Scholarly discussion of the music of John Cage.

[David Bellows <>]

Rod, thank you very much!

As I was checking my results against Nicki's, I noticed that hers
would leave out words in the wing words which she justified with the
following:

"Cage: 'Then I take out the words I don't want.'
My interpretation: Don't add all possible words between spine letters."

So I then added the option to randomly remove words within the wing
words. Originally the software would preserve connected words and then
allow an option to truncate the wing words from random chosen
positions. So given:

"to be As accurate to cage as i can"

might be:

"be As accurate to cage"

or :

"As accurate to cage as I"

and so on. So at least I had it correct the first time!

The software also truncates wings at 45 characters which is something
else she mentions. Meaning, that if the number of characters is
greater than 45 the software removes enough words from whichever is
the offending end to get the total below 45.

> Whether there are 7 more words or only one word in the "branch” is 
> something that might better be decided by a choice-deciding brain.

Yeah, there's no way around that one. The software is primarily about
creating music using as many musical ideas I can possibly program in.
I need text generators for lyrics or even for stand alone poetry and
mesostics was an obvious one. The results of the random calculations
are actually determinate (pseudo random number generator) whose
results are entirely dependent on the initial input of the user. So if
I use your name the results will be different than if I use my own.

For example, using the spine "John Cage" with Walden and using your
name as the seed we get:

                              Judge
                              Of
                              Him.
                             fiNest
                          whiCh
                             hAve
or stolen time, robbinG
                            crEditors

Same spine and source but with my name as the seed:

                                           Judge
                                           Of
                                           Him.
                                          fiNest
                                       whiCh
                                           hAve
borrowed or stolen time, robbinG
                                          crEditors

Not a huge difference, but as you generate more poems you get more 
differences.

So even though aesthetic choices aren't being made, hopefully the
oracular nature of tying the output to the initial input (name, birth
date, or literally any combination of characters and then "casting
lots") provides its own kind of aesthetic experience. At least that's
the philosophy behind the project. The difference between this and an
oracle is that the result will always be the exact same given the same
initial input (seed).

Dave

On Mon, Jan 14, 2019 at 2:33 PM Rod Stasick <> wrote:
>
>         On Jan 14, 2019, at 13:17, David Bellows <> 
> wrote:
>
>         The other point is that once you have two spine words are the wing
>         words chosen randomly from all the valid words between them? Right 
> now
>         my software chooses up to 11 words before and after each spine word
>         (meeting the mesostic rules) and then, depending on the option 
> chosen,
>         randomly chooses from those words.
>
>
>
> If I’m reading your comment correctly, this was the idea that Nicki Hoffman 
> had,
> at first, implemented incorrectly, so I wouldn’t try doing that hack on 
> your software.
> Think of the “wing words” as simply extensions of what has come before the 
> mesoletter
> and another extension of what has come after the mesoletter. You can also 
> think of the
> words as a branch of the mesotree that is just extended left or right of 
> the mesoletter:
> some are short branches, and some are longer ones - each keeping within the 
> confines
> of the mesorules. There are no breaks in a branch, for instance, where you 
> can cut out
> middle segments and so on.
>
> An example using your words (and only the first two letters of your name):
>
> “...I'm definitely trying to be as accurate to Cage as I can and
> especially so once he was using computers to assist the process...”
>
>          i’m Definitely trying
>      to be As accurate to cage as i can
>
> [I hope this lines up, but I think you’ll get it].
>
> Anyway, if you look at the second line,
> you aren’t allowed to pick and choose which words
> you want from the branch. For example you can’t decide to say:
>
> “to be As accurate as i can”
>
> nor can you say:
>
> “to be As cage as i can”
>
> or any variation where you are choosing to jump around.
>
> Also, notice:
> You can extend your branches to varying lengths
> as long as the words are connected in a string,
> so with the above mini-mesostic you can have
> either of these versions
>
>          i’m Definitely trying to be
>                 As accurate to cage as i can
>
>          i’m Definitely trying
>      to be As accurate to cage as i can
>
>
> My comment concerning the limits of the programming language
> was more about how you can’t rely on the random number generator
> to decide something that may actually be an aesthetic decision about
> “wing word" lengths. Whether there are 7 more words or only one word in
> the "branch” is something that might better be decided by a choice-deciding 
> brain.
>
> hope this helps,
>
> Rod
>
>
>
>
> *************
> Now Playing:
>
> ‌Edward Avis‌ - Evening in Birdland
>
> from:
>
> ‌Ecstatic & Wingless: Bird-Imitation on Four Continents, Ca. 1910-44‌