h a l f b a k e r y
(Rolling in flour, halfbaking my ass off)
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
Follow the link if you don't know the Library of Babel.
It would be very easy to generate a 3-D virtual
environment resembling the Library of Babel. The
architecture of the Library is merely a series of modules -
hexagonal rooms separated by air shafts with spiral
staircases between the floors
and a mirror. Two small
closets, one for sleeping, one for the bathroom. These
modules tessellate indefinitely in all directions and some
claim form a space mappable onto a hypertoroidal surface.
Within each of the rooms is a standard number of volumes
on a standard number of shelves with a standard number of
pages, lines, characters and so forth, each of which is
The layout of the library is relatively simple to simulate, as
it only requires one room lined with bookshelves plus a
hallway, a mirror, a couple of closets and a section of
spiral staircase to be represented. Regarding the rooms,
for example, really only a sixth of the room needs to be
stored as images because of the symmetry. See the graphic
on the link to illustrate the simplicity of the situation.
Beyond that, the books and their textual content need only
be procedurally generated. The location of the user along
with that of the book they're reading can be used as the
seed for a pseudorandom number to generate the text of
the book. The user can select any book from the shelves
and read the appropriate text, which might just have the
occasional comprehensible string in it. Any other user in
the same place in the library on another occasion would be
able to select the same book and read the same text
because of the algorithm.
The Library is not infinite. I estimate the number of
possible books to be of the order of the cube of one
hundred vigintillion on the long scale, which is a big
number but not so big as cannot be expressed in natural
language. The trick would be to choose an algorithm
which doesn't lead to repetition for generating the text.
Users access the library via VR headsets and gloves, if they
so wish, plus a treadmill, or they can access it in the same
way as they would any 3-D first person game. They can
either be allocated an arbitrary location within the library
or choose a specific room. The probability of any two
users allocated to random rooms within the library is
probably rather small even if the entire population of the
world is using the MUD simultaneously, but if they so
choose, they can be located near each other, or even in the
same room. In fact they could even try to find each other
while located in locations less than, say, 10 000 rooms
To me this seems eminently bakeable, but of course I'm not
a computer programmer.
The Library of Babel by Borges [nineteenthly, Feb 06 2017]
[pocmloc, Feb 08 2017]
[pocmloc, Feb 08 2017]
||Use DNA databases to procedurally generate text. The data is being added to from both forward and back in time.
|| Admittedly there be a lot of recurring stuff but I like the idea of a real world anchor to this infinite text shadow immersion.
||I've seen a few shelves with the human genome on it. That
could be a case of engine licencing, as it were. No reason
why you can't have a database of known genomes of various
organisms, but even with a trillion species it would still be
tiny in comparison to the size of the Library. It could be a
small section of the main library.
||Note that the code for creating this Library program already
lies within the LIbrary. As does the code for creating a
simulation of the Library within the Library program. As
does the code for creating a simulation of the simulation of
the library program...
||That's very true. Also, that code exists with different
degrees of optimisation, in different assembly languages
and high level languages, and possibly with more efficient
algorithms than anyone will ever consider.
||Jonathan Basile has made an algorithmically generated stable version of a universal library, and Keiwan Donyagard has made an Android VR app which uses Basile's site to visualise the Library pretty much as described here. <linkeez>
||If you had a quantum supercomputer acting with a
linguistic filter to can out the complete nonsense, how
long would it take to find any useful information in the
library I wonder?
||Take the output of that, which will still largely be useless
gibberish, and broadcast it to space. Maybe somewhere
in there there will be a deadly insult or compliment to
the nearest alien species that they will respond to and
wonder how we 'learned' their language.