# Got formula?

This is a little late for any Londoners out there, but Edge had a collaboration with the Serpentine Gallery that debuted last weekend. It’s a collection of scientists’ and technologists’ and artists’ answers to the question, “What is your formula? Your equation? Your algorithm?” All of the answers are on display at the site now, so have a browse. There’s a little bit of everything, from obvious truisms (like mine) to detailed, specific formulae, to weird guesses, to stuff that is outright crazy, but nobody said exactly the same thing (which is probably a reasonable outcome from Pinker’s equation, if nothing else).

1. says

My algorithm for composing music:

1. Get up early, boot up music software (Digital Performer, Sibelius 5.1, Finale).

2. Play music in head into software.

3. Revise music.

4. Revise music again.

5. Repeat steps 3 and 4.

6. Lunch

7. repeat steps 3 -5.

8. Dinner.

9. Family, practice piano, read, listen to music, concerts.

10. Sleep.

11. Repeat from step 1.

It is the most enjoyable life one can imagine, except when the music won’t come out. Then it’s hell, and the algorithm stops. Fortunately, that doesn’t happen too often.

2. MH says

Sorry this is off topic, but I thought you might like to tear appart this article by Dinesh D’Souza.

3. says

Of course, I have to go with
(int) cabin dcabin
= log (cabin)

(But really it’s log (cabin) + C, aka “Houseboat”)

An oldie, but not a goodie.

4. Tom says

John Gabriel’s Greater Internet Fuckwad Theory:

Normal Person + Anonymity + Audience = Total Fuckwad

5. says

MH,

D’Souza claims to have challenged Dan Dennett in that article and that Dan Dennett replied on his website. Assuming he means Dan Dennett’s homepage, I’m unable to find anything addressing D’Souza on there.

As for the argument itself, from what I can see, D’Souza sounds about as intellectually serious as a rock. Most scientists out there seem comfortable with the fact that our cognitive and sensory apparatus is often suboptimal when making observations (quantum theory is counterintuitive for a reason). D’Souza once again seems content to attack big, fat strawmen.

6. says

Dinesh D’Souza claims to be piercing atheism but what Kant argues in not in tension with atheism at all.

Kant argues correctly that our capacity to reason is limited by our ability to perceive. This limitation places an implied limit on any kind of truth claims people can make. Indeed, this realization is far more commensurate with a limited-knowledge epistemology such as atheism than anything that D’Souza has in mind.

Basically, what D’Souza is saying is “You don’t now that this God theory isn’t true.” Except that he’s saying “Kant says: ‘you don’t know that this God theory isn’t true'”. Even Dawkins admits that the God theory cannot be conclusively proven to be false merely by our capacity to reason. The typical arguments made by atheists are not that God Theory X is “untrue”, but rather that God Theory X is unsubstantiated (not to mention inconsistent, transient, shifting, historically varying, unfalsifiable, hostile, patriarchal, etc.)

I kinda wish I could have a brain accident so I could be as clever as D’Souza and write books like “Gosh, Why Christianity is so Freakin’ Wonderful” and get paid lots of money to be a “science expert friendly to religion”.

7. TomS says

A comment on Dawkin’s formula:

Power of a theory = Number of things it explains/Number of things it needs to assume

Unfortunately, for creationism, the Number of things it explains is infinite. All things, even those things which are not true, are explained by creationism.

Of course, you might say that the denominator is also infinite, for the Number of things that it needs to assume is equal to the number of things explained.

But I have two possible rebuttals:

1. The number of things that creationism needs to assume is one. Namely, that an omnipotent being can do anything.

2. The number of things that intelligent design, in particular, needs to assume is zero. For intelligent design does not say anything at all.

8. α = 1/c * (1/τ – 1/τ0)

9. says

I’d have to go with the Nonlinear Schrödinger equation. I work in the field of nonlinear optics, and it one of the most frequently used equations. In my case, it is hugely generalised, containing 10th order derivatives, and a whole host of other nasty things.

Offtopic:

MH, I have had the misfortune of reading that ridiculous article by Dinesh D’Souza. Even if Kant is right about the limits to human experience (which is a very big if), what on earth has that got to do with god? It’s just the god-of-the-gaps argument combined with name-dropping a famous philosopher.

Even if “god” did lurk beyond understanding, why should such an entity be remotely like D’Souza’s Santa Claus style god (which leaves eternal life in his stocking).

