[AFUL] Livre: La bataille du logiciel libre Livre: Libres enfants du savoir numérique Livre: Liberté Egalité Business [No Software Patents] [Free Speech Online] [Privacy Golden Key]


Hardboiled Economics

Bernard Lang
INRIA - July 17, 1997

From my electronic Webster :

          hard-boiled \-'bo[0xC7]i(e)ld\ adj
          (1886)
          1: devoid of sentimentality: TOUGH <a hardboiled drill sergeant>
              <a hardboiled detective story>
          2: HARDHEADED, PRACTICAL <handle aid programs on a friendly but
               hardboiled business basis -- N.Y. Times> 

Wisdom flows from Brooks of experience - (Bokonon)

Why hardboiled ?

Well, there is that story about how a mathematician will cook a hardboiled egg, given a water tap, a pan, a burning stove and of course a raw egg. He does it more or less like you would, or I, i.e. he fills the pan with water, puts it on the stove, and when the water boils he drops the egg in it for 10 or 12 minutes.
The interesting part of the story is a second problem : the mathematician is given a water tap, a raw egg, a burning stove and a pan full of boiling water. Any good mathematician will react instantly by throwing away the boiling water (in the sink because they are all nice people), so as to reduce this new problem to the previous one he already solved.

Although economics is a fairly old science, it is now reaching a maturity that requires an increasing use of mathematics. And economists become more and more like mathematicians, at least in some respects.

For centuries, actually for as long as mankind has existed, economics was for a good part a study of how to manage scarcity. Producing required more manpower than available, and there was not food or goods enough for everyone. So various schemes were devised to structure society, and to structure the flow of goods, to ensure what was considered an adequate (fair ?) distribution of goods, or lack of them, within the population. And it sort of worked, improving all the time, though with lots of side-effects -- such as misery, wars, revolutions, and the like -- which do put some spice in an otherwise dull existence.

But now the world has changed and economists have a new problem. Due to mechanization and other form of technological progress, scarcity is no longer. We produce too much food, too many cars, TVs, and other essentials. What is more, much of the economy is now going immaterial, with cyberspace, movies, computer programs and other intangibles, becoming more intangible as their material support, formerly paper or film, is becoming less defined and localized. And the new thing about intangibles is that they no longer obey what mathematicians call linear logic, i.e. they can be reproduced indefinitely at essentially no cost, well, I mean no labour and no mate'rial expense (sorry about this lapse). If I take your car, you no longer have it, but if you copy this text from my machine, I still have my copy, and most likely I will not even notice. So scarcity cannot exist naturally in an immaterial world.

The hitch is that, with these changes, with this lack of scarcity, the society as we know it can no longer work. For life to be enjoyable, one needs privileges, things that others do not have. And if everyone gets more or less everything, how do we distinguish between those who are deserving and those who are not ? How do we reward ? For it is well known that scarcity is what makes the world turn round, and that no one ever does anything, except in the hope of a reward, like Vincent van Gogh who is now selling his pictures for hundreds of millions, or Paul Erdős who made a fortune with his mathematics, or Richard M. Stallman and Linus Torvalds, the software tycoons.

But economists, who are the theoricians of how the world and its riches should be managed, have learned from mathematicians, and they know how to cook an egg. And for them, the solution to our woes is now clear, our unbearable overabundance of everything is now an easy problem to solve. All we have to do is to reintroduce scarcity where it is now missing, and we are back to the previous situation that we have happily known how to handle with riots, diseases, and ignorance.

So, much thought and ingenuity is now devoted by economists and others to find the best ways to destroy goods (only to preserve the market prices, mind you, which are an essential commodity), to put up barriers to prevent free dissemination of ideas and immaterial ressources, and to make sure that only the rich and deserving (well, they are the same) will get a proper education for their children. Now you know why I call modern economic practices hardboiled.


In a world without fences, who needs Gates ? - (author unknown)

Maybe you do not really believe me, but I have some hard (boiled ?) data to back up my understanding. To be honest I did not look very hard, but being interested in werewolves and software, I chanced upon an IEEE paper by Brad Cox, who will not only solve economic problems the hardboiled way, but will at the same time produce the philosophal stone that will turn our bad programming technology into perfection. Admittedly Brad Cox is only a would be economist, but his paper Objects as Property is such a nice illustration of hardboiled economics that, in my opinion, he definitely deserves an Ig Nobel prize in economics for his contribution.

Noticing, after Leonard Read, that the simplicity of a pencil hides a complex social and production structure, Brad Cox had the wonderful idea to apply the same economic principles to the intricacies of software, forgetting at least one small detail, viz. that the pencil is indeed simple in itself, even though produced cheaply by a complex society, while software is inherently complex, even though the complexity can sometimes be hidden. Pencils have nothing to hide (well, at least mines do not). However software does not necessarily require a complex social and production structure. Since Cox seems to like these words, complexity is an exogenous property of the pencil, while it is an endogenous property of software, unlike conservation of mass. But never mind.

