@[email protected]

Reached what I think is the absolute lowest denominator for making objects for ossia / Max / Pd / VSTs / <anything you can think of which looks like a graph>: the object on the left is generated automatically from the code on the right with just a C++20 compiler #intermedia

@Researchirel @MaynoothChem @debraireland @MU_MathsStats @EnvironnementQc Thibaud Keller @MusicMaynooth also receives €124,000 to advance hardware interfaces for music technology, working with French company ossia.io

@theycallmeken @zachlieberman i'll bring my quneo to the grave

@zachlieberman Check out keithmcmillen.com if you’re looking for MPE gear that’s roadworthy

What are your favorite (midi ?) controllers for live performance. I’m trying to avoid cheap feeling knobs…. Mostly looking for sliders or knobs to control parameters.

Implemented a couple new backends recently in #libremidi: - Network I/O through #OSC - Standard logic for computer keyboard input mapping Testing appreciated! github.com/celtera/librem… #midi #cpp

Very glad that the GStreamer conference happens in Montreal this year! At the SAT we are showing our appreciation for the critical open-source work done by GStreamer contributors, by becoming a gold sponsor of the conference and organizing a networking event tuesday. RSVP link:

@zerolehero @satnam6502 La même que tout le monde: requires (N % 2 == 0) en c++20, c'est vraiment fait pour ça 😁

@satnam6502 @jcelerie au cas où tu ais une suggestion?

I'm looking for help with C++ (yes, really). I would like to know if there is any way that I can statically ensure a function like ilv (show below) takes an even length array as an argument and returns an even length array as a result at compile time. I know I can use a static_assert inside the body of the ilv function, but I want to convey the even constraint in the type signature. I appreciate the C++ compiler in the code below infers a size M for the template matching for the size of the array, and then fails to work out that M = 2*N. If there is a way to statically enforce an array having an even length in the type then I'd love to know about it! Thank you kindly. Normal Haskell Satnam service will resume shortly.

@matthew_meadows @janetacarr Median salary in France is ~36000$ pre tax. 95000$ you're a 1%er.

@janetacarr These salaries seem quite low even outside of the tech hub markets.

Stackoverflow dev survey 2024: Look who's paying the bills!

Our team member Guillaume Riou at the @SATmontreal innovation lab developed a fulldome shader for @godotengine compatible with the SAT's 18-meters, 210° dome. Free and open-source full-dome render pipeline for all, yay :) #fulldome #mediaarts #creativetech #MTL #GodotEngine

ossia score 3.2.4 released with many packaging news: - Now available on Flathub: flathub.org/apps/io.ossia.… (x86_64 and aarch64) - Now available on Nix (unstable channel): nix-env -i ossia-score (x86_64, aarch64 coming in a PR) - Added BSD to CI :) github.com/ossia/score/re…

Hi folks! We're starting a formal user study on ossia score, to improve the UI / UX affordances. If you've ever used ossia for audio / visuals / IoT / media art and can spare 10 minutes, please fill out forms.gle/HWpozsvApUz9JW… !

@supahvee1234 Most likely the non-negative quantity is also not supposed to behave such that x-y for x = 0 and y = 1 gives 4 billion

@supahvee1234 With a type that represents this invariant (which likely is implemented as a signed integer). Most likely the full range of an int even 31 bit is likely invalid, e.g. of it's for representing an amount of items displayed in a GUI app or a size of some geometry in pixels.

How does the "make invalid states unrepresentable" crowd cope with the widespread advice of using signed integers even for quantities that are never supposed to be negative? I am in that crowd, BTW. #cpp #c

@FFmpeg @ossia_io score is an open-source audio/video/any media real-time sequencer which uses ffmpeg for video & audio encoding & streaming :)

Show us interesting things you have done with FFmpeg and we'll share them with all our followers

@ossia_io score reached the 200k downloads milestone, a bit less than a year after the first 100k which took a decade :') thanks for your support friends 🫶

La SAT est ravie de participer au prochain Open Hardware Summit qui aura lieu le 3 et 4 mai 2024 à Montréal ! Inscriptions : 2024.oshwa.org

@m_ou_se How would you handle a native GUI where an operation causing some panic deep down should never crash the app but instead just show a dialog to the user like "error: operation canceled" and let them continue with other GUI actions ?

What if Rust only had panic="abort"? (Instead of panic="unwind", today's default.) It could simplify some parts of the language and prevent subtle bugs. Do you absolutely need catch_unwind? If yes, please leave a reply! (Panic hook and backtrace etc. will still work of course!)

@ossia_io score now available directly from MSYS2 packages! :-) packages.msys2.org/base/mingw-w64…

@tanzislam @vzverovich If you use X11 and defined some themeing the C preprocessor runs on every boot to parse your .Xresources file

@vzverovich I've used the C preprocessor on Dockerfiles. Works pretty neatly.

We have metaprogramming at home

Looking for testers: as of the latest commit, libremidi (github.com/jcelerier/libr…) has primitive support for sending & receiving MIDI 2 UMP messages through the Alsa sequencer. (Linux6.5+). Check out midi2_echo.cpp example :)