a great thread to revive, thanks for doing that! and thanks for the reminder to read the article, bookmarked it back then but forgot. i’ve been thinking a lot about meta / fragment sequencing, about how it would work, extracting transformations and sequencing them instead of sequencing notes. this article articulates it so well, incredibly thought provoking. and i love the fact she mentions computer sequencing / composing specifically.
apologies for the wall of text below, i find this topic so inspriring - and love that people talk about their overall tt experiences as well as posting specific examples!
as @trickyflemming posted you can translate some of the actions directly into scripts - transposition, rotation etc. that’s a good way to approach coding a teletype scene - start with some simple concept (a metro script driven sequencer based on pattern values, for instance) and then treat scripts as actions that apply some transformation. so then script 1 could swap a couple of steps randomly, script 2 could transpose up by an octave, script 3 - transpose down, script 4 could double the rate, script 5 could decrease the rate etc.
how these actions are triggered also plays a major role, meadowphysics / ansible / clock dividers etc, or have different scripts output pulses and cross patch TT itself. or see these as gestures and connect them to monome walk / pressure points etc, so then you are performing a composition. you could do something completely crazy, like having a trigger input call a script where the script number is determined by the param knob, or by some external CV from a ribbon or a theremine controller.
another way: have each script perform a fragment. say, script 1 plays a 3 note sequence and generates a burst of 5 triggers. script 2 plays 8 notes with random gates. etc etc. then use metro script to sequence scripts.
it might also be helpful to approach coding a scene similarly to doing a patch. often a patch is built by creating something simple and then applying modulation where appropriate. using a simple sequencer as an example again, have metro script simply step through pattern values. then start modulating various things. what if instead of keeping the constant metro rate we change it on each step using pattern values as well? what if we use 2 oscillators and use a random number to determine which CV output gets updated? etc etc.
one way to think about the above is trying to identify fundamental principles your scene runs on - in case of simple sequencer we have direction, speed, number of steps, which outputs are used. what would happen if we tried to change those fundamental principles? put them under CV control? it’s kinda like circuit bending it where you change the rules of the game itself instead of just changing the sequence.
meta sequencing - use teletype with multiple modules and have tt coordinate kria / earthsea / meadowphysics / whatever. sequence your sequencers. using the above idea of scripts as actions use some scripts as actions that affect multiple sequencers at once, a script that resets everything being the simplest example, but doing completely unrelated things could be fun too.
final thought, one thing that i think makes teletype a completely unique module is the descriptive nature of it. patching modular is like that “the incredible machine” game sometimes, you know what you want to do but you have to think how to translate that into a patch (if i want this sequence to play after this sequence is done i need… a switch? a flip flop?). with teletype no such translation is needed, you think “when a trigger comes in i want a random voltage generated on this output and i want 3 trigger outputs triggered sequentially with a random delay in between” and then you can just program that exactly.
a couple more threads for ideas: https://llllllll.co/t/orca-style-sequencing-with-teletype-scripts