Cybernethics / Cybernéthique

The G1 has alluring hardware specifications and does everything you could expect a small computing device to do. Alas, it doesn't do any of it well. It is not by any means a decent Internet access device, a decent digital assistant, or even a decent MP3 player. What is worse, it is a computer on which you are not the one in control. It may be an excellent phone, but only because by the telecom industry has set very bleak standards indeed for what a phone may be expected to do. Moreover, if you use any of its advanced functionalities, then the battery situation is even worse on the G1 than on my old dying phone. All in all, it is a luxury I'd rather not afford.

I'm now the happy owner of the cheapest phone (marginally free) with the cheapest plan ($30 a month). This brings me 95% of the satisfaction of the G1 with none of the extreme frustration, for less than half the price and half the weight.

I am looking forward to a future of telephony over decentralized WiFi networks where government-backed telecom companies can no more force consumers into buying outrageously priced plans that can only be used with inferior locked down devices.

A Boston Lisp Meeting will take place on Monday, June 29th 2009 at 1800 at MIT 34-401B, where Eli Barzilay will speak about Implementing Domain Specific Languages with PLT Scheme.

Additionally, we are still accepting proposals for up to two volunteers to each give of a 5-minute Lightning Talk (followed by 2-minute Q&A).

Also, there will be a buffet offered by ITA Software. Registration is not necessary but appreciated. See details below.

The Lisp Meeting will take place on Monday June 29th 2009 at 1800 (6pm) at MIT, Room 34-401B.

A Boston Lisp Meeting will take place on Tuesday, May 26th 2009 at 1800 at MIT 34-401B, where Norman Ramsey will speak about Using Higher-Order Functions and Continuation-Passing Style to Make Dataflow Optimization Simple.

Additionally, we are still accepting proposals for up to two volunteers to each give of a 5-minute Lightning Talk (followed by 2-minute Q&A).

Also, there will be a buffet offered by ITA Software. Registration is not necessary but appreciated. See details below.

The Lisp Meeting will take place on Tuesday May 26th 2009 at 1800 (6pm) at MIT, Room 34-401B.

And so for all I know, his eventual abduction by a ghost may be Don Giovanni's latest invention, staged by himself to get rid of all these former victims who are now running after him and spoiling his fun, all the while silently mocking their superstition and gullibility from behind as they sing their final moralizing Questo è il fin in the front of the stage. At least, that's how things would turn out if I had a say in a production of Mozart's masterpiece. Certainly it would bother me a bit to let the scoundrel get away with murder scot-free, but after all unhappy endings are par for the course in Opera.

Lightning Talks are series of short strictly timed five minute talks each followed by two minutes for questions and answers during which the next speaker gets prepared (by e.g. disconnecting the previous speaker's laptop from the projector and connecting his own instead). The formula was used at ILC'2009 in sessions of 4 to 8 talks, and was a tremendous success.

Indeed you don't need to have that much to say to give a Lightning Talk, and so the formula lowers the barrier to entry and allows for a wider diversity of ideas to be presented at such events. At the same time, the strict constraint forces the speaker to be more concise, maybe even poetic, in conveying his ideas. Several participants remarked that since Lightning Talks were both so short and precisely timed, you could rehearse them many times before you gave them, making delivery more powerful. Additionally, some argued that the short duration limits the potential for either speaker humiliation or listener boredom, once again giving incentive for people to participate on both ends of the communication, promoting the exchange of ideas.

And so, I instituted such Lightning Talks at the Boston Lisp Meeting; so far, they have been well-received, and many people have subsequently volunteered to give one -- which makes me hopeful about their positive impact on the exchange of ideas within the community.

beach reported on IRC "In fact I agree with a French colleague of mine who works for the water company, that you have more and better ideas when under the influence, but then you need to be totally clean when working out the details. Luckily, grad students work out the details these days :)"

My answer was that I fell in the wine barrel when I was a kid, and the effects on me are permanent. Maybe that's why I have all these ideas, but can never work out the details.

rsynnott remarked that he normally attributes that to laziness, in himself. I suppose that could be a valid other name for that. In any case, that's why I'm now working towards getting minions to work for me.

Reisner's Rule of Conceptual Inertia: If you think big enough, you'll never have to do it.

If many top notch computer programmers, in a field where people are largely selected for their ability to understand causes and consequences, fail to understand the difference between a constant and a variable with respect to a given choice, how can you expect people from professions that don't thus select their members to fathom the difference?

