Skip to content

Infinte Monkeys v1.80

January 27, 2012
by

Hi all. I wanted to plug the newest version of Infinite Monkeys. This release features some of the neatest conveniences you’ll ever find in an n-gram generator.

Firstly, because Infinite Monkeys is a standalone local executable, you can now load and save ‘biases.’ The biases are saved n-gram states. For instance, I have a new book I’ve been working on which utilizes the Language of Eden. The LoE is basically composed of the following texts: all my work, Ezekiel, Isaiah, Hosea, Zechariah, The Book of Daniel, and of course, Revelation. Since it is a major inconvenience to have to ‘re-eat’ all of these texts every time you load up the program, I made it so that you can save the n-gram data and reload it whenever. You can also add to the corpus later if you want to.

One minor glitch with this may be that if you do not specify the “.txt” extension, IM may not be able to find the file. You can always go back and rename it though, something to be on the lookout for.

Also, I have added a rather clumsy BiasMaster which allows you to cycle through n-gram data, and edit it from within a rather crappy and none too elegant GUI. I’m an in the midst of thinking of a smarter way to accomplish this. Please bare with me.

One last thing. I have updated the manual to include all the info you need to get started with n-grams. Another neat feature: n-grams can be called from within scripts. For example:

#NG[250]

…would produce 250 words worth of n-gram garble.

Thanks and enjoy, and if you feel so inclined, feedback is appreciated. One issue is load times on files which will cause windows to appear to be ‘NOT RESPONDING’. If you wait long enough, I assure, windows will respond. Apologies.

Until I switch to GTK,

JYNX

code.google.com/p/infinitemonkeys/

Advertisements
2 Comments leave one →
  1. January 31, 2012 6:28 am

    Awright, I finally got around to trying it out! (actually, the hang-up was finding something to unrar it, since windows didn’t handle it natively.)

    I have added a rather clumsy BiasMaster which allows you to cycle through n-gram data, and edit it from within a rather crappy and none too elegant GUI. I’m an in the midst of thinking of a smarter way to accomplish this.

    Why not stick with the text files? At some point you have to figure anyone who wants to work with it to any degree of complexity has got to be willing to at least edit files by hand every now and then.

    My attitude has always been to build the tools that I want to use, and worry about usability for the general public only if I feel like it. I guess that’s the difference between 1) a generator as a tool for you, the programmer 2) a generator as a tool for the least common denominator of an audience and 3) a generator that is itself the aesthetic focus (instead of the generator’s output being the aesthetic focus) as in many of the Flash generators programmed by Digital Poets. Item 3 tends to result in fairly simple generators (since most of the focus is on appearance) and 2 is a vast time-sink. Item 1 gets you a generator that to other people often seems like an alien artifact – but isn’t that in itself kind of cool? Even something as seemingly straightforward as charNG has been described as kind of mysterious by other digital poets (not on this site!) But who cares? For me it’s all about the output.

    I think I’m getting the hang of using the biases… I’ll have to read through your earlier posts and get clear in my head what they are mathematically…

  2. January 31, 2012 9:38 am

    @edde

    In Western ontological systems, the impulse is to keep your ‘ontology’ as small as possible while simultaneously being able to explain as much as you can. So the ontology is a system of committed values and philosophemes which generate ‘explanations.’ The Bias is really no different from regular n-grams, the pitch being that you have more say than you would with a regular n-gram set. The ability to manipulate the bias is the only real innovation here, and I don’t think other generators allow this. But there’s also more work to be done as far as internal alliterations are concerned, and the next time around, you will be able to bias the n-gram data via substrings like say ‘str’ would return: ‘straight astronauts strong astray’ or something.

    When I made IM, I wanted it to be able to be used for general consumption, I wanted it to be a itself an object of art, and I of course wanted it to be a useful tool for me. Do you think its too abstruse? I tried to make it as simple as possible, but I also wanted to talk about philosophical issues of metaphysics and have it be an example of some of philosophical issues it addresses. So, I’m interested in all three possibilities for IM, and will continue to work to see them through. What would you suggest? It’s hard to get any kind of feedback on a generator. People seem more interested in my output. But the tool itself is just as important, down to the fact that it uses a moniker which is synonymous with ‘brute force.’

    Part of the complexity is the fact that I’m interested in programming actual languages, and so I wanted to invent a language for generating language as per props to my langpo homies who turned language into an agent.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: