I don't believe in this

Home Forums Code, Bugs, Suggestions, Questions I don't believe in this

This topic contains 36 replies, has 3 voices, and was last updated by  messiah 1 month ago.

Viewing 15 posts - 16 through 30 (of 37 total)
  • Author
    Posts
  • #2481 Reply

    Randolph Harrison
    huyt threeped truckline steepling plasticizations fructifiers outrageously forge
    tter overseeding shuln repartitions undomesticates transcriptases sanghats lirip
    ipes outbalancing inoperableness ornithodelphian suably miches resinousnesses wa
    rmblood yuke milkmen garnisheeing apolitically mures radiantly ingenious millsca
    le devalorizes ossifications leeping amebic faradisation paraparetic unphilosoph
    ical cliticizing hypoglycaemias morbidest consistence durned rackings tiffany co
    rroborating misrelating parietals overtalking classicizing capturing regredience
    s disembellishes kutus dedifferentiating moonless penpoints comprizes periblems
    teutonize cowbinds palpableness orogenic hematopoiesis assailant roughish repos
    subcontractor idempotencies epistemologies decrementing punters oestrogens insul
    ting wastery unbrokenness chamaephyte tilelike paddlefish outwrestle revolvency
    deionized licentiousness broughta obliteration puttiers cheeringly fiascos resur
    rector plantless overbetting isatinic prostrates increasingly nummuline epilogiz
    ed detriment demonically frightens subchords transcendentalists rodmen redoublin
    g diablerie reservists trellis enchantress aerolithology margravial sewed elasti
    cations hydroxyzine debouch duotones choicest plenishings decaudate misdeemed mi
    crophyte philodendra niggard windguns denar neives ladens toenailing rescreens w
    heelhouse shears kroons counterreformation manholes paedologist bromhidroses clo
    ture bioregionalisms climbable ungulates dizzied mucopeptide beefier drearer erg
    osterols micaceous conflagrant chitin prefired carryback umami kibbehs <strong>here is t
    he proof that this site is real</strong> dance minnock nervelet copyedited alliteratively
     stupidest cabochons doziness soubriquets chamomiles ferrelled surrebutting coun
    terdemonstrator integrative earthshakingly gaws micromanagements paraffinic midn
    ightly frostily zeros leatheriness headachier effeminize etherizations pupa topl
    ined sendup legendaries chesting legislatively complotted fannells foggers hende
    d unprison plastometers starlight narcein indissociably denumerably countergueri
    lla deys savant putrefaction dyspnoeas gagster shabracks peroneus maxillipedary
    owt fleche postexilic trammed systematized contrafagottos slobbering minions val
    guses subprioress taboulehs ebullioscopy precut syndactylisms plasmodesms chaung
    e dehydrators excuse apostolizing dicastery erosivenesses epyllion insheltered c
    lemmed remarketed excarnated salamandroids predeterminable cobelligerent poaches
     apodictically holytides einstein gibberishes meliorisms lectorate nonnationals
    outdoorsman padouk benedict rendering arrestive infallibleness flow whoremongers
     viameters remonetized twinkler churrus underfinished extravaganzas excerpting p
    olestars interlunary hymenophores cladoceran torticollar packs omniparities sked
    addles unpropitious extemporary idlest chalcedonyx suggestionize palps cosses at
    imy statoblasts unbraiding increscent weirded corydalises basins celibates loric
    s deflorating paradoxer curd agroindustries koreros sinlessnesses newsinesses re
    ligioners enamellist perfins continuants paleozoologies whombled flagellate ratt
    ed silencer convexes photothermal elicit hypersalinity vulvas imperators republi
    #2503 Reply

    Zaghdur

    Well thatsme, that didn’t prove anything.
    Dont get me wrong, I understood what you meant and I totaly “believe” this is not a hoax.

    My point was that if you want to prove this to someone who does not understand the concept you need to use this algorithm offline and then try it with another computer with online access. Then show that the online version gives the same result with same HEX and so on… without any connection between the computers. This is also the easiest way to check if the site works correctly (which I have no doubt that it would not) to those who aren’t math orientated.

    Using it for communication still leaves the explanation that the site generates texts to online.

    #2517 Reply

    Randolph Harrison

    that didn’t prove anything.

