1 Networks of tinkerers: the invention of the airplane and a model of open source innovation Peter B. Meyer, U.S. Bureau of Labor Statistics * * Findings and views are those of the author, not the BLS University of Strasbourg June 2, 2009 Agenda: (a) show history of open/sharing period and model of this (b) the transition into “industry”, and model of this. (c) Comments welcome.

2 Development of the airplane (heavier than air, with fixed wings) 1860s and on Clubs and journals appear It’s a niche activity – maybe hopeless, useless, and/or dangerous Publications do not refer much to prior work Wise, 1850; Mouillard, 1881; Goupil, 1884; Lilienthal, 1889; Langley 1891; Means Bibliography by Tissandier 1894 Survey by Chanute, who refers to 190 people/experiments Publications then refer more often to prior work. Means 1895, 1896, 1897; Banet-Rivet, Wright brothers’ powered-glider flight ~1909 An industry arises Many designs were shared openly, and documented. I seek to quantify this activity.

3 Chanute’s 1894 overview Progress in Flying Machines refers to or quotes more than 190 persons Experimenter / groupPages location (background) Maxim33 Britain (US) Lilienthal31 Germany Pénaud22 France Mouillard21 Algeria, Egypt (Fr) Hargrave19 Australia (Br) Moy19 Britain Le Bris17 France Langley16 US Wenham15 Britain Phillips14 Britain These are counts of pages referring to the individual. The people are diverse and international. Later technological histories treat these people as central. Their findings were mostly public.

4 Alphonse Pénaud Box kite findings ~1894 Did not patent, on principle. Presented/published many papers in Australia Lawrence Hargrave Engineer in France Showed importance of tail on aircraft

5 Founded company making steam engines in Berlin 1860s-80s studied bird wings and experiments 1889: published Birdflight as the Basis of Aviation : Flew inspirational hang gliders Otto Lilienthal Why? “... to soar upward and to glide, free as the bird” -- Otto Lilienthal, 1889

6 Samuel Langley Professor, then Director of Smithsonian Institution in DC Tested lift and drag of planes on “whirling table” with 30-foot arm 1891: Published Experiments in Aerodynamics Wrote to and visited other experimenters Helps make aviation study legitimate 1896: Small powered gliders flew half a mile.

7 Octave Chanute French-American railroad engineer, then writer His 1894 book Progress in Flying Machines, surveyed experiments, devices, theories Was in contact with many experimenters. Visited with Langley, Santos-Dumont, Ferber, Huffaker, Herring, Maxim and others. Corresponded with Hargrave, Mouillard, Montgomery, Cabot, Zahm, Kress, Wenham, Moy, Pilcher, Means, Lilienthals, Wrights: Letters back and forth W Wright to Chanute Chanute to W Wright

8 Otto and Gustav Lilienthals’ selected letters and contacts (Schwipps, 1993) PersonLetters Means12 Chanute11 Dienstbach5 Langley Met with, 1895 Pilcher Met with, 1895 Correspondence and references show much overlap with Chanute’s list Most cited in index of Published Writings of the Wright Brothers (Jakab & Young, 2000) Personpages Lilienthal34 Langley29 Chanute24

