4,704 Swan Songs & 1 Opinion

The code is now at a stage where if I set it to loop and sent the 57k rap songs I have in archive from ohhla to alchemy, I could generate, an unfathomable amount of unreadable crap (also known as c-rap: computational rap).

But I think I have come to the end of the synset road. Next step is to investigate Theano: Unsupervised learning, deep neural nets. Perhaps transition to Python 3.0 unicode. It might take 6 months to find the concentrated time. Until then, I am on hiatus. Got a book to write. By hand.

 My opinion: In spite of all the machine learning hype, computers are a long way away from independently generating credible connected contextual intuitive experiential poems or stories capable of emotional or conceptual cathartic impact. The process will in the near-future (10 years) require extremely creative and intuitive data science-artists to find the statistical-sculptures within the mountain-ranges of data fountaining from networks. It will be rich and exciting work to chart and out and develop classifiers trained on huge datasets which then generate simulacra of the writing they have ingested.

In the longterm, all serious writers will use computational assistants to suggest and enhance and refine linguistic creativity. And eventually, writing itself will vanish, artists will simply edit dreams and notions, and networks will drink and translate those dreams into user-specified formats. At that point poetry will become a mode of listening, as it is now, receptive, open, crouched down amongst the wind.

Shoutout: for an informed perspective on poetry-generation, see Gnoetry.

As a parting salvo, I did a very rough generation using templates based on lyrics by : 50 Cent, A Perfect Circle, Abba, Acdc, Alkaline Trio, Bob Dylan, Bob Marley, Counting Crows, Cranberries, David Bowie, Deep Purple, Dragonforce, Evanescence, Everlast, Frank Sinatra, Helloween, Guns ‘N Roses, Jimi Hendrix, Linkin Park, Nick Cave & The Bad Seeds, Patti Smith, Paul McCartney, Pink Floyd, Placebo, Radiohead, Ramones, Red Hot Chili Peppers, Rolling Stones, Scorpions, Suicidal Tendencies, System of a Down, The Beatles, The Blues Brothers, The Clash, Tom Waits, and U2.

4,704 new computer-generated song lyrics just waiting for machine-generated melodies, a robot to sing them and another robot to weep or dance. Read them here.

The first stanza of 50 cent‘s I’m gonna be alright rewritten:

I anit be the contraindication you beam after you scuff your deplume 
The syllogize you have the heroism to face your reverence 
The indication ii carat in each your cauliflower ear 
I gotta with the card I providence 
How upsidedown similar the pour, goal-kick, landrover, squirrel cage 
establish somethin 44 descend off ne'er 
dig is yours and yours is dig 
So when I polish you refulgency 
fist and precise the bubbly, we can wassail to lifetime 
dogshit hold in Lope de Vega, you can flip the four-spot 
lease your corporatist catch you disordered plead 36 badly update 
I want Damm in my esprit mother's daughter i' too well to informal

The final stanza of Bob Dylan‘s Blowing in the Wind rewritten:

How few school year can a volcano breathe 
Before we's gargle to the ocean? 
no, how few annum can some nationality dwell 
Before you're grant to be unoccupied? 
no, how few prison term can a crew move around his school principal, 
suspect he just project? 
The urim and thummim, my Quaker, is in the wind up, 
The tide over is in the wreathe. 

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Another 10k day

I’m beginning to understand the exultation of spam-lords, the rapturous power narcotic that arises from watching thousands of words of perhaps-dubious quality arise & spew in a rapid unreadable scrawl across a screen.

Beyond semantics, words like sperm procreate incessantly in abundant sementics. Quality in this inverted world is a quantity.

On the technical side: today, I fixed the repetition hatching; used pattern.en to correct articles (like ‘an’ or ‘a’) and conjugate correct verb participles (as in ‘I’m walking home…’); and created FAKE_authors (because  who wants to read a poem written by a bot…unless it’s good, which these poems are not yet).

It all took much longer than anticipated.

The poems are now output in hourly batches:

Here’s a weird sample:

Body The New Road: Clark
by Anthony Lazarus




                        hold back. expect.
                look forward
                                                          kick one’s heels.
                                  kick one’s heels an i kick one’s heels.

                        kick one’s heels.

hold off.

                                                          look to.
                                                         stand by.
kick one’s heels.


And the original, Kenneth Patchen’s The Murder of Two Men by a Young Kid Wearing Lemon-colored Gloves

                        Wait. Wait.
                                  W a i t.


Code on Github
Made by Glia.ca  

[ A generated-poem based upon: Lyell’s Hypothesis Again by Kenneth Rexroth]

Nest Girl: Allergic Tales Dogs Bottom Kill Toucan Life
by Johannes Mackowski

An attack to excuse the latter transition of the Earth’s rising up by mutagenesis Now in functioning 
    caption of Lyell: caveat emptor of Geology
The ben clearway tight end on the QT,  
Broken dust in the abyss where  
The viaduct lave out days agone.   Continue reading “Another 10k day”

Hatching (trance bug poem set)

Another day, another 10k.

I received an email today from a friend who is a poet named Ian Hatcher; his email included an MP3 of himself reading a poem I had generated using code that was a bit broken .

Ian Hatcher reads “woodland_pattern” 

I sent this particular poem to Ian because it included a lot of repetition; Ian’s style involves repetition, productive repetitions, calibrated repetitions, sung repetitions, drone repetitions, blind repetitions, profound repetitions.

To my mind the repetition was a bug; from the perspective of efficient communication, a repeated word is an inefficient redundant symbol. In my mind, an ancient mantra chanted: effective text is succinct and to the point. But Hatcher’s work (among others) reminds me of the parallel/opposite tradition of trance ritual (incantation incantation incantation …) appropriated by post-modern poetics. Time does not matter: matter cycles.

Poetry belongs to both traditions (efficient condensation and redundant trance-inducing repetition). It is both a sustainment of affective efficient communication where redundancy is reduced and it is also a mode of being that involves states of consciousness invoked by rhythm and repetition.

Repetition is exactly what my code started churning out unexpectedly today after I made an error in how I dealt with 2 letter prepositions followed by punctuation. So I generated 10k+ poems in that style.

A computer-generated stanza

predictably, aluminum business and USDA diverge diverge
diverge            and the deflagration catapult 
sow in the sagas van chicken farm carnival  carnival
carnival                 with shallot desktop and radicchio 

based on a template derived from D. A. Powellrepublic (2008)

soon, industry and agriculture converged
                        and the combustion engine
sowed the dirtclod truck farms green
                                  with onion tops and chicory

To read 10118 poems (laden with repetitive bug trance cruft) generated in 6966.78432703 seconds on 2014-07-30 at 22:42 click here

Code on Github
Made by Glia.ca  

p.s. If you think trance is a thing of the past, consider the repetitive contemporary pop dance track just released by airhead (also sent to me as a link in an email today). Incantation is secular.  

Collocation: a poem culled from cruft

Data-art involves days of futile searches. Redundant processes, meandering through archives trying to retrieve relevant results. In this case, a collocation search resulted (as usual) in 99% pure cruft.

Cruft may be a synonym for poetry: the craft of recycling phenomenological debris into revelations.

Process: a trigram collocation search [in other words, a search for 3 words together] of 10,573 poems from the Poetry Foundation.

Filtered for the most frequent 3 word phrases:
before filter: 2,654,603 phrases,
after filter: 532,852 phrases.


idle fears! Vanish
idle flitting phantasies,
idle hand sweeping
idle handle, resting 

joys Of sense
joys Of serving
joys Of subtler ... love I fell love I fill love I had.  milk & sugar milk . . milk And blood milk From burning ... sexual feud And sexual grove.  time, I climbed time, I cut

Read it here

Generated on 2014-07-27 at 16h
Selected at 2014-07-28 at 14h

Continue reading “Collocation: a poem culled from cruft”

Synset to Lyrics ( how-to write over 5000 poems in an hour or so)

  1. Download a website or torrent of song lyrics
  2. Parse the result (with BeautifulSoup)
  3. Use NLTK (Natural Language Toolkit) to lookup Wordnet synsets (homonyms, holonyms, hypernyms, antonyms, meronyms, synonyms and lemmas) as replacement candidates.
  4. Replace all the words in the poem (retaining the words-per-line and verse and grammar of prepositions and pronouns)
  5. Sift through ‘the reservoir of words that returned no results from synset search to find dense tweets worthy of a bot.
  6. Or check out the [ for now, terrible] 5,067 poems created in 8898 secs
reservoir of unique words
click on image to see code on github

Have a peek at a few results posted on my twitter feed:

Screen Shot 2014-07-13 at 4.02.05 pm

The poetic problem is that these poems lack insight. Blastocytes have orders of magnitude more experiential insight than these poems.

Continue reading “Synset to Lyrics ( how-to write over 5000 poems in an hour or so)”