10. gray lensman says

Down in Texas during my childhood this passed for science:

1. Pie are square.

2. No, cornbread are square; pie are round.

11. says

I’d have to go with the Nonlinear Schrödinger equation. I work in the field of nonlinear optics, and it one of the most frequently used equations. In my case, it is hugely generalised, containing 10th order derivatives, and a whole host of other nasty things.

Offtopic:

MH, I have had the misfortune of reading that ridiculous article by Dinesh D’Souza. Even if Kant is right about the limits to human experience (which is a very big if), what on earth has that got to do with god? It’s just the god-of-the-gaps argument combined with name-dropping a famous philosopher.

Even if “god” did lurk beyond understanding, why should such an entity be remotely like D’Souza’s Santa Claus style god (which leaves eternal life in his stocking).

12. says

I’d have to go with the Nonlinear Schrödinger equation. I work in the field of nonlinear optics, and it one of the most frequently used equations. In my case, it is hugely generalised, containing 10th order derivatives, and a whole host of other nasty things.

Offtopic:

MH, I have had the misfortune of reading that ridiculous article by Dinesh D’Souza. Even if Kant is right about the limits to human experience (which is a very big if), what on earth has that got to do with god? It’s just the god-of-the-gaps argument combined with name-dropping a famous philosopher.

Even if “god” did lurk beyond understanding, why should such an entity be remotely like D’Souza’s Santa Claus style god (which leaves eternal life in his stocking).

13. Oh, and mine is the equation for the absorption coefficient of an absorbing species in Cavity Ringdown spectroscopy. τ and τ0 are the times it takes light to decay out of a cavity in the presence and absence of the absorber, respectively.

14. says

Damn it, triple post.

15. says

RickD – I think you need to invert your cabin first. At the moment, it’ll just be square.

Bob

16. says

Actually, Geno Segre (physicist) and Marvin Minsky (comp sci) did have the same equation.

I still like Newton’s Law of Motion – it works in so many contexts.

17. Janicot says

Not entirely formulae but:

1. Never believe anything, but question only what is worth questioning. — George Polya

2. The secret of success is sincerity. Once you can fake that you’ve got it made. — Jean Giradoux

18. Christian Burnham says

0!/0^0=1

19. says

Dammit, Bob, you’re right. (Don’t tell Rutgers or they’ll revoke my Ph.D.!)

20. Janicot says

Oh yeah, one more:

1 nanocentury = PI seconds

21. Colin M says

gaaaaaaaaaaah! Dawkins used Comic Sans! My eyes!

22. Rey Fox says

Life is my formula, babe.

23. says

@ #7

Two things,

1) I think the implication was _true_ things the theory could explain, but thats open for interpretation.
2) Creationism actually must assume a few more things than “An Omnipotent Being can do Anything” It must assume the existence of such a being, which is an axiom with enough complexity to rival Euclid’s Parallel Postulate. It must assume that this being is inclined to do things to our world in particular, etc.

so maybe a better equation would be:

Power(T) = [(# of true things the theory predicts) – (# of false things the theory predicts)]/[(Sum of the Relative Powers of each axiom) * (number of axioms)]

Where T is a theory.

so Darwinism assumes a pair of relatively simple premises: “Genes Exist” and “Genes can Mutate”, and provides “Millions of individual species” with no apparent contradictions, so we have F = number of failures, contradictions, etc. 10^6 number = number of true things
A1 and A2 the relative powers of the axioms, we then have:

(10^6 – F)/2(A1+A2)

since the powers of the axioms are fairly small, 2(A1 + A2) is fairly small, and assuming its less than 10^6 – F, then the power is fairly big.

Creationism, on the otherhand, provides a fairly small number of true predictions, in contrast to the number of false predictions. I’m, however, somewhat biased against it, because I think Creationists are dumb. So I won’t go that route.

anywho, my mantra/algorithm/formula for life:

A paraphrase from M. Ghandi
“Let them ignore you, Let them laugh at you, Let them Hate you, You will win.”

That always reminds me of Cantor.

and my Formula for life: Aleph_Null = 2^(Aleph_Null) — The infinite is always attainable.

24. says

Click my name below to see my post regarding the God function. It’s “mathy.”

25. mxracer652 says

Christian:
That’s one of my favorites, along with infinite geometric series for showing just how ill prepared the mind is at dealing with very large numbers.

0.999… = 1

26. Tatarize says

I kind of like Dawkins’ formula about the use of a theory as to what it explains divided by what it assumes.

God, Everything / (Everything + God)
Evolution, Billions of well adapted species / Genes Exist

27. CL says

Well, back when I was an undergraduate studying physics, it was probably the formula for determining the temperature at which a Bose-Einstein condensate would form. But I’m in law school now, and I’m not doing quantum optics anymore.