9 References in histories of aviation Last nameFirst namePages WrightWilbur and Orville443 ChanuteOctave303 LangleySamuel Pierpont240 CurtissGlenn Hammond198 LilienthalOtto177 StringfellowJohn117 CayleySir George103 BlériotLouis98 HerringAugustus Moore97 patents81 Smithsonian Institution75 HensonWilliam Samuel66 BellAlexander Graham65 ManlyCharles Matthews60 ZahmAlbert Francis56 MaximSir Hiram Stevens49 AderClément47 VoisinGabriel45 BreareyFrederick W.44 MeansJames44 WenhamFrancis Herbert44 PenaudAlphonse43 Counted references to persons or institutions in the books below, combined: Crouch’s A Dream of Wings (1981/2002) Dale’s Early Flying Machines (1992) Garber’s Wright Brothers and the Birth of Aviation (2005) Gibbs-Smith’s The Invention of the Aeroplane. (1966) Hallion’s Taking Flight (2003) Hoffman.Wings of Madness (2003 biog of Santos-Dumont) Jakab’s Visions of a Flying Machine (1990) Penrose’s An Ancient Air (biography of John Stringfellow) Randolph’s Before the Wrights flew: the story of Gustave Whitehead. (1966) Runge and Lukasch Erfinder Leben (2005) (biography of Lilienthal brothers) Shulman’s Unlocking the Sky (bio of Glenn Curtiss) Preliminary; almost all this is in English. Now up to 2000 persons referenced. Again the same names appear.

10 Most U.S. patents by people with aircraft- related patents before 1907 Falconnet 6 Quinby 5 Beeson 3 Bell 3 Blackman 3 Cairncross 3 Fest 3 O’Brate 3 Most German patents by people with aircraft-related patents Lilienthal, O. 25 Lilienthal, G. 9 Baumgarten 7 Gaebert 6 Lehmann 6 Hofmann 4 Ozeyowski 4 Wellner 4 Czygan 3 Fischer 3 Israel 3 Riedinger, A. 3 Many fixed-wing flying machine patents were filed before [Data: Simine Short and Otto-Lilienthal Museum] Patent counts tell a different story Little overlap with the other lists. To my knowledge none of these patents were ever licensed Aviation historians treat the patents and most patent-filers as irrelevant to the main inventions. So did Chanute and the Wrights. Claim: Intellectual property ownership was mostly irrelevant. Q: French patent database?

11

12 Motivations of experimenters Would like to fly Curiosity, interest in the problem Prestige, recognition Belief in making world a better place Make one nation safer Nobody refers to expected profits “... A desire takes possession of man. He longs to soar upward and to glide, free as the bird...” -- Otto Lilienthal 1889 “The glory of a great discovery or an invention which is destined to benefit humanity [seemed] dazzling.... Enthusiasm seized [us] at an early age.” - Gustav Lilienthal

13 Parallels to open source software and open-, user- or distributed- innovation Autonomous innovators (not hierarchy, not cult)... with various goals  Want to fly!  Hope for recognition, prestige, fame, maybe fortune  Curious, interested in the problem  Bring peace, or make own nation safer... who share technical info with international public  Intellectual property set aside Authors, evangelists, organizers have valuable role

14 Systematic measures of innovators & innovations Who was involved before the industry existed? To whom do the major inventors refer? To whom do later historians and analysts refer? Who patented? How much? Did patents matter? Such measures help distinguish open source/distributed innovation from secret or intellectual-property-type innovation These measures have problems. But one gets a super-index with “everyone relevant” The “airplane case” should be at the open extreme.

15 Microeconomic model (Meyer, 2007) Imagine self-motivated tinkerers making progress on some project They invest time, effort, money into experiments Let two tinkerers’ experiments add value to one another’s projects Say they are not in competition because they cannot foresee a marketable product for now high “technological uncertainty”)  They’d agree to share findings with one another  They’d specialize to avoid duplication  They’d standardize on modular designs and tools (Market processes are not necessary for these effects)  They don’t bother with intellectual property (there’s no gain)  There is a role for an author / organizer / evangelist to expand the network and reduce duplicative efforts.  A tinkerer might change if the technological uncertainty lifts

16

17

18

19

20 Investment to gain more from network Experimenters and programmers specialize within project  avoids duplication; develops focused skills and tools They standardize and modularize interfaces  so some progress by other tinkerers will snap right in They re-design for ease of use, learning, and development They recruit/evangelize/publish technology & network  As Chanute and Lilienthal and others did with publication and demonstration  This raises inflows of info and reduces duplication  Like open source programmers Stallman, Torvalds, other charismatic founders In model: Tinkerers receive fraction f of progress by others Suppose for cost c s, could raise that inflow to f 2 > f