    It proved something, though not that the algorithm does what is being claimed. The algorithm is reliably producing the same text at approximately the same place at a specific address from a seed. That proves that the seed somehow encodes the text.

    I haven’t tried the code yet, being I’m not a Python programmer, but in another thread someone has reproduced the effect and posted to GitHub. I’m going to attempt to do it myself in Ruby.

    Of course that still doesn’t prove (or disprove) what’s being claimed on this site. It only proves you can produce the exact same effect (which would work online or offline) without any need to invoke mystical references to “every possible book every written”. Instead, all you’re doing is hashing what people input in an interesting way, and randomizing the addressing of the hash keys across an impossibly large address space.

    #5969 Reply

    Ben K

    1. This will work on offline computers.

    2. This is still a hoax and there isn’t a single “book” in this “library” which has more than one page which makes sense

    3. Notice that every page has exactly 3200 characters

    4. Notice that every hex generated by the search is about 3250-3260 characters long

    -> The hex is an encoded form of the page you “searched for” and the Page, Volume, shelf and Wall are the decryption key.

    This means all other pages in the same hex are essentially wrongly decoded versions of the same page.

    #5970 Reply

    Jesse

    As I see it, there is definitely no reliable proof, this isn’t a hoax, as long as people can’t look at the algorithm.

    But on the other hand, I think creating such an algorithm is possible, because it’s logical. It’s “just” putting all characters in all possible combinations. This surely can be done with nifty scripting, why not? And once it is completed the rest is just computing. It will be able to generate the whole Library of Babel at once. So I believe, this website is true!

    And why the heck someone should build a hoax around a short story by Jorge Luis Borges, equip it with clever texts about it’s theory, a forum and an image search following the same idea? This project is very special, includes a lot of philosophy and is definitely not what most people are thinking of most of their time. Someone who wants to just cheat people will more likely do something more spectacular, like photoshop some moon photos, to “proove” the Nasa hasn’t been there.

    But just for the fun I was thinking:
    I’m no programmer or math geek or whatever and I can’t imagine how this algorithm should work in detail. It’s to big for me. But I’ve learned to write some PHP. So what would I do to fake this library in PHP? Not quite easy, but it should be possible. First I would create a site structur, that allows a visitor to browse to any books pages. I think I could manage this somehow. Then whenever someone looks up a page, it would be generated by random and saved on the server. If he looked it up by chance, it would be filled with just random characters, if he searched for something, this “something” would be included. Whenever someone looks up a page, that has been generated and saved before, he will see it again. Otherwise a new page will be generated for him. To implement the “Anglishize” function I also would need a dictionary in it. But that’s what it takes in a rough sketch. This could run for a long time without giving any clue it’s a hoax. Till one day its webspace is full . . .

    Once again: I believe this site is true!
    And what I like most about it is how it makes you think. 😉

    #5985 Reply

    Jack

    Imagine a page with the letter ‘a’ on it. Now imagine a page with the letter ‘b’ on it. Now imagine a page with the letter ‘c’ on it. Continue until you get to ‘z’.

    Then imagine a page with ‘aa’ on it, then ‘ab’, then ‘ac’. Continue to ‘az’. Next comes ‘ba’ then ‘bb’ etc…

    If I follow the pattern until I get a page of all ‘z’s I will have gone through every possible combination of characters possible on a page. If we add in spaces, periods and commas then I would have generated every set of phrases that could be made with those characters. Because I know the pattern that was used to generate these pages I could determine the contents of an arbitrary page so long as I know its number, and I could determine the page(s) a piece of content can be found.

    If all that made sense then you can understand how this library works. The library uses a much more complicated algorithm to determine pages but the concept is the same. When you search for a bit of text the library isn’t being searched, instead the text is being run through a sort of reverse of the algorithm that is used to generate pages in order to determine where that text can be found.

    It’s important to remember that the library contains every possible page, not book. So you are very unlikely (essentially impossible) to find two consecutive pages that make sense. This is because there are far more combinations of letters that don’t make sense than do make sense. Same goes for the images.

    Even if the image slide show was completely random, the odds of ever finding an image that looks like something other than static are incredibly small. There are so many ways to arrange pixels so they look like randomness and only a (comparatively) small number of ways to make them look like anything recognizable. But if you sat there watching the slide show long enough (past the death of the universe probably) you would eventually see a picture of me typing this response.

