This topic contains 56 replies, has 2 voices, and was last updated by Anon 2 weeks, 6 days ago.
October 7, 2017 at 12:53 am #22135
I’m a bit late, but…
Explain to him that: The information behind the location (which is generally very complex, with massive hex IDs) is put through an algorithm, which generates text based on that location. Because the algorithm is the same, the result is always the same for the same location (2 * 2 will always be 4, 5+4 will always be 9).
This allows any random order of characters to be generated via this algorithm. Because of this immense variety, you’ll never find much of anything by wandering about, since the chances of finding something coherent are astronomically low. However, because it’s an algorithm, you can reverse engineer it to find locations (I can find “9” with the locations “3 * 3”, “6 + 3”, “7 + 2”, “27 / 3”, etc.).October 8, 2017 at 6:34 pm #22158
Are you going to share the source code of the library?October 10, 2017 at 1:49 am #22223
I have an idea and I want to share it with you all.
Let’s say you want to generate all the possible numbers that can exist that meet the next criteria: 1.- use base 10, 2.- max 5 digits.
So, you will have all the possible numbers from “0” to “99999”, agree? Ok, if I search for the number ” 768″ for example it will be repeated a lot of times cause it can be arranged in different ways like: “00768, 07680, 07681” etc, BUT, if I want to find the number “55555” for example, there can’t be more than one? Haha. The library use base 29(all the letters plus the space, coma, and period), 3200 is the max of digits, if I write 3200 times the letter “a” like “aaaaaaaaaaaaaa…….” Why it would give me more than one possible result?October 10, 2017 at 6:50 am #22227
Great question. 3200 is the max number of characters per page. The max number of characters per book is 1,312,000 (3200 * 410). Page 1 could be “aaaaaa…” and page 2 could be “aaaa…”, “bbbb…”, “cccc…”, etc.October 15, 2017 at 3:27 am #22386
The idea of the library is fascinating, but without the source code, it’s a dichotomy. There are far too many books to keep them all pregenerated on a hard drive. So either the code generates the proper location and and book, or (and skepticism refuses to be silent), it would be very easy to insert the search query into a random book. This is why it’s hard to deal with gigantic numbers, larger than the atoms in our universe. The closest thing we can get is a simulation. So the question becomes, ‘is it an authentic simulation?’ once again, the code is the only proper answer to the question, but I’m skeptical because of how quickly all the books can generate. Even a simple sentence has millions of variations. I’m also not familiar with the creator of the website, so I don’t know if he would be more inclined to create a gimmick that just injects the search into a random book and makes it look nice, for the feeling of authenticity without the hassle of hard drive space.
I think the better question is whether or not the experience of the library is authentic. The story even says that something so massive and complex, maybe not infinite, but close enough for our purposes, could only have been created by a god. Would it actually be scandalous for the site to simply inject the search into characters? I don’t think so. Like Minecraft, that’s the only way so much data can be managed.
But that also raises the question, do all the things mentioned in the book, a description of God, autobiographies of angels, the stories of our lives, actually exist in this library? Technically, in Minecraft, the areas you visit are generated when you enter them, and they are saved in the files. BUT they only exist after you cause them to.
SO the distinction between the website and the real library is that these things can only exist after you search for them. Wheras in the short story, that information is already there waiting to be found.
I hope that technology advances to the point where we could truly generate the full library, in all of it’s gigantism. From there, searching for full sentences would be an easy way to sort the gibberish from the relevant, and so make the information more accessible. In the book, the librarians were hard pressed in their task because their physicality limited them to looking at one book at a time, in one place at a time. Technology circumvents this issue.
Thoughts on a digitized and indexable library of babel?October 15, 2017 at 4:42 am #22388
Many of your concerns have already been answered numerous times. You need only look for a few minutes, but for convenience I will address them all here.
Yes, the Library is not pre-generated and stored on hard drives. As that would be physically impossible, it is instead simulated via random generation. This causes many people to question its authenticity.
Rest assured, the Library is authentic. They way it works is like any other pseudo random number generator (PRNG). If you know how that works, then you know the concept of a “seed”. If not, in a nutshell: Computers cannot naturally produce random numbers. They instead use a mathematical equation which you provide a beginning number to (a seed), and then it outputs another number. That new number is used as the next seed and the cycle continues. Any seed that you provide will result in the exact same pattern of “random” numbers – which is why Minecraft will generate the same world given a seed. Typically (if you were to click “New World” without a seed), a PRNG will use the current UNIX time, which is the number of seconds that have elapsed since January 1, 1970, down to the microsecond – it’s unlikely that 2 people making a new world at the same time will get the same seed that way.
The difference with the Library though, is that Jonathan created two things:
1) A PRNG that will generate all unique books without overlap (under the limit of the hex’s length)
2) An inverse PRNG. You’re able to provide a page to it, and it will find many books where that page is located. If you are searching for a small piece of text (under 3200 characters), the website will have to generate the rest of it.
What this means is that, yes, all pages “exist”, but not until you discover them. That could be through the Random button or searching for it yourself, but it will always be there in perpetuity. Nobody is trying to deceive here.October 28, 2017 at 10:48 am #22675
the truth is out there
it was the best of times
it was the blurst of times
we (ed and ais and I) think that this webiste is for real
emma does not
oscar is tired of us and likes the floorDecember 12, 2017 at 12:31 pm #23608
in response to your answer of Alejandro’s question, can you explain why there are duplicates of pages in light of this fact taken from the About section of the library:
If completed, it would contain every possible combination of 1,312,000 characters
At present it contains all possible pages of 3200 characters
it seems that if completed, you would expect to find duplicates of pages but not books. However at present, you would never find a duplicate of a page. Or has the library been ‘completed’, and the About section hasn’t been updated?December 12, 2017 at 12:37 pm #23609
Looks like my question was answered. Apparently the library has been completed to contain every possible combination of 1,312,000 characters and the About section still needs to be updatedDecember 12, 2017 at 12:39 pm #23610
Or maybe not, which is it?December 12, 2017 at 9:55 pm #23617
“If completed” might actually just be a rhetorical statement, like “If [the Library has been accurately] completed” rather than a “When it’s completed”. Maybe it’s not possible to test if there are no repeating books, unless the algorithm was designed in such a way that it’s not mathematically possible. I think that claim was made a while back, but I’m not sure of the validity without the code to look at.December 29, 2017 at 6:38 am #23836
After reading this discussion, being a simple person without any advanced knowledge on the mathematical operations of the site, I decided to conduct a small experiment to validate the point made about text appearing in the middle of the gibberish.
It went as follows:
1. Log on to the library of babel
2. Enter the “Search” menu
3. Type “Proving time”, fittingly
4. view results to compare points made
As it turns out, the site appears to have a function which searches for results that happen to have the searched text placed in the middle, however upon selecting the “more random char matches” option, I discovered that the site also contains a very large variety of results for the searched text, with such text appearing all over the page.
A simple argument made by someone jumping to conclusions, proven wrong by a simple experiment by a simple person.