In his short story The Library of Babel , Jorge Luis Borges writes of a vast library. It contains every possible book with precisely 410 pages each with 40 lines of 80 symbols from a list of 25 letters and punctuation.
This is a staggering number of books: 2 5 1 3 1 2 0 0 0
Now suppose they needed to be reordered (don't ask how), in how many orders could the librarian arrange them?
Don't enter the number of possible orders.
Don't enter the number of digits in the number of orders.
Enter the number of digits of the number of digits of the number of orders.
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
The story The LIbrary of Babel is a fascinating concept, the idea of a "total library", thanks for bringing to my attention. In theory, "all information is there", but being able to distinguish between useful and useless information itself becomes an information gathering task---which requires getting another library! And unless that additional library is not yet another total library, that becomes useless as well. Hmm...
Log in to reply
There are so many books that there can't even be a book that can serve as an index. The index could span many many volumes, but there would be many many more volumes containing false indices. Of course, the vast majority of the books are complete gibberish.
Log in to reply
It's been said that in the infinite decimal expansion of π all of Shakespeare's works can be found. But is it really meaningful to say that there's "inexhaustible information" in that decimal expansion? Without selection, there is really no "information". In creating his works, Shakespeare chose a very few of all books he could have theoretically written, and it's by his choice that he has become famous. So, any index that would at least point out useful books is itself a much more valuable source of information that the total library. Indeed, just listing the addresses of such useful books is the virtually the same as writing such useful books.
Log in to reply
@Michael Mendrin – There's problems with that infinity though. How will you find the part you want? The Bible might repeat itself three times before the beginning of Romeo and Juliette.
In the Library, you could have a shelving system and the books arranged in some order. Book 1: all spaces. Book 2: 1311999 spaces with the letter a at the bottom of the last page, ... Book {something}: "two households, both alike in dignity, in fair brilliantania..." close enough...
Log in to reply
@Jeremy Galvagni – ...close enough, now where's the book with fair Verona in it?
If every single book in the total library has an unique address that may or may not be included in some marvelous "index of useful books", then the contents of any book is that book's unique address. If there are 2 5 1 3 1 2 0 0 0 books in the library, then that means there are 2 5 1 3 1 2 0 0 0 addresses from which the index can selectively include. So, how can this be economized?
Any such "index of useful books" is already itself a library of books it has deemed to be useful. In other words, a conventional library.
The number of books can reordered in 2 5 1 3 1 2 0 0 0 ! ways.
For very large n , Stirling's approximation is ln n ! ≈ n ln n − n . Using the change of base formula ln x = lo g e lo g x , the equation converts to lo g e lo g n ! ≈ n lo g e lo g n − n or lo g n ! ≈ n ( lo g n − lo g e ) . This means the approximate number of digits in n ! is n ( lo g n − lo g e ) , and the number of digits of the number of digits in n ! is ⌈ lo g ( n ( lo g n − lo g e ) ) ⌉ = ⌈ lo g n + lo g ( lo g n − lo g e ) ⌉ .
Therefore, the number of digits of the number of digits in 2 5 1 3 1 2 0 0 0 ! is ⌈ lo g 2 5 1 3 1 2 0 0 0 + lo g ( lo g 2 5 1 3 1 2 0 0 0 − lo g e ) ⌉ = ⌈ 1 3 1 2 0 0 0 lo g 2 5 + lo g ( 1 3 1 2 0 0 0 lo g 2 5 − lo g e ) ⌉ = 1 8 3 4 1 0 4 .
Problem Loading...
Note Loading...
Set Loading...
Relevant wiki: Stirling's Formula
The number we seek is lo g ( lo g ( ( 2 5 1 3 1 2 0 0 0 ! ) )
We need to make use of Stirling's Approximation and since the number is so huge, any of the approximations will work. I will use n ! ∼ 2 π n ( e n ) n because it doesn't involve logarithms and I plan to use common (base 10) logs.
First rewrite 2 5 1 3 1 2 0 0 0 as 1 0 lo g 2 5 ⋅ 1 3 1 2 0 0 0 = 1 0 1 8 3 4 0 9 7 . 2 9 1 and just call this exponent A
The number of orderings is 2 π 1 0 A ( e 1 0 A ) 1 0 A = 2 π ⋅ 1 0 2 A ( 1 0 A − lo g e ) 1 0 A
Taking the first common logarithm and separating by the product and power rules results in
lo g 2 π + 2 A + 1 0 A ( A − lo g e )
It's not possible to take the logarithm of this sum very easily, but note the first term is minuscule and even the second is tiny compared to the third. They can be ignored. Even the lo g e can be ignored because it is also tiny compared to A . The second logarithm gives the answer.
A + lo g A ≈ 1 8 3 4 1 0 3 . 5 5 5 which means the number we seek is 1 8 3 4 1 0 4