And so most people, most bright people even, have the greatest difficulty understanding the difference between variables under their control, and variables not under their control -- be them variables under someone else's control, hypotheses of a given thought experiment, natural phenomena, or universal constants.

Actually, the people who are trained to understand and explain the nature of a choice, its benefits and costs, are economists. And even most economists fail (cám ơn, Constant). Only austrian economists are properly trained, but they are marginalized in academia and media by statist economists whose very raison d'être is to produce cover stories for the depredations of the State, by denying of the very basic principles of economics.

This is how otherwise intelligent people believe that taxes and prohibition can abridge the consumption of services or goods with inelastic demand -- when by definition of inelastic demand, it will instead displace other consumption, and lead the victims of prohibition to poverty and crime when desperate measures are required to indulge in the inelastic behavior. What is constant is the inelastic demand for addictive drugs, that depends on the will of the people targeted by the prohibition, rather than the propensity of said people to abide by the law when the law is changed. And so what the prohibition law does is introduce more crime and law enforcement, displacing peaceful and productive activities at a huge cost to society.

The same people may believe that raising taxes and prohibitions (such as minimal wage) on services or goods with elastic demand will help them increase transfer of wealth from the designated victims (taxpayers, employers) to the designated beneficiaries (tax consumers, employees). But once again, what is constant when you enact laws is not the behavior of the victims, but their will, goals, desires and preferences. And so the consequence will instead be that less productive employees will not find a job, that taxed capital will either run away or be exhausted, taxed work will be less intensive, and black market activities will increase.

All in all, the only people who benefit from laws are for the professionals who specialize in either enforcing or countering the law: scammers, whether legal (politicians) or il-, racketeers, whether legal (bureaucrats) or il-, and their goons with guns, whether legal (cops) or il-.

Oppression thrives on people being systematically unable to properly understand the nature of choice and to apply this understanding to Political Economics. And yet, most anyone can be trained into understanding it. If you want to stop Oppression, start by extirping the seeds it sows inside you. Learn (Austrian) Economics. Knowledge will make you free -- and others with you.

Statists will argue that individuals on a free market aren't really free to choose, because they are constrained by reality and society. Then they propose Government as the solution. But Government isn't above reality, and neither is it above society. Its choices are just as constrained. All Statists actually argue is that they support the arbitrary use of force by rulers to coerce citizens into doing their bidding.

I am always amazed when people fail to distinguish between constants and variables. I am all the more amazed when the victims of such confusion are the otherwise brilliant implementers of programming languages. You'd think that if anyone knows the difference between a variable and a constant, it would be a programming language implementer.

Arguing that the current average experience and proficiency of programmers is pretty low, and that programming malpractice incurs huge costs including the risk of occasional death of end-users, MIT Professor and Scheme co-inventor Gerald J. Sussman has proposed at ILC'2009 and on several other occasions [Update: as a joke] that programmers should be certified before their are allowed to practice. I heartily agree, but for the small prevention that follows.

While discussing with fellow libertarians about the foolishness of sacrificing one's own welfare for the sake of avoiding State taxes and control, I remarked that as a Frenchman, I well knew the blessings of unilateral surrender.

In this recently unearthed posthumous book by J.R.R. Tolkien, the fantasy world filled with dragons and magicians has evolved from warring medieval times to peaceful modernity; furthermore, in the second volume, it is resolutely entering post-modernity.

As previously announced, there will be a Boston Lisp Meeting taking place Next Monday, March 30th 2009 at 1800 at MIT 34-401B, where Carl Eastlund will speak about Modular ACL2. http://fare.livejournal.com/140695.html

Additionally, having observed the success of the formula at ILC'2009, I'm instituting Lightning Talks at the Boston Lisp Meeting. At every meeting, before the main talk, there will be two slots for strictly timed 5-minute talks followed by 2 minutes for questions.

Our first Lightning Talks will be given by Matt Knox and François-René Rideau. http://fare.livejournal.com/140960.html

Matt Knox has been hacking on Ruby and Scheme for the last 3-4 years. Looking at Rails, and how to fundamentally improve it, he came up with GoaloC (Generate on a Lot of Crack).

I, François-René Rideau, am a story-teller who dream of myself as a software author. I will address designers of computer languages and systems about Better Stories, Better Languages.

