Been nerdseq'n for a couple of years, only got involved in feature requests around pattern export. I have thanked Thomas many times, but I'll do it again: thanks! You're a wizard.
As background to my user case, I have a lot of obligations and there are long periods where I can't touch music. Nerdseq has been tremendous in that it allows me to create a groove box where a _lot_ of elements are controlled by the Nerd, so I can leave it and come back, but also work on multiple ideas that I can switch between quickly without repatching, so Nerd is acting in part as a massive preset controller. It takes away some of the spontaneity of starting with an unpatched setup, but I have a "b" rack for that and for doing more elemental synthesis.
Adding the mapping matrix screen has created an entire rabbit hole -- nay, rabbit metaverse -- that is incredibly useful and extends how one can think about the Nerd. It also adds tremendous complexity. Perhaps some day I'll have a single consistent template for all my projects, but the matrix offers so much flexibility that I can't really imagine not using it differently for different works. But that takes the user way beyond menu diving -- there isn't even a menu for one's own matrix code -- so in a few projects I've found myself coming back and then wondering why I'm still hearing a note repeating, or jumping back in the timeline unexpectedly. I'm not talking about debugging, but rather use -- you create something interesting with the mapping matrix that is unique to a project, but different projects use the matrix differently, and so I'm no longer learning an instrument but creating a mutant instrument per composition. A lot of complexity!
What I'm missing -- from before the mapping matrix arrived but even more so now -- is a documentation function. Documentation for code is important, especially for teams, or (in my case) for a time later when you've stepped away from something, and you just need to refresh why you made certain choices or how something works.
To be clear, I recognize that Nerd should NOT turn into Evernote.
But is there a way of doing documentation productively? First, for memory and UI reasons I assume that it would have to rely on external files loaded directly to the screen buffer, ie not saved to working memory. Similarly, I'm guessing the texts should not be in the project file but rather pointers to discrete associated files on the SD. I don't know the details of how you're loading from SD, and I know that saving project files pauses playback, so if there's no easy way to avoid blocking on read we can stop right here.
If, on the other hand, there's a lazy/async way of loading, at human (not tick) speed, I would suggest an addition to the FX column which would allow one to specify a file. DOCU 001 would be doc1.txt on the sd in the project folder, DOCU 002 doc2.txt, etc. Hitting OK on the FX value would load the text to view, up and down arrows could scroll notes and left and right scroll between texts. Shift OK would open a text to edit.
I could personally imagine using the NERD arcade-game-style text entry system to write quick texts, but if the texts were saved as text files that would be amazing -- one could edit them externally and load them to the SD, or if using a qwerty keyboard that could accept text in this mode as well. (I will refrain from requesting VIM bindings ) The key point is that reading is somewhat time critical (should load in a second or two), but writing is not. A block on write would be acceptible.
An example text I can imagine leaving my future self:
"CVin1 mappedd -> euclid hats, CVin2 -> vibrato on EnsOsc, CVin4 max repeats island"
Sure, I can read the mapping matrix and follow patch cables to remind myself of all of this, but it takes time and cognitive energy that I may not be able to spare as I'm loading the next song for a performance.
Similarly, yes one could take and view notes all on a phone. I have made spreadsheets of connections, made notes on my phone, etc. But I'm already looking at the nerdscreen, and again, pausing to look away and manipulate something else is suboptimal.
Finally, while a single text per song would be a huge help, I like the idea of up to 256 notes/project, as then the texts can be specific to parts of a project. I could see having a basic refresher DOCU text at the top of the song, then if the mapping matrix or interactions get tricky in a particular island, I'd have a more specific text adjacent to that island.
Nerd (especially with other modules) is approaching the complexity of a Max or a Supercollider, and it deserves to have documentation! Thanks for coming to my Nerd Talk.
As background to my user case, I have a lot of obligations and there are long periods where I can't touch music. Nerdseq has been tremendous in that it allows me to create a groove box where a _lot_ of elements are controlled by the Nerd, so I can leave it and come back, but also work on multiple ideas that I can switch between quickly without repatching, so Nerd is acting in part as a massive preset controller. It takes away some of the spontaneity of starting with an unpatched setup, but I have a "b" rack for that and for doing more elemental synthesis.
Adding the mapping matrix screen has created an entire rabbit hole -- nay, rabbit metaverse -- that is incredibly useful and extends how one can think about the Nerd. It also adds tremendous complexity. Perhaps some day I'll have a single consistent template for all my projects, but the matrix offers so much flexibility that I can't really imagine not using it differently for different works. But that takes the user way beyond menu diving -- there isn't even a menu for one's own matrix code -- so in a few projects I've found myself coming back and then wondering why I'm still hearing a note repeating, or jumping back in the timeline unexpectedly. I'm not talking about debugging, but rather use -- you create something interesting with the mapping matrix that is unique to a project, but different projects use the matrix differently, and so I'm no longer learning an instrument but creating a mutant instrument per composition. A lot of complexity!
What I'm missing -- from before the mapping matrix arrived but even more so now -- is a documentation function. Documentation for code is important, especially for teams, or (in my case) for a time later when you've stepped away from something, and you just need to refresh why you made certain choices or how something works.
To be clear, I recognize that Nerd should NOT turn into Evernote.
But is there a way of doing documentation productively? First, for memory and UI reasons I assume that it would have to rely on external files loaded directly to the screen buffer, ie not saved to working memory. Similarly, I'm guessing the texts should not be in the project file but rather pointers to discrete associated files on the SD. I don't know the details of how you're loading from SD, and I know that saving project files pauses playback, so if there's no easy way to avoid blocking on read we can stop right here.If, on the other hand, there's a lazy/async way of loading, at human (not tick) speed, I would suggest an addition to the FX column which would allow one to specify a file. DOCU 001 would be doc1.txt on the sd in the project folder, DOCU 002 doc2.txt, etc. Hitting OK on the FX value would load the text to view, up and down arrows could scroll notes and left and right scroll between texts. Shift OK would open a text to edit.
I could personally imagine using the NERD arcade-game-style text entry system to write quick texts, but if the texts were saved as text files that would be amazing -- one could edit them externally and load them to the SD, or if using a qwerty keyboard that could accept text in this mode as well. (I will refrain from requesting VIM bindings
) The key point is that reading is somewhat time critical (should load in a second or two), but writing is not. A block on write would be acceptible.An example text I can imagine leaving my future self:
"CVin1 mappedd -> euclid hats, CVin2 -> vibrato on EnsOsc, CVin4 max repeats island"
Sure, I can read the mapping matrix and follow patch cables to remind myself of all of this, but it takes time and cognitive energy that I may not be able to spare as I'm loading the next song for a performance.
Similarly, yes one could take and view notes all on a phone. I have made spreadsheets of connections, made notes on my phone, etc. But I'm already looking at the nerdscreen, and again, pausing to look away and manipulate something else is suboptimal.
Finally, while a single text per song would be a huge help, I like the idea of up to 256 notes/project, as then the texts can be specific to parts of a project. I could see having a basic refresher DOCU text at the top of the song, then if the mapping matrix or interactions get tricky in a particular island, I'd have a more specific text adjacent to that island.
Nerd (especially with other modules) is approaching the complexity of a Max or a Supercollider, and it deserves to have documentation! Thanks for coming to my Nerd Talk.
