A Niche in the Library of Babel

Click for site

There are lots of reasons to love the story of the Wright brothers.  They came from nowhere to solve the great problem of flight, one that had defeated so many others.  They showed straight-up physical courage when flying these dangerous machines, and would do it wearing suits and ties.  They built a secret base in the wilderness of Kitty Hawk for their amazing experiments.   They’re like Jules Verne characters come to life, but without the whole conquer-the-world thing.

Yet one thing that David McCullough’s new biography makes clear is that they succeeded when so many others failed because they had a better approach to invention: systematic and incremental development instead of the flash of genius.

The flash story is how invention is almost always described.   An apple falls on Newton’s head and he wonders if the Moon is falling around the Earth.  Edison thinks “What if you put so much current in a wire that it glowed white-hot, but kept it from burning up by putting a glass envelope around it?”  Tim Berners-Lee gets annoyed that all these connected computers have all these incompatible file formats, and adds a few text markers to them so that a single browser program can display them all.

It’s a good way to tell the story to children, because it makes invention seem like much less work.   Come up with the good idea, patent it, and fame and wealth are yours!

It’s nonsense, of course.  To start with, creative people have ideas all the time.    David E. H. Jones, author of the wonderful old Daedalus columns in Nature and New Scientist, describes the process in his book “The Aha! Moment: A Scientist’s Take On Creativity”.   He calls it the Random Idea Generator, or RIG,  a subconscious process that is constantly throwing up combinations of things.  The trick is not to get ideas; it’s to winnow them.  You have to filter the ideas down to get the ones that are useful, feasible, and doable by you.   Useful means something that a few people want a whole lot, or a lot of people want a little.  Feasible means that it can be done in a reasonable time frame, like a year for a nice idea, and a few years for a great one.   Doable by you means that you have some means of actually making it work.  Patenting ideas that you can’t implement is trolling.

So in the case of the Wrights, it was NOT that they had one great idea that made flight possible.  They had a whole set of them, each related to the problem they faced at the time.  They started by writing to the Smithsonian in 1899 to get any literature on the then-current state of aeronautics.   They then carefully studied what Lilienthal and Chanute had discovered.  They realized that control was as important as mere lift, and studied the flight of birds to come up with their famous wing-warping method.  They knew that Lilienthal himself had been killed while flying a glider, so they had to do unmanned trials first.   Then they needed an open place to practice, one with steady winds, so they wrote to the US Weather Bureau to find the best such place, and so came to Kitty Hawk.  It had long stretches of soft sand to crash into.  Better still, it was away from prying eyes – their only neighbors were a Coast Guard station a couple of miles away.  They needed a place to stay while there, so they built their own cabin and workshop.  They practiced with gliders, and found that the existing equations and tables for lift were wrong.  They invented the wind tunnel to measure it themselves.  They then needed a light, powerful engine, so they had a colleague cast and machine one out of an aluminum block.   The plane then needed a push to get it going, so they devised a rope catapult driven by a falling weight.

This then gets them to one of the most iconic pictures in American history, the one McCullough uses for his book cover:

First powered flight in history.  Orville at controls, Wilbur running along beside.

There they are, cooperating and focused on something ingenious and wonderful.   This is how Americans like to see themselves!   It’s the animating story of every startup.

Once they flew for the first time in December 1903, they gave up on Kitty Hawk.  It was too remote, the biting flies were hideous, and they had almost been swept away in a hurricane.  They returned home to Dayton Ohio and practiced in a nearby field, constantly improving their machine, until the 1905 Flyer could stay up for up to 40 minutes.

Having solved the main technical problems, they then turned to sales.  They pitched it to the US Army, who wasn’t all that interested.   The US military didn’t get concerned about air power until much later, December 7th, 1941, to be exact.   The Wrights were already getting feelers from European officers, though, so they packed up a machine and went to France in 1908.   It was a sensation.   Thousands turned out to see every flight, including royalty.   The Euros had been trying for flight for the previous few years, and were abashed that these outsiders had gotten it first, but also tremendously excited.

In November 1908 the Aero-Club de France threw a huge banquet in their honor and awarded them gold medals and a $1000 prize.   At the banquet Wilbur gave a particularly gracious and eloquent speech.   Let me copy it here just to show that he was not the stern, taciturn mechanic that he’s usually portrayed as:

For myself and my brother I thank you for the honor you are doing us and for the cordial reception you have tendered us this evening.

If I had been born in your beautiful country and had grown up among you, I could not have expected a warmer welcome than has been given me.  When we did not know each other, we had no confidence in each other; today, when we are acquainted, it is otherwise: we believe each other, and we are friends.  I thank you for this.

In the enthusiasm being shown around me, I see not merely an outburst intended to glorify a person, but a tribute to an idea that has always impassioned mankind.   I sometimes think that the desire to fly after the fashion of birds is an ideal handed down to us by our ancestors who, in their grueling travels across trackless lands in prehistoric times, looked enviously on the birds soaring freely through space, at full speed, above all obstacles, on the infinite highway of the air.

