Scholarly discussion of the music of John Cage.

[Stefano Pocci <>]

On 04/27/2013 06:33 AM, Zachary Bond wrote:

I was sad that the online "mesostic" generator was taken down and replaced with a juvenile message instead of being fixed, so I whipped up a generator in the Python programming language.  It obeys the 100% rule, but I was surprised at how many other decisions I had to make about how to limit the number of wing words, how to actually go about selecting the words with the mesostic letter, etc.  Basically, I choose the word by searching from a random point in the text.  Then, I select up to 12 wing words, with the 12 distributed randomly on either side of the center word (for example, a line might have a max of 3 on the left, max of 9 on the right).  Only rarely do you ever get 12 since usually a rule violation typically occurs first.   I also made it circular, so that the first letter obeys the rule with respect to the last letter--that's to avoid having conspicuously long starting and ending lines.

I did not try to prevent the same word being used more than once--I seem to recall repeats in some of Cage's mesostics, and sometimes you don't have a choice if your source text doesn't have many Zs, Js, etc. 

The next step is to figure out how to get this on a website with a usable interface.  It's been a fun way to refresh my Python skills.

Here's an example based on 10 iterations of "Mesostic" and the Wikipedia entry on Cage as a source text.  Suggestions welcome.  Hope you've got a fixed width font.

-Zac Bond

Hi Zac, thanks for your message that seemed to be quite appropriate in this moment.

After the removal of the mesostic generator, I decided in fact to build one myself too. Instead of Python, which I don't know, I chose Java (which I know a little) and just yesterday I managed to complete the most difficult part - I thought - that is the construction of a mesostic according to the 100% rule. I still haven't limited the number of wing words and if you say that's quite a feat also... well, then I'm not half way through yet :-) I originally thought to trim the mesostic once obtained a very wide one (with a lot of wing words), but your on-the-fly random process approach seems more challenging, from a programming point of view at least.

So far in my trials, I found that long horizontal lines might appear only at the very beginning, in the very first line where you need to pay attention to the first mesoLetter only. But it depends on the text and the mesostic words of course. And the language too: I'm Italian and I haven't yet considered the letters with accents such as è, ò, à, ù, ì that are so present in my language. Unlike the umlauts and strange marks above or below letters inside the Finnish, Estonian, Swedish and other languages vocabularies, those accents are not new letters of the alphabet. It is just an accent added to the original letter, which I presume, must be then considered during the construction of a mesostic.

What I created is more a program that 'writes through' a whole text, but I bet my program will have some troubles with very large texts. The reason while I embraced Java is that I would like to make my application run on a website, therefore I will need to study how to make it available to everyone once it is finished. What I did not consider however, was that maybe languages like Python work really fine with strings... perhaps I'll talk about these more technical details with you off list, if you don't mind.

Keep you posted, ciao

-- 

Stefano


In Italy for thirty years under the Borgias they had warfare, terror, murder, bloodshed —but they produced Michelangelo, Leonardo da Vinci, and the Renaissance. In Switzerland they had brotherly love, 500 years of democracy and peace, and what did that produce? The cuckoo clock.

- Harry Lime (Orson Welles) in the "The Third Man" -