warming up the crystal ball

Regular flight in heavier than air machine, much of medical science and making use of the strong interaction; but certainly not the Internet, television or the car...

 

A few of us were sitting around a lunch table at Bell Labs in Murray Hill, New Jersey in the early 1990s playing the game where you can transport someone 100 years ahead in time and asking what would surprise them.  For fun we were trying to come up with notions that would really surprise the anyone.

 

We came to the conclusion that communications networks wouldn't be a complete surprise as telephone networks were expanding rapidly.  Science fiction in the early 1890s was full of references to forms of communication that look like radio and communication. Novel as the Internet might seem, it wouldn't be a complete surprise.

 

We debated computers.  The ways computers are used would come as a shock - in fact modern use would come as a shock to most as late as the 1940s before people like Vannevar Bush began to connect the dots of what was happening in research facilities.¹  But the work of Charles Babbage on the difference engine and Ada Lovelace on programming came a full century before Bush's insight.  Digital computation and its use may have well been a shock, but computational engines not.

 

We were lucky enough to grow up in an environment where there was always much encouragement to children to pursue intellectual interests; to investigate what ever aroused curiosity. - Orville Wright

 

Today marks the 109th anniversary of controlled heavier than air flight.  By the late 1800s people were beginning to advance past semi-controlled flight in gliders, but with no real success. A real stagnation in design combined with desire and over-funding to kill several would-be innovators and strengthen the notion that even with the advances of the Industrial Revolution, flight was only for the birds and lighter than air machines.

 

An airplane, or a bird for that matter, has to properly balance four forces to fly - lift, drag, gravity and thrust.2  It must also have a viable control system.  

 

The early aircraft designs were ok on gravity, but lift, thrust and drag were all problematic and viable control systems were nonexistent.  

 

A huge problem that required advances on multiple fronts.

 

The best way to approach such problems is to address each issue empirically scientifically.  The Wright brothers spent years learning how to build and fly kites, understand aerodynamics and simple fluid mechanics to built good-enough airfoils, to lower the drag on their machines,  to use a good-enough engine, and to build a control system and learn to fly with it. 

 

The investigations took several years and a careful methodical approach.  They built their own airfoils and had the revelation that a propeller could be made much more efficient using a proper airfoil.  Air-cooled gasoline engines could have a very high power to weight ratio for the day. Wing warping could control roll and elevators and rudders, pitch and yaw.

 

By early December 1903 they had enough of the pieces to make the first real flight.  The pieces were far from optimal and several of them had been discovered before by others - but no one else had put together all of the pieces.³ What makes this interesting was their funding was modest.  Some competitors had enormous funding, but were using designs that were simple extrapolations of earlier non-functioning designs.  

 

One can go on and on about the brilliance of the approach used by the Wrights, but it also needs to be underscored that their design, while good enough, was very primitive.  They patented it and proceeded to prove they were as bad at business as they were at fundamental revolution.  A problem was their patents and business model were very restrictive and others, once the fundamental problem had been cracked, made new discoveries that made dramatic improvements.⁴

 

Last year I was part of a panel charged with making projections for technologies five, ten and twenty years out.  While such events are fun (even more fun when someone is paying and the others in the group are smart), they rarely are useful as predicting the future is a hazard at best.  But at the same time it is possible to get a lot of useful information from a well run process.

 

When I look back on some of the things I have done it is possible to say I saw and even invented bits and pieces of the future.  But saying that turns out to be a combination of selective filtering and hubris.  In reality many of us saw and created many pathways to potential futures.  That was the job of pure research at Bell Labs where the rubber met the sky.  Of course innovation is the useful application of these ideas and making the rubber meet the road is much more difficult.⁵

 

I think it is fair to say almost all of the technologies we'll be using in 2022 are up and running in pure and applied research labs around the world.  The combinatorics of factors that come to bear on which will survive are an extremely difficult problem to solve and there is truth to Alan Kay's famous quote - "Look, the best way to predict the future is to invent it..."

 

Many of the discoveries are impacted by something outside of what is considered to be their normal domain.  Television was originally seen as a point to point communication, portable atomic clocks combining with several technologies to create a new class of navigation - the list goes on and on.  Ideally you can gather a group of people with an understanding of the technology being studied, wide ranging interests, delight in connecting the dots and an inquisitive nature.  From that a variety of useful scenarios can be invented that can form a guide for those who must plan for the future.

 

Some are expert at understanding the human side of technology along with the rate of sophistication of component technologies and can create new products and even product categories - Steve Jobs and his team being an example on several occasions. 

 

Rarely something is created where its basic design survives for a long time.  The invention of the safety bicycle is a great example.   In the early and mid 1880s many inventors tried to come up with a light, easy to ride and safe bicycle.  There were multiple failures until John Kemp Starley of Coventry, England put the dots together and came up with the Rover in about 1888.  He had come up with a design and geometry that is basically unchanged to this day... a design that happens to be the most efficient way for a single human to move around on the ground ...  a design that is nearly four times as efficient as walking.

 

On this day it is sweet to note the connection between bicycles and early flight.  The skills required to build reliable safety bikes proved to be one of the dots the Wrights used in their arsenal of ideas.

 

I should probably quit now as I'm nearly out of time, but it is also very important to note the adoption rates of new technologies.  This the percentage of the population using a technology from in years from the first product - a very different number from the first invention!

