SAGE graphics terminal in every home” as part of an “information utility” as an analogy to the water and electricity utilities already connected.

ARPA Add-on

Science + Maths + Engineering: ARPA-IPTO (and other DoD funding)

In 1962 “spare funds” at ARPA as the space program shifted to NASA were given to JCR Licklider

This major funding was joined with smaller sources to eventually create an entire community of about 20 large “projects”--about 3/4 were at universities, the rest at think-tanks--devoted to formulating problems from the vision and inventing working results that wound up constituting many of the basic technologies for the computing of today

Science + Maths + Engineering: Xerox Parc

In 1970, when Congress blindly started to curtail ARPA, Xerox Parc was set up to “finish the job” with most of its computer researchers (all young) drawn from the ARPA projects.

With regard to the “immense challenges” we face, part of the solution processes of “new levels of thinking” has to be reflected by inventing “new levels of thinkinghelper-tools” and much better education processes to help more people learn how to use them

Basically: re-engaging with Licklider’s visions and lifting them above the “consumer mire” that has keep them from helping global citizens outside of science and engineering

ARPA + Xerox Parc Research

1. The goodness of the results correlates most strongly with the goodness of the funders

Soon, Lick had funded about a dozen projects, later growing to about 20 at 16 venues, at major universities (Carnegie-Mellon, Stanford, Illinois, etc.) and several government research think tanks (RAND, Mitre, etc.)

2. Visions instead of goals

Whenever he was asked what he was doing, he would only say: “Computers are destined to become interactive intellectual amplifiers for all humans pervasively networked worldwide”.

Whenever he was asked what he was doing, he would only say: “Computers are destined to become interactive intellectual amplifiers for all humans pervasively networked worldwide”. This came to be known as “The ARPA Dream”

3. Cosmic metaphors really help imagination

When asked why he used “Intergalactic” in this early memo, he said “Engineers always give you the minimum. I want an world-wide network, so I asked for an ‘Intergalactic’ one!”

4. Fund people not projects

“We will accomplish this by finding and funding special people who will have their own ideas about how to go about realizing the vision. They will come up with their goals and their processes.”

5. Fund problem-finding, not just problem-solving

The secret of science is to ask the right question

6. No peer review

Peer reviews tend to be “too reasonable” and it is also very difficult to find “real peers

7. It’s baseball, not golf!

we are not going to cry about losing a stroke here and there. We are playing something more like baseball, where successfully getting a hit 30% of the time is considered excellent

8. Fund great people (MacArthur for groups!)

This means that only absolutely top people should be engaged to do the creative work

9. It’s a research community not a research project (Research Lab)

Lick’s grants were similar but were much larger: large enough so that big projects requiring many people could be supported

10. Important results include new great people

In addition, the funding also covered considerable “student and intern development”. The idea was to develop young people into more “great people” who could be principle investigators in the not too distant future

Even though the percentage of the most unusual types might be quite small, a large enough population will produce enough high ability prime contributors to fill out most needs.

11. Separate responsibility from control

propensities for “control” will tempt the managers to try to “command and control” the processes they are supposed to be helping

12. Synergy requires constant messaging

the students and interns also wound up acting as the “messengers” and “cooperators” in the larger community

This larger overview was another reason that Bob Taylor aimed at getting young ARPA researchers for Xerox Parc

13. “No one can have good ideas inside the Beltway”

14. Train your successor and get back to work!

He felt it would be much more productive to continually bring in new directors from the research community.

It is very often the case that a new executive will dismantle projects by the previous executive. Quite the opposite was the case with ARPA. In part because each new director bought into the “ARPA Dream” vision, the scope and perspective of the work was amplified in many ways, and left room for brand new projects.

15. Argue to make progress, not to win

be able to argue deeply about ideas without personal attacks

16. If you have the ability to invent and make new tools that are needed for your problem, then you must.

Much of the computing hardware and software used for computing research was made by the researchers themselves, and later adopted in parts or in whole by existing manufacturers. The computers that led from EDVAC in the late 40s to Whirlwind to TX-2 to Project Genie and NLS to the many different Xerox Parc architectures were all invented and built by the researchers

17. Think and work in the future, not the present or past

it is possible to “buy the future” by using and making supercomputers in the present that can be confidently predicted to be generally affordable in the future.

Parc went very far by making 1000s of “personal supercomputers” (in today’s money, costing about $130K each!), in order to invent the 80s and 90s in the 70s.

18. Take an idea immediately 30 years out to evaluate

if “it would be ridiculous if the idea weren’t possible 30 years out”, only then is it worth thinking about how to do it

19. What Is Actually Needed (WIAN)

If we aim at WIAN and not just “better” then the process to get there might have a learning curve that starts worse than those striving for “better”. But achieving WIAN crosses a qualitative boundary that opens the door to stronger and very different kinds of thinking

Barriers

the question I’ve been most asked about “all this” is: Given that the methods used in these and other examples have worked so astoundingly well, why don’t funders, organizations, governments, universities, etc., set up similar processes to not only deal with our “immense challenges”, but also just to generally make great improvements in many areas?

The “Six Core Biases” in the Ostrich Paradox are:
Myopia in time and environment
Amnesia (quickly forgetting past difficulties)
Optimism (“things will work out”)
Inertia (especially where there is uncertainty)
Simplification (cognitive load, etc.) Herding (basing decisions on societal consensus

these were all pluses for almost all of human history where daily survival in an unkind environment was what needed to be given close attention

Other

Another related cognitive bias is against “other”.

An interesting form of “other” is the otherness of things one didn’t grow up valuing, and especially those who are very good at them. A good example: in the US, at least, almost the only people who are allowed to be publicly exceptional are sports stars

Loss Aversion

Dunning-Kruger

Until I believe it, I can’t see it!”

Back to the issues at hand

How Society Can Get Better At Dealing With “Immense Challenges” That Are Not Yet Like Wars

global warming

Charles Keeling, a chemist turned geologist, in the mid-50s devised the first highly accurate instruments for measuring the CO2 content of the atmosphere. His first measurements were 310 parts per million (ppm) and rising on average year by year. By the early 60s it was scientifically clear that the amount and pace of the rise was dangerous, and the first warnings to the public and the government were given

56 years after the first clear warnings, the general public, their governments, their industries, etc., still cannot summon enough informed imagination to see this as an approaching global disaster on many fronts

In Sum

No one and no committee knows enough to lay out all the direct problem solving actions that are needed to deal with “immense challenges”. But past history of dealing successfully with a number of such challenges shows that processes involving many thousands of people can be organized well enough to yield enormous synergies of effort and produce new ways to understand the challenges and invent and build new ways to handle them.