InĀ What If?: Serious Scientific Answers To Absurd Hypothetical Questions, Randall Munroe, creator of the popular web comicĀ xkcd,Ā provides pages of in-depth responses to questions you probably never thought to ask.
Q. How many unique English tweets are possible? How long would it take for the population of the world to read them all out loud? ā Eric H, Hopatcong, NJ
A. Tweets are 140 characters long. There are 26 letters in English ā 27 if you include spaces. Using that alphabet, there are 27^140 = 10^200 possible strings.
But Twitter doesnāt limit you to those characters. You have all of Unicode to play with, which has room for over a million different characters. The way Twitter counts Unicode characters is complicated, but the number of possible strings could be as high as 10^800.
Of course, almost all of them would be meaningless jumbles of characters from a dozen different languages. Even if youāre limited to the 26 English letters, the strings would be full of meaningless jumbles like āptikobj.ā Ericās question was about tweets that actually say something in English. How many of those are possible?
This is a tough question. Your first impulse might be to allow only English words. Then you could further restrict it to grammatically valid sentences.Ā
But it gets tricky. For example, āHi, Iām Mxyztplkā is a grammatically valid sentence if your name happens to beĀ Mxyztplk. (Come to think of it, itās just as grammatically valid if youāre lying.)
Clearly, it doesnāt make sense to count every string that starts with āHi, Iāmā¦ā as a separate sentence. To a normal English speaker, āHi, IāmĀ Mxyztplkā is basically indistinguishable from āHi, Iām Mxzkqklt,ā and shouldnāt both count. But āHi, Iām xPoKeFaNxā is definitely recognizably different from the first two, even though āxPoKeFaNxā isnāt an English word by any stretch of the imagination.
Our way of measuring distinctiveness seems to be falling apart. Fortunately, thereās a better approach.Ā
Letās imagine a language that has only two valid sentences, and every tweet must be one of the two sentences. They are:
āThereās a horse in aisle five.ā
āMy house is full of traps.ā
Twitter would look like this:
What If?
The messages are relatively long, but thereās not a lot of information in each one ā all they tell you is whether the person decided to send the trap message or the horse message. Itās effectively a 1 or a 0. Although there are a lot of letters, for a reader who knows the pattern of the language, each tweet carries only oneĀ bit of information per sentence.Ā
This example hints at a very deep idea, which is that information is fundamentally tied to the recipientās uncertainty about the messgeās content and his or her ability to predict it in advance.
Claude Shannon ā who almost singlehandedly invented modern information theory ā had a clever method for measuring the information content of a language. He showed groups of people samples of typical written English that were cut off at a random point, then asked them to guess which letter came next.
Based on the rates of correct guesses ā and rigorous mathematical analysis ā Shannon Ā Ā Ā Ā Ā determined that the information content of typical written english was 1.0 to 1.2 bits per letter. This means that a good compression algorithm should be able to compress ASCII English text ā which is 8 bits per letter ā to about 1/8th of its original size. Indeed, if you use a good file compressor on a .txt ebook, thatās about what youāll find.
If a piece of text containsĀ n bits of information, in a sense it means that there are 2^n different messages it can convey. Thereās a bit of mathematical juggling here (involving, among other things, the length of the message and something called āunicity deistanceā), but the bottom line is that it suggests there are on the order of about 2^140 x 1.1 = 2 x 10^46 meaningfully different English tweets, rather than 10^200 or 10^800.
Now, how long would it take to read them all out?Ā
Reading that many tweets would take a person nearly 10^47 seconds. Itās such a staggeringly large number of tweets that it hardly matters whether itās one person reading or a billion ā they wonāt be able to make a meaningful dent in the list in the lifetime of Earth.
Instead, letās think back to that bird sharpening its beak on the mountaintop. Suppose that the bird scrapes off a tiny bit of rock from the mountain when it visits every thousand years, and it carries away those few dozen dust particles when it leaves. (A normal bird would probablyĀ deposit more beak material on the mountaintop than it would wear away, but virtually nothing else about this scenario is normal either, so weāll just go with it).
Letās say you read tweets aloud for 16 hours a day, every day. And behind you, every thousand years, the bird arrives and scrapes off a few invisible specks of dust from the top of the hundred-mile mountain with its beak.
When the mountain is worn flat to the ground, thatās the first day of eternity.Ā
The mountain reappears and the cycle starts again for another eternal day: 365 eternal days ā each one 10^32 years long ā makes an eternal year.
A hundred eternal years, in which the bird grinds away 36,500 mountains, make an eternal century.
But a century isnāt enough. Nor a millennium.Ā
Reading all the tweets takes youĀ ten thousand eternal years.
What If?Thatās enough time to watch all of human history unfold, from the invention of writing to the present, with each day lasting as long as it takes for the bird to wear down a mountain.
While 140 characters may not seem like a lot, we willĀ never run out of things to say.
Ā
Excerpt fromĀ WHAT IF? by Randall Munroe. Copyright 2014 by xkcd. Used by permission of Houghton Mifflin Harcourt Publishing Company. All rights reserved.
