This topic contains 51 replies, has 2 voices, and was last updated by Edward Edrewardloar 3 weeks, 4 days ago.
February 24, 2017 at 1:26 am #12999
Oh and messiah, how did you go long enough without breathing to type your post? You clearly don’t have the mental capacity to use both language and maintain autonomic functions at the same time.
Or in monosyllsbles for your benefit: bite me.February 24, 2017 at 2:47 am #13001
Liquid, I can select the blank spaces just fine. Are you sure it’s not your computer?February 24, 2017 at 11:15 am #13006
u must b a fagFebruary 26, 2017 at 6:58 am #13032
Holy crap people in this thread are dumb. Going to answer a few questions people might have about the Library of Babel and how it may be coded.
1: You can import dictionaries into your program, thus you be able to search for a page filled with english words.
2: No, we do not know for sure if its just making up a random page with your name on it, but its probably not. You can easily create something in python that searches for what you enter (a string) in another string.February 26, 2017 at 10:54 pm #13049
look m8 the logic behind the page generation is that the input accepts some numerical shit (hexagon, wall, shelf, volume, page), and with a clever algorithm – unclearly described in https://libraryofbabel.info/theory4.html – will ‘translate’ it to a page of text u see afterwards. do u get it?February 26, 2017 at 10:55 pm #13050
I mean alphanumerical of course. sorry for the typo.March 1, 2017 at 9:07 am #13134
I do think the library is “fake” but not for the reasons mentioned above. What the search function does is generate a random page with your text on it. It’s not like it actually searches a practically infinite amount of strings; that would be stupid and pointless. For all intents and purposes the search function creates the page in the process. The contents of the page are encoded in the hexagon string so the only way to find a specific page is to know its exact contents (which is not really interesting).
So why I think this is “fake” is that for example these two pages are identical:
I went to browse and picked the first page, searched it, and it returned the same page in different coordinates. It kinda ruins the magic when there are countless copies of the same page. This means that the system can’t locate any single page, it just creates a new page that contains the text you want.
I know programming is hard, but there must be a way to create an actual Library of Babel. This site toys with the idea but fails to implement it properly.March 1, 2017 at 9:18 am #13135
>For all intents and purposes the search function creates the page in the process.
EXA-FUCKING-CTLY! this is the main point of this site. a translating algorithm works in the background, for some input it generates an output, very similarly to the cryptographic functions.
>This site toys with the idea but fails to implement it properly.
fuk uMarch 1, 2017 at 7:27 pm #13160
Renato J. Couré
“In the vast Library there are no two identical books”, the Borges’ story states. It does not say that are not two identical pages.March 1, 2017 at 8:01 pm #13161
Renato J. Couré
Edit: It does not say that there are not two identical pages. (I’m sorry, not an English speaker over here)March 1, 2017 at 8:12 pm #13162
Random words doesn’t need a search that is English words. All it needs is a string of characters, which it then finds somewhere that it coincides with a page of entirely words. These pages, while extremely rare to find randomly, are very easy to locate for an algorithm.
Something tells me you haven’t read the theory, or how this website was created. It’s run by a random seed that is put through an algorithm. That way, the algorithm can be reverse through a search query to find the different places that your string would appear, as well as where it would appear with other things (English words for example).March 2, 2017 at 11:13 am #13189
“In the vast Library there are no two identical books”, the Borges’ story states. It does not say that are not two identical pages.
That’s a good point. But I don’t see how that could be guaranteed either because in the end everything is randomly generated. Seems to me that it’s actually guaranteed that there are also duplicate books.
But yeah, I didn’t consider that when I made my initial criticism, and only now did I realize that the search apparently returns all the locations the exact page is found in, “~10^29 possible exact matches” it says, which is weird because wouldn’t there actually be 410*29^3200^409 places a page can appear in (when every possible book exists and there are no duplicate books)?
Also, the theory page only talks about producing all the unique pages (29^3200), and not producing all unique books (that’s 29^3200^410, an even more insane number).
But maybe I’m being too harsh. The system is kinda neat. I don’t think I could do better.April 2, 2017 at 12:59 am #14507
Either I still don’t get the concept or it seems fake/useless.
I searched a long sentence, copied it and searched it several times again and again.
Every single time the result pointed me different places. I understand that my sentence may appear in various contexts. But why does the same search generate different results?
Also, it seems to find even very long sequences of words when searched, but browsing through multiple existing pages almost never yields even a short sequence of 3 or 4 real words.
Can someone explain the first observation and did anybody succeed to find something new and meaningful just browsing?April 4, 2017 at 3:54 am #14569
People who don’t understand the basics of what’s involved have no business crying foul. If you’ve read the Theory page and you don’t even have passing familiarity with the mentioned maths or computer science, then it’s a wonder to me that you value your opinion so highly on the matter.
Consider a much simpler implementation: An algorithm that systematically produces every 3200-character permutation of the alphabet. This starts with the very first page, which is filled with 3200 ‘a’ characters in a row. The next page is 3199 ‘a’ characters followed by 1 ‘b’ character. The third page is 3199 ‘a’ characters followed by 1 ‘c’ character. This continues on for a while. The twenty-sixth page, which is 3199 ‘a’ characters followed by 1 ‘z’ character.