Dinner: ITA Software, is kindly purchasing a buffet to accompany our monthly Boston Lisp meeting. Anyone who attends is welcome to partake. However, we appreciate it if you let us know in advance you're coming, and what food taboos you have. Tell us by sending email to boston-lisp-meeting-register at common-lisp.net (you remain anonymous, no spam, no ack unless requested).

Finally, our next next meeting will happen on Monday 2009-04-27, and will feature Noah Goodman on MIT-Church, a non-deterministic Scheme. The previously announced speaker, Norman Ramsey, will hopefully be speaking in May, on purely functional dataflow optimization (in Haskell).

Carl Eastlund will give a talk about Modular ACL2.

The Lisp Meeting will take place on Monday March 30th 2009 at 1800 (6pm) at MIT, Room 34-401B.

You can now find on my web site my essay From Creationism to Evolutionism in Computer Programming, subtitled The Programmer's Story: From Über-God to Underdog. As compared to the previous MSLUG conference (see slides), the essay contains a vastly expanded second part on the prospective future of the evolution of programming, and my vision for TUNES. Enjoy!

Abstract: Programming tools imply a story about who the programmer is; the stories we tell inspire a corresponding set of tools. Past progress can be told in terms of such stories; Future progress can be imagined likewise. Making the stories explicit is a great meta-tool... and it's fun!

If you don't care for your survival-or-not, I have good news for you: your survival-or-not doesn't need your care; thanks to this free voucher your survival-or-not will take care of itself, picking the "not" branch by default.

This Special Darwin-Day Voucher, is valid for any "you". We do not discriminate against individuals, races, genes, memes, or life substrate. You may print or photocopy this voucher at will. Redeem at your nearest cemetery.

Dimitris Vyzovitis will give a talk about Programming gerbils: Distributed programming with PLT-Scheme.

The Lisp Meeting will take place on Monday February 23th 2009 at 1800 (6pm) at MIT, Room 34-401B.

Common Lisp pathnames have long been a source of frustration for users and implementers alike. I'll argue that the deep reason for this frustration is that the Common Lisp standardization process resulted in declaring as fixed an interface to a moving functionality, preventing users' needs to be addressed with no one in charge of addressing the discrepancy.

Once you understand programming language documents not as mere technical documents, but as social tools to coordinate people, you can critique them from a new perspective. Who promises to whom to take responsibility for what in exchange for what? What about the contract brings value to both parties, and what doesn't? These are questions you should be asking when working on programming language design. Contracts are useful when they promote an efficient division of labor, whereby the more competent in a specialized field easily take responsibility for work in said field at the benefit of others, at low cost. Contracts are harmful when they create unnecessary work, when they assign responsibilities to the wrong people, when they impede future improvement in division of labor. Design your contracts carefully; include what works, exclude what doesn't.

A sane moral person shall not feel responsible for other person's pain, except when the first person indeed caused the pain by his own actions. And even despite any interference that one may suffer from other people, to claim that the primary responsibility on one person's fate lies in the acts of other people is to deny the fact that one is one's own moral agent. Temporarily denying one's responsibility on one's own fate is indeed to treat one as a child. Permanently denying one's responsibility on one's own fate is indeed to treat one as an animal. Those who deny this responsibility are not one's friends, but one's most mortal enemies.

Why should one be responsible for the fate of other people one has never interacted with? There is no reason, and cannot be one. On the other hand, there is a moral duty of justice, to not actively harm other people, to not violate on their life, liberty and property -- or to have compensate them for the trouble. And this basic duty specifically forbids us to sacrifice anyone in the name of any Grand Scheme.

If you neglect some aspect of computation in designing your programming language, then the people who actually need that aspect will have to spend a lot of time dealing with it by themselves; and if you don't let them do it by extending your language, they'll have to do it with inarticulate barkings rather than civilized sentences.

Suppose some causal relationship between some measurable phenomenon A and some measurable phenomenon B, through some undetermined positive feedback loop (respectively a constant statistical proportion, or some other simple mathematical construct). But how much does A depend on B? Well, do it scientifically, by computing with lots of experimentally measured numbers! You put some historical numbers in, you fit your model to those postdictions, and ask your simulation to make predictions about the future. What do you get? An exponential future growth of A due to B (respectively, a proportional growth of A, or whatever follows your causal law). And what happens if you force B to be higher or lower in the future? Then the growth of A correspondingly goes faster or slower. This establishes a clear dependency of A to B, doesn't it?