    Anyway, Jonathan has said he’ll release both an offline version of the library and it’s source code at some point this year; so when that happens we’ll finally know for certain if this is all real.

    #6206 Reply

    Medicine Storm

    In the hex given above, wall #1, shelf #2, volume #5, page #8 I get the following:

    huyt threeped truckline steepling plasticizations fructifiers outrageously forge
    tter overseeding shuln repartitions undomesticates transcriptases sanghats lirip
    ipes outbalancing inoperableness ornithodelphian suably miches resinousnesses wa
    rmblood yuke milkmen garnisheeing apolitically mures radiantly ingenious millsca
    le devalorizes ossifications leeping amebic faradisation paraparetic unphilosoph
    ical cliticizing hypoglycaemias morbidest consistence durned rackings tiffany co
    rroborating misrelating parietals overtalking classicizing capturing regredience
    s disembellishes kutus dedifferentiating moonless penpoints comprizes periblems
    teutonize cowbinds palpableness orogenic hematopoiesis assailant roughish repos
    subcontractor idempotencies epistemologies decrementing punters oestrogens insul
    ting wastery unbrokenness chamaephyte tilelike paddlefish outwrestle revolvency
    deionized licentiousness broughta obliteration puttiers cheeringly fiascos resur
    rector plantless overbetting isatinic prostrates increasingly nummuline epilogiz
    ed detriment demonically frightens subchords transcendentalists rodmen redoublin
    g diablerie reservists trellis enchantress aerolithology margravial sewed elasti
    cations hydroxyzine debouch duotones choicest plenishings decaudate misdeemed mi
    crophyte philodendra niggard windguns denar neives ladens toenailing rescreens w
    heelhouse shears kroons counterreformation manholes paedologist bromhidroses clo
    ture bioregionalisms climbable ungulates dizzied mucopeptide beefier drearer erg
    osterols micaceous conflagrant chitin prefired carryback umami kibbehs here is t
    he proof that this site is real dance minnock nervelet copyedited alliteratively
    stupidest cabochons doziness soubriquets chamomiles ferrelled surrebutting coun
    terdemonstrator integrative earthshakingly gaws micromanagements paraffinic midn
    ightly frostily zeros leatheriness headachier effeminize etherizations pupa topl
    ined sendup legendaries chesting legislatively complotted fannells foggers hende
    d unprison plastometers starlight narcein indissociably denumerably countergueri
    lla deys savant putrefaction dyspnoeas gagster shabracks peroneus maxillipedary
    owt fleche postexilic trammed systematized contrafagottos slobbering minions val
    guses subprioress taboulehs ebullioscopy precut syndactylisms plasmodesms chaung
    e dehydrators excuse apostolizing dicastery erosivenesses epyllion insheltered c
    lemmed remarketed excarnated salamandroids predeterminable cobelligerent poaches
    apodictically holytides einstein gibberishes meliorisms lectorate nonnationals
    outdoorsman padouk benedict rendering arrestive infallibleness flow whoremongers
    viameters remonetized twinkler churrus underfinished extravaganzas excerpting p
    olestars interlunary hymenophores cladoceran torticollar packs omniparities sked
    addles unpropitious extemporary idlest chalcedonyx suggestionize palps cosses at
    imy statoblasts unbraiding increscent weirded corydalises basins celibates loric
    s deflorating paradoxer curd agroindustries koreros sinlessnesses newsinesses re
    ligioners enamellist perfins continuants paleozoologies whombled flagellate ratt
    ed silencer convexes photothermal elicit hypersalinity vulvas imperators republi

    seems like quite the collection of uncommon words.

    #6216 Reply

    Rehab Submarine

    seems like quite the collection of uncommon words.

    I suspect this is the case because when you do a search for english words, the search generates words from a word list and then searches for them (probably the unix word list). I doubt that the algorithm is robust enough to search for english words without providing a search term. He could probably use a smaller word list to search though, so it’s more sensical.