21 Worth doing if: A player benefits more from this if, ceteris paribus: other tinkerers produce a large flow of innovations p 2 ; gain in useful innovations from the others (f 2 -f) is large. cost c s is small For β=.95, p=.07, f 2 =.55, f=.5, payback is 1.33; worth usual investment.  Specialization and standardization are natural in tinkerers’ network. Tinkerers in model would be willing to pay.  Don’t need market processes to explain this behavior Specialization, standardization, modularization, redesign, evangelism

22 Intellectual property and secrecy In each episode (airplanes, computers, open source software) many people want to avoid intellectual property and/or secrecy Hargrave thought all aerial navigation work should be published and nothing patented till something really worked. Chanute wanted to get all information out in the open Analogously: Stallman, Fogel, other open source programmers In model: intellectual property payments for sharing results of experiments would introduce noise and friction. Incentives aren’t needed, and don’t help, and there are no profits to split.

23 Wright brothers as open-sourcers Wilbur and Orville Wright ran a bicycle shop. They read up on gliders and try flight experiments. Motivations: “I am an enthusiast... I wish to... add my mite to help on the future worker who will attain final success." -- Wilbur Wright, 1899 "At the beginning we had no thought of recovering what we were expending, which was not great..." -- Orville Wright, 1953 They published articles They spoke at conferences Chanute, others visited and stayed in contact

24 Wright brothers as open-sourcers First letter to Chanute, May 13, 1900: “Assuming then that Lilienthal was correct...” [Wilbur explains what he will do differently.] “.... my object is to learn to what extent similar plans have been tested and found to be failures, and also to obtain such suggestions as your great knowledge and experience might enable you to give me. I make no secret of my plans for the reason that I believe no financial profit will accrue to the inventor of the first flying machine, and that only those who are willing to give as well as to receive suggestions can hope to link their names with the honor of its discovery. The problem is too great for one man alone and unaided to solve in secret.” “the apparatus I intend to employ... is very similar to the "double-deck" machine with which the experiments of yourself and Mr. Herring were conducted in ” Chanute’s reply May 17, 1900: “I believe like yourself that no financial profit is to be expected from such investigations for a long while to come.” Wrights help test Octave Chanute’s glider, Oct 1902 Wrights’ 1900 glider

25 Wright methods and inventions Wind tunnel with smooth air flow Tested many wings systematically Propeller invention: shaped like wings, with lift going forward This produces ~40% more pulling power. This design idea lasts to the present. They are skilled, precision-minded toolsmiths, in a workshop every day. They flew craft as kites and gliders both, many times No landing gear, no engine. Their piloting invention had to be learned, like on bicycle

26 Wrights exit the open-source network Late 1902: they become more secretive, apparently because of wing design success 1903: They filed for a patent on their control mechanism for the wings. Their secrecy and tight hold on patent rights lead to permanent conflicts with Chanute, Curtiss, and others. Wrights’ first powered, controlled fixed-wing flight Dec, 1903

27 Model of exit from network, starting firm Suppose a tinkerer envsions how to make a profit from project A worth more than the present value of staying in the tinkerers’ network. Then tinkerer can exit network agreement conducts directed R&D stops listening to network becomes an entrepreneur, owns a technology

28 Suppose at the start of each turn a tinkerer may see through the technological uncertainty and how to design a product With small probabilities π 0 and π 1 each turn has a valuable insight like that, and quits the network to start a firm, modeled as taking a large expected utility payoff of M. (Anticipating this could happen they may agree on a fee x to exit the network) Then there is a payoff expression for that, something like this: Previous results hold, while the tinkerer’s network exists But incentives change once tinker sees how to make the quantum leap to being an entrepreneur. Modeling exit to startup firm

29 Notes: Sometimes guess at start year. Some of these failed quickly too. Counts do not include nonprofit research, govt/military, nor subsidiaries. Most make airplanes; also engine makers, propeller makers, flying schools, exhibition companies. Little overlap with earlier lists! Almost no overlap with 19 th century experts.

30 Conclusions (1) Key assumptions of this model:  motivated tinkerers, perceiving progress  no perceived path to profit  cheap communication This generates inventions, as by:  Hobbyists  “Skunkworks” inside organizations  Basic researchers  Better communications – an effect of the Internet is that it enables networks. An industry can arise this way

31 Conclusions (2) Experimenters aren’t in economic models of employees, managers, investors, consumers, government.  Need model of self-motivated “tinkerers” cooperating Phases observed:  Tinkerers worked in small groups ( , here)  Tinkerers networked more (1894 to )  Commercialization (since then) Future work More lists of publications, patents, clubs, references, bibliometric statistics, companies/organizations, designers,....

32 Founded company making steam engines 1860s-80s studied bird wings and experiments 1889: published Birdflight as the Basis of Aviation : Flew inspirational hang gliders 1896: Died after crash. Otto Lilienthal Motivation: “... to soar upward and to glide, free as the bird” -- Otto Lilienthal, 1889 “The glory of a great... invention... destined to benefit humanity [seemed] dazzling.... Enthusiasm seized [us] at an early age.” -- Gustav Lilienthal, 1912

33 Lawrence Hargrave Retired young Ran many creative diverse experiments Several flapping-wings designs Innovative engines Box kites showed layered wings were stable and had lift Did not patent, on principle. Published a lot

34 Octave Chanute Chicago railroad engineer 1880s: takes interest in flying machines 1894: Publishes Progress in Flying Machines, which surveyed experiments, devices, theories 1896 and on: ran experiments on gliders Corresponded with, supported, visited many experimenters Chanute preferred findings to be shared so as to speed progress “[By writing lucid articles], personal correspondence, and visitation, he inspired and encouraged to the limits of his ability all who were devoted to the work.” --Wilbur Wright, 1911

35 An explicit reference to the person’s name or quote from the person In a relevant book (11 so far; at least 15 to go) Text in main content, preface, forward, introduction, appendices, pictures, tables, and figures Table of Contents and indexes don’t count References to something named for the person count. (Should they?) Events after 1909 shouldn’t count (not done yet) Only events related to aircraft work should count (not done yet) On this view, biographies “over-refer” to the subject person sometimes they leave the subject person out of the index (!) Groups (brothers Wright, Lilienthal, Montgolfier, Tissandier, Voisin; likewise institutions or groups are referred to as groups and other times as individuals)  Counts are preliminary and can never be perfect What’s counted as a reference

36 Role for author / moderator / evangelist Chanute corresponded with, visited, introduced experimenters, and published book In model: A tinkerer’s best opportunity for progress may be editing, writing, speeches, evangelism  To welcome future tinkerers who could generate progress  To avoid duplicate efforts, thru standards and specialization  a uthors/evangelists are another kind of specialist tinkerer Octave Chanute, 1894: “The writer’s object in preparing these articles was threefold: 1. To satisfy himself whether... men might reasonably hope eventually to fly To save... effort on the part of experimenters trying again devices which have already failed. 3. To... render it less chimerical... to experiment with a flying machine....” Analogously: Lilienthal’s public demonstrations; Felsenstein at Homebrew; open source programmers Stallman, Torvalds, etc.

37 (1) Network: a population of agents with i nterest in a problem (a 0 ), worthwhile opportunities (p), information flows between them (f)  experimentation and socially constructed “progress” No pool of information, or incentive structure, or technical measure of improvement. (2) Race to be first (space race; genome project) (3) Collective invention (Allen, 1983)  but those are (a) firms, (b) not paying costs to experiment (4) To earn income or wealth indirectly  Start company, or license patented invention  signal to employers; get hired as engineer (Lerner and Tirole, 2002) Alternative models of invention