Cox's main argument is that we need more well designed small software modules, but there is no economic incentive to build them, because people get rewarded only when they produce, and sell, large objects. So let's find a technical way to reward financially small modules (by adding usage accounting and billing complexity to the code) and we'll have good ones. He fails to tell us what the incentive will be to use them. Actually, the people from Free Software Foundation and the League for Programming Freedom have exactly the same worry. They also assume that complex immaterial structures, should be built from well engineered smaller ones. Their difference with Brad Cox is that they believe in people and good will, and they try to take advantage of the natural non-linearity of software, its lack of mass conservation that allows duplication at no cost, rather than try to fight it to get it back in the old material molds, the hardboiled way.
Their point of view, thus, is that, since software is non-linear, and once produced can be made available for all and for ever, we should produce good components, and then make them free (after all, work is no longer needed). Then people producing larger structure will have a definite incentive to use the good quality small components that have been made available, and software will get both better and easier to produce (hence, hopefully, cheaper). Having to pay for every small nut and bolt is a deterrent from reuse, even if only because it is hardly manageable in a non-linear world. Free programmers have not just proposed a theory, but they have also put it into practice, and produced components and systems that are widely used and superior in quality to those produced by the commercial world (as attested by several studies), while we are still waiting for the hint of the beginning of a feasability study of Brad Cox's proposal.

I guess this shows the difference between people who believe in money, like most modern economists, and people who believe in people.

The fact is that this is the (free)way mankind has been following for centuries with knowledge, in mathematics for example. And it has allowed for the development of knowledge modules (mathematical theories, for example) nicely built on top of each other, and being changed or refined over time by competent people to improve their effectiveness and expressiveness, a necessary social process that is impossible for copyrighted software. But no one is being paid when a theorem is used, and that is why we use them so freely for everyone's benefit. The only incentives are quality, recognition, personnal accomplishment and general welfare.

The main advantage is that scientists, or free programmers, take the time to do things right, and to redo them. Quality is the ultimate goal. Industry, on the other hand, has all kind of strange constraints, like time to market, or doing it like the guy next door. Money is the ultimate goal. The main reason they buy some software is because others are buying it, even when they know it is not the best. Or worse, because they believe the others are going to do it, and they all con each other into doing it wrong : "follow today's fashion and you won't be blamed for failure". The same holds with technological choices. But free programmers are free in many senses, including the freedom to chose the right solutions over the popular ones. If science had stuck to popular solutions, the Earth would still be flat. Industry as a whole is a completely irrational structure, even though individual decisions are rational, because its individual components keep playing some variant of the prisonner's dilemma with the globally losing strategy. And we all lose because those decisions govern our lives.

To achieve better results in the world of intangibles, we must change our ways, rather try artificially to impose them on this new universe. How they are to be changed, to still somehow account for contributions, is still to be debated. But to throw away the boiling water is certainly not the answer.


The Cox steering for the banks should expect shallow results - (rowing manual)

Post Scriptum. It appears that the value of software increases with the number of users, while producing more copies costs nothing (unlike, for example, telephone lines). Hence Microsoft should be paying each new user of Windows, rather than the opposite, for increasing the value of its asset. Microsoft actually endorses this point of view, and has been supporting it concretely, for example by giving free subscriptions to the Wall Street Jounrnal On-line or other freebies to people willing to use Microsoft Internet Explorer. This is why Microsoft is activeXly opposing subversive organizations like BSA whose ludicrous economic analyses and irresponsible actions undermine market progression and proper deployment of Microsoft software base, particularly in the former communist block and in third world countries.

Juillet 1997, Free reproduction © Copyright Bernard Lang [F1450324322014] Licence FDDL v1
http://pauillac.inria.fr/~lang/licence/v1/fddl.html

[end of document]


[HOME] Modifié le: Wednesday, November 12, 1997 : Last modified
S.V.P. Veuillez envoyer commentaires & contributions à Bernard.Lang@inria.fr,
Please: report bugs, send comments & contributions to Bernard.Lang@inria.fr,
Original URL of this page: http://bat8.inria.fr/~lang/ecrits/hardboiled/index.html

Le contenu de ces pages relève de la seule responsabilité de leurs auteurs, et ne représente pas le point de vue officiel de l'INRIA. Les informations sont présentées de bonne foi, mais leur exactitude ne saurait être garantie.
All documents are under the responsibility of their authors, and do not represent the official views of INRIA. The information is given in good faith, but its accuracy cannot be guaranteed.

Powered by Apache