So now, my formula is one that comes from my lawyer grandfather:

“If you can’t win on the facts, argue the law. If you can’t win on the law, argue the facts. And if you can’t win on the facts or the law, argue like hell.”

28. David Harmon says

My cynical formula is Sturgeon’s Law: “90% of everything is crap”.

My addendum: The business of life is digging out the other 10%.

29. says

I’ve always been partial to the Drake equation, even given all the problems it has.

30. says

Unfortunately, for creationism, the Number of things it explains is infinite. All things, even those things which are not true, are explained by creationism.

Not true. Creationism doesn’t explain anything. “Goddidit” is not an explantaion, it is the semantic equivalent of “Because I said so” or “Go ask your mother.” It may stop the question, but doesn’t answer it.

31. Sven DiMilo says

To equationize Robert MacArthur:

(The only rules of scientific method) = (honest observations) + (accurate logic)

32. cm says

Looking over a number of those equations created a strange and growing–and surely undeserved–contempt for the respondents.

For me, the proliferation of symbols (the various letters and such in the formulae) disconnected or poorly connected from understanding is very irritating somehow. I guess enough of the respondents failed to explain the formulae such to trigger this reaction. Also, there seemed to be a number of cases of what might be called “symbol abuse” in which the respondent piled up big heaps of symbols or words in the manner of someone waking up in the middle of the night and writing down something “really important”, only to wake up the next day and wonder what the gibberish on the page was.

I think it also speaks to a suspicion I have, which I will put in a formula for the heck of it:

C is far less than C1

C = The minimum necessary complexity of the descripitions of our present understanding of the world.

C1 = The complexity actually used (on average) in our descripitions of our present understanding of the world.

My point being: we overcomplexify things. Some of that is justified as we struggle toward clarity, but much of it is egregious, as we bluster and flaunt our symbology.

33. Nathaniel says

Ok, not exactly MINE, but the one closest to my life:

P(nu_x->nu_y) = sin^2(2 theta) sin^2( 1.27 Dm^2 L/E)

… the oscillation mixing equation for two neutrino species.

34. SteveM says

Being an EE, I’m kind of partial to Maxwell’s Eqn’s.

35. gobsmack says

eπi = -1

36. Bill Dauphin says

The 90/90 Law of Project Scheduling:

The first 90% of the work required to complete a project will consume the first 90% of the schedule time available; the remaining 10% of the work will consume the other 90% of the schedule.

37. [elephant] X [grape] = elephant grape sin(theta)

[elephant] X [mountain climber] = undefined because you can’t take the cross product of a scaler.

38. TomS says

I like that.

39. Joshua Schraiber says

To cm:

I disagree, actually… I think that the symbols are probably the most clear way to discuss a lot of these things. Consider if we had to do math in language… it would cause a lot of headaches and not be nearly as easy to verify. Of course, though, you bring up the valid point that there are a lot of random symbols to know: the thing is, you would have to define words or symbols at some point, you know? You’ll have to add SOME definitions somewhere!

My own contribution: the simple differential equation kx=ma. It describes the motion of a harmonic oscillator, and includes some of the most fundamental ideas in differential equations and can also lead to more complex things, like the wave equation (d^2y/dt^2 + (v^2)d^2y/dx^2 = 0, iirc)

40. moioci says

Colin M @ #21:

I think it’s actually Chalkboard, Apple’s less ugly counterpart to Comic Sans, and appropriate for this use.

41. Flex says

Two algorithms are important in my life:

1. Ask others (and yourself) “why?” – If you don’t know the reason for something, ask about it, figure it out yourself, or admit you don’t know.

2. Ask yourself (and others) “why not?” – If you are asked to do something, determine the reasons you shouldn’t do it. If there are good reasons to not do the task, then don’t do it. If there are no good reasons (and fear alone is never a good reason) then do the task.

Note: your reasons may not appear good to someone else, but so long as you know what the reasons are and can explain them to others this doesn’t matter. Saying you can’t cook dinner because you are watching a football game may seem a perfectly reasonable reason to you, but others may not see that reason in the same light.

Reasons which cannot be articulated (e.g. “I don’t feel like it.”) are never good reasons.

I’ve always felt that everyone could get along so much better if we admit the limits of our knowledge and help those who request it to the limits of our ability.

42. Dan says

Holy SHIT, I’m going to be reading all of this (and the web sites of those featured) for the rest of the day, and possibly tomorrow as well.

Thanks a lot.

