Part 1, ‘Generative Text Generation’
Daniel Shiffman
is Assistant Arts Professor at the Interactive Telecommunications
Program at NYU’s Tisch School of the Arts.
The professor has developed a
Markov chain-based Generative Text computer programme which he describes (and provides the code for) here.
The programme is able to generate English text which appears at first
glance to make sense, but in reality has more in common with
‘context-free-grammar’ or, as some might say, gibberish.
It should be
noted that the words themselves are not altered from the original
standard English, but the order in which they appear is unconventional.
Here is an example ‘poem’
* * *
Nonsense, Matriona will feed it.
It is always either vodka or brandy.
Yet I am sorry to leave.
You should not handle youthful egoism so roughly, sister.
What did I hurt my poor boy?
No, indeed, are ambition; for the first day.
Yes, they are singing across the water.
It is like a beggar beneath your window.
* * *
Coming soon: Computing Gibberish (part 2)
Bonus : Here are
gibberish exponents extraordinaire Eric (Monty Python) Idle and
colleague Henry Woolf explaining things on Rutland Weekend Television.
(circa. Nineteen seventy five and a half):