Well, yes, of course -- it establishes the dependency you introduced in the model, to begin with. The positive feedback loop is a conclusion of the model only because it was inserted as an explicit premise of the model. To wave around your model with lots of numbers and graphics as a proof that something alarming is happening and that production of B should be stopped (respectively started), is a fallacy: it is a petition of principle hidden under a lot of verbiage, it is intimidation using the usurped authority of "official truth", it is confusion of the layman by a mass of numbers and complex maths, it is numerology dressed up as science.

Unhappily, that kind of pseudo-science is what is taught in universities and used as the basis for public policy, in such domains as economics, sociology, and as of late, climatology. And it is pseudo-science precisely because it is used as the basis for public policy, in a positive feedback loop where government officials fund and establish intellectuals who justify their power.

Next time you're shown a model, skip all the numbers, all the mathematical formulas, the graphics, the computed projections, the "scientific" conclusions, and the policy recommendations. Ask what were the hypotheses on which the model resides; what are the relationships assumed between various phenomena, and why are these relationship assumed not to move in general, and in particular when the recommended policy changes the system?

More often than not, you'll find that the general conclusion was already assumed as a hypothesis to the model, and that the whole model is a swindle to smuggle the premise as if it had somehow been established by the model.

Sure, some keynesian economist will show you some beautiful equation, with aggregates that add apples and oranges, whose multiplicative relationship to some government-controlled variable is magically supposed to be constant, unlike simpler variables. "This equation is true by definition!" will he even claim. Sure. Just like it's true by definition that being hit by a magic long sword +2 will cause you 1d8+2 of hit point damage. That's also the definition, straight from the rulebook. The definition is tautologically true in the corresponding model. Now whether the model accurately describes reality, that's quite a different question. At least the D&D rulebook lets me have fun with friends and doesn't serve as a pretext to massive armed robbery.

In the conclusion of his article "When Are Two Algorithms the Same?" Yuri Gurevich (who BTW fails to cite Mitch Wand on FEXPR) notes:

In addition to the Curry-Howard correspondence, there is another connection between proofs and computation. If we take an algorithm for computing a function f and we run it on input x obtaining output y, then the record of the computation can be regarded as a proof, in an appropriate formal system, of the equation f (x) = y.

But as Philip Wadler remarks, for each programming language and type system, there is a Curry-(deBruijn)-Howard isomorphism between that language and the proofs of some logic proposition system. The connection that Yuri noticed, and that I incidentally used as the basis for my 1998 mémoire de DEA (~ Master's Thesis), is thus not a wholly novel connection but indeed a particular case of such a generalized Curry-Howard isomorphism, for the type system whereby evaluation trace or tactic T is of type E↓ iff it shows how evaluation of expression E can terminate.

The computational content of the proof that E terminates is a concrete evaluation trace. In the proper framework, an abstract evaluation tactic or strategy may be of type E↓ if it can be reduced to such a concrete evaluation trace in the "type"-directed context of evaluating E. Thus, if you're only considering expressions from a system that has strong normalization properties, then the trivial strategy "reduce wherever you can" has type E↓ for all E in the system, which isn't very informative. If on the other hand you're considering a richer system where termination is not decidable, then you can have informative proofs that include important information on what choices to make at various points of the evaluation attempt. If evaluation of an expression E doesn't terminate, there is indeed no terminating evaluation trace and no computational content. To achieve a full intuitionnistic logic, you can conceivably extend your computation system with expressions such as E→F that terminates iff termination of E implies termination of F, yielding a function that takes an evaluation trace of E as input and output an evaluation trace of the F. Use ⊥ for F and you have the usual negation — and of course, just because E doesn't terminate doesn't necessary imply that E→⊥ terminates, which is the stronger statement that you can actually prove from within the system that E doesn't terminate. Hopefully for your system, evaluating the Gödel sentence Y(G↦G→⊥) won't terminate.

NB: if you're willing to stretch it a bit more, E itself is of type ⊳V iff some trace or tactic shows it evaluates to value V; but then your proof object is trivial and you moved all the intelligence to the opaque evaluation metasystem, which makes the correspondence useless, unlike the previous point of view, that is actually useful in enabling the same tools to do more things (or the same things to be done by fewer tools).