 

The first chart is from a now defunct visualization site.  The point that is problem important is the ten percent mark as well as the slope from there through fifty percent adoption.  Some technologies compete with seriously mature technologies and require enough further evolution so as to have very gentle slopes, even though they eventually come to become important.  Some forms of alternate energy power production and electric cars are good examples.6

 

The second chart involves consumer "gadget" technologies compiled by Alexis Madrigal - areas that are generally expected to explode rapidly and quickly generate enormous competition. Note the leading examples are all for media consumption rather than communication or creation. There are a variety of interesting reasons why this is so, but there is not enough time to probe that here.  Gadget spaces where first moves often are not those making money as the technologies mature.  Ideally companies in this space have a deep understanding of how and why people use these things.  If they can add this information to the other dots they're connecting they can create more serious contenders.  Apple is a prime example as were Polaroid and Sony back in the day.

 

So while an accurate prediction of winners a decade out is nearly impossible, it is easier to eliminate losers and the process, properly done, can offer a good deal of illumination into just how you think about the future.  On the other hand surrounding yourself with narrow subject matter experts frequently leads to noisy blindsidings. ⁷

 

__________

¹ Vannevar Bush was an amazing mind - inventor, scientist, engineer and brilliant administrator.  A real renaissance person.  He was one of the people who started the Manhattan Project, ran NACA, CMU, the MIT school of engineering, the NDRC, the OSRD and was a core force behind the creation of the National Science Foundation. His article As We May Think article that appeared in The Atlantic on information indexing and processing was a brilliant piece you should read if you are unfamiliar with it.

² Gliding only requires three - thrust isn't used.

³ This is a critical point. Some of the tricks had been uncovered at places like MIT and Stevens, but were only known to ship designers.  Very little of this was known in the aviation community.  Orville and Wilbur, with their experiments, created a problem relevant body of information that was revolutionary in their field.  

⁴ Ailerons are a prime example.  There was considerable prior art, but the other pieces of powered flight had not come together.  They proved to be superior to wing warping, but accomplished the same basic task - roll control.  The Wrights fought this tooth and nail in the courts.  They won in the US and demanded very heavy fees fundamentally crippling aircraft development in the US for about a decade as the Europeans quickly caught up and moved forward.

⁵ Bell Labs had a good mechanism for that -  linkages between pure and applied research and applied research and manufacturing.  The problem was the last link became weak through a series of consent decrees and the value of the place was diluted.

⁶ These are rich connect-the-dot regions and I spend some time worrying about them.  Get in touch if you have an interest

⁷ There are a few individual subject matter experts who have enough outside areas of interest and their own forest of dots to connect that they can not only be useful in these investigations, but often central.

 

__________

Recipe Corner

 

With the holidays nearly on us I've been experimenting with gingerbreads.  The one I offer here is a recipe in progress - I'm on the third version, but this one appears to be a keeper.  Unlike many gingerbreads it isn't great warm, but shines after chilled in the refrigerator for at least eight hours.  Sukie thinks it is even better with eggnog poured over it or with peppermint ice cream on the side (yes - I've made a batch of peppermint ice cream).

 

There is an issue of what to do with the holiday experiments as we can't eat all of them ourselves.  The trick is a group of always happy-to-try-it-out neighbors. 

 

This one makes a lot.  A nine inch round pan is marginal, a nine or ten inch square pan probably works and making it in a two piece angel food cake pan may be best.  Be sure and put a sheet pan under the cake pan!

 

I've begun to measure cake temperature with a thermocouple.  The cake is done went the core temperature is about 88° C - don't let it get as high as 90° for any prolonged period.

 

Ginger-Ginger Refrigerator Cake

 

Ingredients

 

Cake 

° 85g (3 oz) peeled an sliced fresh ginger

° 200g (1 cup) white cane sugar

° 3/4 tsp fine un-iodized sea salt

° 1/2 tsp ground ginger

° 1 tsp ground cinnamon

° 1/4 tsp ground mace

° 225g (1 cup) water

° 2 tsp baking soda

° 340g (1 cup) dark molasses

° 2 large eggs

° 225g (2 sticks) melted unsalted butter

° 275g (about 2-1/2 cups) all purpose flour

Glaze:

° 125g (1 cup) powered sugar

° a pinch of fine salt

° a splash of milk

° a bit of vanilla extract

 

Technique

 

Cake

° Preheat the oven to 350°F and butter or pam a cake pan

° Put the fresh ginger, sugar, salt, ground ginger, cinnamon and mace into a food processor and pulse until the ginger is ground into the sugar

° Microwave the water for about a minute, stir in the basking soda until dissolved and then the molasses until well blended. Set it aside

° Add the eggs to the food processor one at a time pulsing it a bit to blend them in.  

° Turn on the food processor and stream in the melted butter.

° Stop the food processor, add the flour and then pulse until just blended - don't overdo it!

° Turn on the food processor and stream in the water and molasses stopping when it is just absorbed.  Again don't overdo it.  

° Pour the batter into the cake pan and bake until a toothpick comes clean - about 50 minutes to an hour for me depending on the pan.

Glaze

° Pour the sugar into a big enough bowl, add the salt and whisk in some milk until it is on the verge of being a liquid.  Add a bit of vanilla and whisk to blend.

° Drizzle the glaze over the top of the cake and put it in a refrigerator for at least a few hours.

 

 

Comments

[Jean Russell]

Thank you for this Steve. I always enjoy your posts on exploration and innovation, and I particularly like your sharing of your experience with future visioning here.