Scarcely ten years ago, all hope of flying had almost been abandoned: even the most convinced had become doubtful, and I confess that, in 1901, I said to my brother Orville that men would not fly for fifty years.  Two years later, we ourselves were making flights.  This demonstration of my inability as a prophet gave me such a shock that I have ever since distrusted myself and have refrained from all prediction – as my friends of the press, especially, well know.   But it is not really necessary to look too far into the future; we see enough already to be certain that it will be magnificent.  Only let us hurry and open the roads.

Once again, I thank you with all my heart, and in thanking you I should like it understood that I am thanking all of France.

That got a standing ovation!   And he had to autograph over two hundred menus.

But Orville didn’t see this triumph.  Two months earlier he had been in a terrible crash while showing the plane to the Army in Fort Myers, Virginia.  A propeller had cracked, which caused so much vibration that a guy wire snapped, wrapped around it, and shattered it, sending the plane straight down into the ground. It killed his passenger, an Army Lieutenant, and broke Orville’s leg and four ribs.   His passenger could have been Theodore Roosevelt!   Their sister Katherine immediately rushed to his side, and nursed him through months of recovery.

They both made it over to Europe by early 1909, with Katherine drawing as much interest as her brothers.   Orville did get back up in the air six months after his accident, and was able to train both French and Italian pilots.   Then they all returned to the US to more acclaim, and a spectacular flight up and down the Hudson and around the Statue of Liberty.  A million New Yorkers came out to see them.

But after that their lives got darker.  Everyone with technical talent got into aviation, and their records were soon surpassed.   Wilbur spent a lot of time defending their patents, and trying to get a proper business set up for building aircraft. That seemed to be where their enormous ingenuity reached its limits – they didn’t have the heart or skills for scaling up their inventions.   It’s a well-known problem among startups to this day, and venture capitalists will often replace the founder as CEO when things get serious.

The strain killed Wilbur.  He was traveling incessantly to raise money and sell product, and it wore him out.   He caught typhoid fever in Boston in 1912, and died six weeks later at age 45.

Orville had neither the interest or energy, given his injuries, to expand the company, and sold it in 1915.  By the time of the US entry into World War I in 1917, the Feds were sick of the patent feuding among the inventors, and forced them to all to enter a common pool for a payoff of $2 million each.  Orville last piloted a plane in 1918, but was involved with NACA, the predecessor of NASA, almost until his death in 1948 at age 77.

Their ultimate tribute came much later.  When Neil Armstrong stepped on the Moon in 1969, he was carrying a piece of muslin from the 1903 Flyer.

But, yes, they stopped contributing to aviation by about 1910.   They were surpassed, as everyone is in the end, and got too involved with growing their business and maintaining their IP to keep up.  But they showed everyone else the method of success – identify each issue as it comes up and knock it down.  It’s not so much a stream of brilliant insights – it’s insights focused on what prevents progress right now.   Don’t Think Different, Think Straight.

I work in a field, semiconductors, that is the paradigmatic example of the Singularity.  When people talk about technology zooming up the exponential growth curve, they’re talking about what I do.  Since the field began in the early 1960s, there has been more improvement in integrated circuits than in any other technology except magnetic storage.  Each transistor has not only gotten cheaper by many orders of magnitude, but they’re faster, more reliable, and use less power.  Yet in spite of all this, there’s been a lot less change in the underlying ideas than you would think.

I’ve seen the progress part directly.  The first chip I worked on, the V-11 microprocessor at DEC in the early 80s, had 110,000 transistors with a gate length of 3 µm and ran at 5 Mhz.   My latest has 130,000,000 transistors with a gate length of 28 nm, and runs at 750 Mhz.   That’s not actually a big chip by today’s standards!  Yet it’s a factor of 1200 in density and 250 in raw speed.  Overall it’s about 2000 times faster, and 20 times cheaper to boot.

It’s a story of terrific technological success, which naturally leads to vast hype.  People have compared the invention of semiconductors to that of printing, writing, and even fire.  Some writers – most notably the SF author and CS professor Vernor Vinge, and the inventor Ray Kurzweil – go farther still and claim that they will bring about the transcendence of humanity, what they call the Singularity.   Things will move so far that no one today will be able to comprehend the people (or their uploaded avatars) of that era.   We’ll either be amoebas crushed by battling AIs, or we’ll all be as gods.

Yeah right.  I’ve actually been in this field for over 30 years, and what strikes me  is NOT how unfathomably different it has become, but how constant the underlying technologies have actually remained.     Let me start at the lowest level and work up:

