Rust Audio Development Forum
rust-audio.discourse.group
rust-audio.discourse.group
Sure, I would have loved for this to be a Matrix server, but beggars can't be choosers. If the Rust audio devs are using Discord, that's where you have to go to find them.
They list their official Dicourse/Discord chat there. I believe this is a set of Free Software projects of a sort.
But the UX of everything Free is just horribly bad. I've written a toy softsynth for myself in Ruby and was looking for some MIDI editing software that I could use to quickly throw some notes in a piano roll and draw some curves for various CCs, and export to a .mid. But everything was incredibly cumbersome. Of the several I tried, only Ardour even allowed the control of arbitrary CCs, and I had to dig to find it, and it wanted to create an entire directory hierarchy before I could even draw a single note.
If there was going to be a Free revolution of people donating and volunteering to overthrow the proprietary software world, it would have happened a long time ago. Free software has an insurmountable design, development, and discovery problem.
If I were looking to start a community today, and actually have people join it, I'd go where the people are, and that is Discord.
Edit; my midi sequencer recommendation would be sequencer64/seq66.
Someone in the 80s probably wrote that free software has an insurmountable tooling, compiler, and operating system problem.
The surmountable hurdle is for key free software developers to learn enough about design and discovery to honor and encourage the very few people who currently show up to do that work. If they made the experts feel like rock stars I'm sure more would turn up to do work for free.
I offer as evidence the number of free software devs who work on free software tooling, compilers, and operating systems as evidence of this.
The usability and accessibility of a platform such as Discord simply can't compare with FOSS alternatives.
The chat started off as a Telegram chat for some of us working on the Rust VST crate, but as we started discussing wider Rust audio topics, we found that the lack of "topics"/channels/whatever wasn't great. For better or worse, we chose Discord to migrate to, and it solved the channels issue for us.
Migrating a community like this is a tough call in the first place; there's always going to be some people that don't want to use or install yet another chat app. Our goal was to fragment the community as little as possible, and a fair amount of people at the time already used Discord for other stuff. So, Discord it was.
I think if we were migrating today I'd push harder to move to something open (probably Matrix?), but as of right now Discord has been a decent enough platform.
Bridge the entire community to Matrix.
Matrix allows rooms and communities from other services to be bridged in so that discussion can continue with people still tied to proprietary platforms. The Pine64 community has done this very well and I regularly use their chats via Matrix while others use IRC, Discord, or Telegram.
I'll have to look into how well the bridges work; I don't know if I'd be too excited if all the cross-platform messages just showed up as a bot message with the user's name prepended to the message? Plus, it seems like it would be yet another thing to pay attention to for moderation/administration purposes (make sure something doesn't break, make sure nothing nefarious isn't going on in each server, etc).
I will bring it up to the other admins/mods of the discord, though.
The Rust Embedded folks saw what was coming and chose Matrix even back when it wasn't nearly as good as it is now.
One of the things about the voting to make the choice was that a lot of people gave an informal thumbs down on alternatives to their main choice. Discord came in for a HUGE amount of thumb downs. No other option even came close.
For some reason, a lot of people fail to estimate how many people really loathe Discord.
You port WebAudio API to a compiled language or use a JS interpreter that has it?
Would be interested to try this
LabSound [0] has done something similar with C++ and WebKit. They took the WebAudio implementation and made it into a C++ framework.
Ah yup, for Servo that would make sense eh?
Well neato. I hope you do, seems useful. Audio landscape needs more options besides JUCE and iPlug2.
The TL;DR is that you make your computer spit out a certain amount of distinct values per second (like, say, 44100 floats per second, times two for stereo), and your DAC will convert that stream of numbers into an audio signal that your speakers will be able to use to produce sound.
The way I like to mess around with ideas is to grab a library that writes to WAV files and see what I can make -- a Sine oscillator at a certain given frequency is a good place to start -- and then you can load that WAV file into Audacity (or your other favorite sound editor of choice) and analyze what you just made.
More complex audio topics will require some DSP knowledge; I started teaching myself by reading http://www.dspguide.com/pdfbook.htm, but there are a decent amount of other resources to learn from too. Especially ones that focus more on the audio aspects of DSP.
There are some interesting programming challenges in the audio space; for example, if you want to pipe sound directly to your speakers (or in a plugin for a DAW), you're going to have to make your code real-time friendly. Designing an "audio graph" to hook devices up together (as if it was a modular system) can be a pretty interesting problem space when you're designing for real-time, polyphony, efficiency, etc. It's a fun topic if you're into both programming and music!