43. says

PZ, are you saying that phenotype is the Fourier transform of genotype? Isn’t that a little too linear? =)

44. Hank Fox says

Jeez, I have to pick ONE?? I have too many of ’em to count:

1) Never lie to a child or a dog for any reason.

2) When two people meet in a narrow corridor or doorway, the one with the heavier load has the right of way.

3) After you lose half your hair, the pressure’s off.

4) There is a date in the near future at which you are already dead. Think and build and love in the knowledge that your supply of Todays is sharply limited, and learn something new every day of your life.

5) The real world is the real world. There are no secrets, lies or tricks built into it, and no part of it is deliberately hidden. Potentially, anyone can understand it. Potentially, every part of it is subject to eventual understanding. Every preacher or government official who tells you that there IS secret knowledge, which you have no hope of obtaining and which only they can understand and interpret for you, already thinks of you as their slave.

6) Tattoos are for people who want to be noticed, without going to the trouble of actually becoming interesting.

7) The instant you think something nice about someone, give it to them as a gift. If the girl checking your groceries has pretty hair or a beautiful smile, tell her … and then walk away. Repeat with everybody around you.

7) Etc., and very much so.

45. Hank Roberts says

“Once is happenstance, twice is coincidence, three times is enemy action.” — attributed variously

“Life is like stepping onto a boat that is about to sail out to sea and sink.” — Shunryu Suzuki

“… No matter what you might do
There’s always someone out there cooler than you …”
— Ben Folds

“Airspeed, altitude, ideas — while you have two out of three you’ll live.” — my hang gliding instructor

46. josh says

I would say the ratio of explanations to assumptions in Intelligent Design, or most any religious explanation, is nealry one, and this is the worst score you can get. That is, anything can be “explained” simply by saying it happened. If my pencil falls off my desk I can say, “it fell because pencils fall of desks which are mine on Thursday afternoons,” and there is nothing formally wrong with that explanation, it is just useless. The power of a good explanation is that it explains many things with relatively few postulates/principles. So for instance, Newtons Laws (lets ignore quantum and relativity) purport to explain the motion of every massive object in the universe. There are only three (really only two) laws. Newton’s laws are pretty good. God-riddled explanations, on the other hand, can apply to anything so the numerator is infinite, but since no such explanations follow logically from the concept of God you’ve got a seperate goddidit for every phenomenon you wish to explain. So the ratio goes to one. God does no better than treating every occurence as an independent event unconnected to anything else.

47. Eric says

No One of Consequence @37,
I believe that should be (mosquito)X(mountain climber)= undefined because you can’t cross a vector with a scaler.

48. says

E=mc

It’s a postmodernist version of Einstein’s famous equation; an improvement that refrains from that other equations’ shameful privileging of the speed of light over mass and energy.

49. John Bode says

My algorithm for scheduling:

2. Multiply the total estimate by π

An irrational schedule calls for an irrational number.

50. roystgnr says

lim u_N -> u

There are countless sets of equations whose solution you might call “u”. Every researcher might have a different way to define a sequence of approximate solutions u_N. The norm you measure convergence in may change, and the rate of convergence may depend on the approximation type and norm. Still, I think the above equation manages to summarize all of applied mathematics:

“We may not be able to find an exact answer, but tell us how close you need to get and give us a big enough computer and we’ll find an acceptable answer eventually.”

51. David Marjanović, OM says

Work expands so as to fill the time available for its completion.

52. David Marjanović, OM says

Work expands so as to fill the time available for its completion.

53. Chris R. says

Mine would have to be:

dU=δQ-δW

54. Susan B. says

Merge sort is my algorithm of choice. Yes, I know Quicksort is faster on average, but I just find merge sort easiest to implement.

55. Crudely Wrott says

From my father, concerning livestock: You must be the boss.

From my step-father, concerning people: Be wise, kind, and a little bit blind.

From my first love: Keep a socket on ya. That’s keepin’ a lotta light.

From me: Life without music would be pretty hard to dance to.

From me: What you see is indeed what is real, assuming certain corrections for instrument error and line noise.

From Kurt Vonnegut: And so it goes, Billy Pilgrim.

From his son: The price of eternal vigilance is insanity.

Me, finally: Eat when you’re hungry, sleep when you’re tired and scratch where it itches.

56. says

Joe Fredette: Assuming an adjunctive logic, how do you count axioms of a theory, since in principle any theory can be stated in one axiom by suitable conjunctions? (Well, maybe that sometimes would require an infinitary conjunction, but …)

gobsmack: That can be rendered funkier – eπi+1 = 0