1. The base material of chips is still silicon, as it was in the early 60s.   People tried other compounds like gallium-arsenide and silicon-germanium alloys, and they’re only used for niche products, and fewer and fewer of those.
2. The base device is the MOSFET transistor, invented in the 1960s.  They look a little different today – they’re built out of etched fins on the chip surface instead of lying flat along it, and use hafnium oxide instead of silicon dioxide for the dielectric, but it’s the same basic device that Noyce and Moore worked on. There are no more bipolar transistors, no tunnel diodes, no magnetic bubble memories, and no exotic non-volatile devices like MRAM.    Maybe the last will happen at some point, but people have been trying new schemes for decades.
3. The base circuits are the CMOS static combinatorial gate, the 6-transistor SRAM cell, the DRAM capacitive storage bit cell, and the floating gate flash memory cell, all from the 70s.   These account for practically all of the 10^21 (billion trillion) transistors made per year.   Dozens of other circuit styles have been tried, and all have failed because of excess power or poor noise resistance.  I’ve worked in some of these styles, such as boot-strapped NMOS, pre-charged dynamic, and cross-coupled cascode, and they’re all gone.   Few people even design at the circuit level any more, except for those doing standard cell libraries (and most of those are auto-generated) or analog circuits.
4. The base data types are character, integer, fixed point, and floating point, all known to Mauchley and Wilkes in the late 1940s.  Characters, at least, have expanded from the  5-bit Baudot code to the 8 to 32-bit UTF-8 Unicode.    There are no logarithmic number systems (which make divide and square root easy), or redundant binary (which avoid carries), and hardly even any support for 128-bit floating point, which we built in the 80s.
5. The base processor architecture techniques were all discovered by the 1970s.  Pipelining, caching, branch prediction, vector instructions, and out-of-order execution were all known by then.   They used to be only available in supercomputers, and now every widget that goes for more than $30 can afford an ARM Cortex-A9 processor, which uses all of them.   Even the newer multi-core processors use ideas like write-back cache coherence, multi-threading, and distributed cache directories from the 80s.  Massive SIMD and dataflow never made it.
6. The base processor language is still C from the 70s.  Lots of code won’t even rely on the enhancements in the 1999 update, C99.   People code today in a huge range of languages, but somewhere underneath them is usually an interpreter written in C, and C is in all the libraries and the OS.  It’s about the only binary compiler that most machines support.  That and FORTRAN, which is even older.  The object-oriented enhancements of C++ and Java are from the 80s. Lisp is still considered exotic, and it’s from the 60s.

Overall, it’s gratifying that the basic concepts of the field have persisted for so long.  It means that one’s initial education has not become irrelevant.  Learn the fundamental concepts and you can work for a long time in a wide set of sub-fields of the chip world.

But I don’t mean to say that there haven’t been any advances.  Chips are now in every corner of the manufactured world.  There have just been less underlying changes than what the breathless promoters of futurity would have you believe.  And I don’t mean that there won’t be significant consequences of further work.   Twenty years from now we may be living in clouds of smart dust that support terabit links to our augmented-reality sensoria.  But it’s very likely that that smart dust will be made of silicon, use standard gates and arithmetic, run code on standard processors, and that that code eventually will depend on C.   Any smart kid of the last 30 years can understand all of it.

The scandal about lead in the Flint Michigan water supply reminded me of my own brush with lead issues.  In the 2000s a program started up in electronics called the Reduction of Hazardous Substances, RoHS.   It was intended to get lead out of the solder used to attach chips to boards, and also to get rid of a lot of other dangerous materials like cadmium, hexavalent chromium, and a lot of nasty organics used in plastics.

This was a royal pain for chip makers like my company.   The lead-free solders are more expensive and don’t work as well.  They can form tiny whiskers of tin that short out adjacent wires. Worse, they don’t have as much give when the chips and boards expand and contract during temperature changes.  When a chip heats up under heavy usage, it can easily pop right off the board.  Going to lead-free solder almost brought down Nvidia when their latest graphic processor chips started losing connections on customers’ boards, and also almost did in the Xbox 360.

This is also the reason why lead paint is better.  Because it has some give, it can stay attached to surfaces through the cold of winter and the warmth of summer.   To this day lead paint is more durable than the alternatives, and is still providing good coverage decades after it was taken out of the paint supply.

In the electronics world we grumbled that no one should care if there was a little lead that was tightly encapsulated underneath the chips.  It’s not as if it was going to be eaten by children like the sweet-tasting chips of lead paint.   But Third Worlders care.  All this stuff eventually gets sent to desperately poor places for recycling.  When they grind up the boards to recover the gold that’s used in contacts and bond wires, they spread lead everywhere.  It’s a nightmare gift from the First World.  The EU was the first to actually care about this, and the US and Asia followed, reluctantly.  Now we’ve actually adapted to doing without lead, and don’t miss it.

Ultimately we have to get rid of it everywhere.   The managers of the Flint water supply thought that the old lead pipes there would be fine, even with the corrosive water from the Flint River going through them.  They were wrong, and have now harmed a generation of children there. Every poison is going to hurt somebody, somewhere and someday.

In the early 90s I began collecting stories of engineers who came to bad ends.  That’s literally bad ends, as in executions and suicide, not just failed projects.    My own career was at a difficult phase at that time, and I found it cheering.  I posted them at an ISP called The World, which was based in Brookline MA, and one of the first anywhere.  Unfortunately, they were ultimately overwhelmed by spam and left behind by other providers, and I had to leave.    The site did stay up for over 20 years, which is an eternity in Internet time.

I’ve now re-formatted and updated the posts here – Doomed Engineers – Careers Even Worse Than Yours .   If you’re suffering from excess contentment and happiness, click over and have a look.   If you know of similar stories, email me!  I’m always interested in hearing more cautionary tales.