The twenty-seventh page in this series is 3198 ‘a’ characters, 1 ‘b’ character, and lastly 1 ‘a’ character. The pattern continues like this. The fifty-second page is 3198 ‘a’ characters, 1 ‘b’ character, and lastly 1 ‘z’ character. The fifty-third page is 3198 ‘a’ characters, then 1 ‘c’ character, and lastly an ‘a’ character.
This pattern continues ad nauseum, until the 26^3200 iteration, which is a page filled with 3200 ‘z’ characters.
This algorithm is perfectly straightforward and predictable. Based on one page, you know exactly what the next page will be. It will be the exact same, except with the pattern ‘incremented’ by one.
The creator of this website did not want the library to be a systematic creation of every page in order; he wanted it to appear as gibberish- a “random” assortment of 3200 characters- from one page to the next to add to evoke the mystique of the library from the short story. So, he used tools that are utilized in cryptography to create a “scrambled” order of all the 3200 character permutations of the alphabetic characters, the space character, the comma character, and the period character.
Because there are many, many, many, many, many, many, many, MANY more permutations of 3200 characters that are utter nonsense than those that contain even a single recognizable piece of information SOMEWHERE within the entire 3200 character sequence on the page, this means you will, in all likelihood, never find one by pure chance.
The reason you can search for a given string that is fewer than 3200 characters and have different pages returned each time is because the algorithm actually requires a search that is exactly 3200 characters long (given the above legal characters) to serve as input for the inverse of the algorithm in order to pop out the corresponding reference hex. Your search may only be for “my dog is the best dog in the whole world”, but the input must still be 3200 characters long. So, as another poster previously mentioned, the way this is probably done is that your short string has its position in the 3200 character sequence pseudorandomly determined, and then the rest of the 3200 character sequence is filled up pseudorandomly with legal characters. From here, the determined 3200 character sequence (which contains the target string) is converted via the inverse algorithm to pop out the appropriate reference number for that page.
The reason inputting this exact reference number will always return the same page is because the algorithm is, at its core, deterministic. For a simplified example, with y = 8 * x, y will always equal 64 when x equals 8 (and vice versa) no matter how many times you perform the calculation. It’s just how maths work, and you should be thankful for that because otherwise our universe would be nonsensical.
The reason inputting a sequence exactly 3200 characters long will not always return the same reference number has to do with the very large number of duplicate pages that exists for each 3200 character permutation. I’d guess that there is some time-based seed that determines which reference numbers you will be directed to each time you perform the search. Why? Either it’s because of the way the algorithm was designed, or because the creator didn’t want identical searches producing the same lowest value reference numbers that are applicable to the search each time that a given string is searched for (thus obscuring the vast nature of the number of different possible permutations). Or perhaps the algorithm was designed that way with the latter in mind – regardless, it is how it is.
The reason each possible permutation has multiple matching reference numbers has to do with the elegance of the algorithm. I have no doubt that the current implementation exceeds my capabilities, but my own ideal for the algorithm would yield each permutation exactly once. (Given the task, I imagine such an execution would require a genius among geniuses to work it out, but I couldn’t say for sure as I’ve never tried in the slightest to produce such a thing.)
The server does not store prior searches, because it has no need to do so. It’s as simple as inputting the reference number into the algorithm, which dutifully spits out the same 3200 character combination every time, as consistently as 8 * 8 = 64.
If you do not understand what I am talking about in this post and you still have doubts, then I encourage you to learn more on your own of the subjects until you understand why this concept’s execution is plausible.
If you’d like an example of strangeness arising from the vast number of permutations systematically produced by some algorithm, see Numberphile’s explanation of “Tupper’s Self-referential Formula”: https://youtu.be/_s5RFgd59ao
If you doubt the veracity of Tupper’s Self-referential Formula, then I encourage you to examine why you weight your gut feeling over the expertise of academics who have forgotten more mathematical knowledge before breakfast than most people will ever attain in their lives – but I sincerely doubt more than a handful of laypeople would genuinely trust themselves over experts in the matter of maths.
If you skipped reading this post because you don’t wish to try to learn more and challenge your doubt, then there is no place for you here until you are willing to try.
Best regards, and thank you for this site you’ve made. Hopefully, I did alright with my explanations.April 4, 2017 at 4:07 am #14570
To SeriosDoubter, with regards to the library being “useless”:
For one part, it is as useless as literature, a painting, a poem, music. Personally, I feel there is a beauty to it, and so I am glad for its existence.
If you want a more “practical” use than art or philosophy or any such thing, however, then that’s a matter best left to the experts. Much work done in mathematics appears useless, but is sometimes later discovered to be useful to some other function that can be used in the “real” world (e.g., Fibonacci numbers and biology).
Does this site have beyond artistic or philosophical merit? I couldn’t say for sure- perhaps time will tell- but I can say that the subjective value that exists for the curiosity of it and for the pleasure Jonathan had in its creation is enough to justify its existence.
All that being said, it might not appeal to you, and that’s all right.