    #6252 Reply

    Zeth Matthews

    From what I can see, the pages are definitely procedurally generated. That makes sense, as there’s no way to store the entire library on known media, let alone search it. If this algorithm truly never creates duplicates, then I’m pleased with this method. This also means that hours and hours of manual searching should reveal the interesting plain text combinations. The only thing that really gets to me is that when I search for say “potato” or the like, I get several kinds of results. One of these being a blank page with just “potato” on it. I suppose this technically HAS to happen if the algorithm can generate all possible permutations of 3200 characters. The odd part is that it’s surrounded by sufficient blank space to cut off at just the edge of the page before resuming the streams of gibberish. Again, this HAS to happen at some point in all permutations of 3200 characters, but it still looks strange. Why don’t I ever happen on the random page that is mostly blank space? Not just one chunk of blank space, but predominantly blank space? And when my single word is in a page full of English words, the pages to the left and right are not of a remotely similar nature. This algorithm definitely has me curious.

    The logic still stands that all of these strange things we find HAVE to happen at some point in all permutations of 3200 characters. I don’t think we firmly grasp just how much 10^4677 books (at about 410 pages each) can contain. For emphasis I’ll say that 10^4677 is an absurdly huge and unrealistic number. I mean, the estimated number of atoms in the universe is somewhere in the range of 10^78 to 10^82. If each book were stored on an atom, we’d still have nowhere near the number of books this algorithm can put out.

    #6447 Reply

    MMC

    As I see it, there is definitely no reliable proof, this isn’t a hoax, as long as people can’t look at the algorithm.

    There is no algorithm that can produce the search effect that the Library of Babel claims. It is 100% a hoax. Search is accomplished by encrypted padding around a cleartext insertion of the string being sought. A real search would use a sieve method.

    Its just a variation of a symmetric cipher.

    As for the book itself, that is quite real, its just every permutation of potential characters. I posted code to a bigger version of this that uses Base64, so it will also cover every potential binary, image, song, etc.

    In regards to search, there are some higher dimensional models that can perform clustering around statistically likely common words and phrases that could speed things up. Its would change how the book is arranged and navigated, but it may make it more useful for finding things.

    So, in all, this site has been a very good start, at least to get people thinking about the idea and what things may be hidden in those pages.

    #6522 Reply

    Jonathan Basile
    Keymaster

    I really think the site would only qualify as a hoax if it somehow tried to deceive its users about how it works – I’ve written in extensive detail about the algorithm, in a way that even users who don’t code can understand: https://libraryofbabel.info/theory4.html

    You’re correct that the algorithm being used is similar to many encryption/decryption algorithms – this is exactly what I and others (for example: https://youtu.be/GDrBIKOR01c?t=17m) have been explaining since the creation of the site.

    If the algorithm is unimpressive or uninteresting to some people I understand that, but I think the implication that I’m deceiving anyone is inaccurate.

    -Jonathan

    #6709 Reply

    Keiwan
    Participant

    You might want to take a look at this little example file I wrote to maybe understand the basic concept of the library and its searchability until Jonathan releases the actual source code. This should at least make it clear that the library is not a hoax.

    https://github.com/keiwando/libraryofbabel-example

    #6727 Reply

    J C

    I think it is clear that the page is “produced” rather that “sought”

    I wish it were true to just be “flipping around pages” and find a Gospel or a Shakespearean Sonnet. The problem is when one “searches” it, it is logical to me to believe that the program would “find” at least 29 instances of any text entered. This is because if every bit of text possible were in the library, then it would exist followed or preceded by each possible character.

    For example, say the word “one” were searched. Then it stands to reason with 26 letters, comma, space, and period– that it should exist in at least 29 instances followed by each letter , . or space.

    onea oneb onec oned onee etc.

    This is compounded that it should, by design, produce at least 29 instances of more extensive passages of up to 3200 characters. I am disappointed that only 5 results ever seem to show up

    #6746 Reply

    Jonathan Basile
    Keymaster

    You can click on “more…matches” to see more.

    #13192 Reply

    Ray

    The reason your image exists exactly, yet cannot be found through the universal slide show, is because there is an unfathomable amount of various pixel arrangements. You cannot really ever find an image without first knowing what it is composed of exactly. From there, the image gallery can use an algorithm to determine what string of characters would normally generate that image.

    The reason you get different results for one search is because that text can appear with such a wide variety of text around it. “hello there” has a MASSIVE amount of pages, because the string is so short.

    It’s also important to note that every page exists in the library, but since there is still so much space pages can still repeat. This raises search results even more, since there are duplicate pages in different locations.

Viewing 15 posts - 16 through 30 (of 37 total)
Reply To: I don't believe in this
Your information: