Machine Poetics

NOTE DUMP AHOY!

IDEA:
take a text — a speech, a long poem, part of a book
list the parts of speech

IDEA:
http://islamicmusingsfromthefringe.blogspot.com/
graham aldis: political speech generator based on bush’s speeches, emphasizing the generic re-combinatory nature of the source texts

play a madlib
print some out: go outside

Japanese Haiku:
http://www.writing-world.com/poetry/haiku.shtml

Originally haiku was the hokku, or starting verse of a renga (a collaborative poem containing several stanzas, each stanza written by different or alternating poets). The hokku was about nature and gave a season word so that the collaborators knew what time of year the renga encompassed. Eventually the hokku became independent of the renga and became known as haiku.

The traditional form of Japanese haiku has seventeen onji. Onji, most of them considered as one syllable in English, led modern haiku to having three lines containing seventeen syllables (5-7-5). But onji has shorter sounds than our English language. Sometimes two or three onji characters can be translated to one syllable in English. Many haiku translators believe ten to twelve English syllables would best be used to mimic the original Japanese sound- length form.

Three translations of Basho:
http://www.haikupoetshut.com/basho1.html

Bio of Basho:
http://www.geocities.com/Tokyo/Island/5022/

Haiku Gen

sequence:
“The” adjective noun

verb adverb

“It is” adjective

\n = unix line break

mt_rand:
http://www.phpdig.net/ref/rn35re672.html
http://us3.php.net/mt_rand

Mersenne Twister algorithm
http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/emt.html
Far longer period and far higher order of equidistribution than any other implemented generators. (It is proved that the period is 2^19937-1, and 623-dimensional equidistribution property is assured.)

FAQ:
* Want to use for cryptography.
Mersenne Twister is not cryptographically secure. (MT is based on a linear recursion. Any pseudorandom number sequence generated by a linear recursion is insecure, since from sufficiently long subsequence of the outputs, one can predict the rest of the outputs.)

Mersenne primes:

Many early writers felt that the numbers of the form 2n-1 were prime for all primes n, but in 1536 Hudalricus Regius showed that 211-1 = 2047 was not prime (it is 23.89). By 1603 Pietro Cataldi had correctly verified that 217-1 and 219-1 were both prime, but then incorrectly stated 2n-1 was also prime for 23, 29, 31 and 37. In 1640 Fermat showed Cataldi was wrong about 23 and 37; then Euler in 1738 showed Cataldi was also wrong about 29. Sometime later Euler showed Cataldi’s assertion about 31 was correct.

Enter French monk Marin Mersenne (1588-1648). Mersenne stated in the preface to his Cogitata Physica-Mathematica (1644) that the numbers 2n-1 were prime for

n = 2, 3, 5, 7, 13, 17, 19, 31, 67, 127 and 257

and were composite for all other positive integers n < 257. Mersenne’s (incorrect) conjecture fared only slightly better than Regius’, but still got his name attached to these numbers.

Definition: When 2n-1 is prime it is said to be a Mersenne prime.

What’s Up with Primes?
large random prime numbers are used in most cryptographic systems

more:
http://math.youngzones.org/Math_2213_webpages/prime_numbers.html