every compiler project expands until it creates its own version of TableGen. after that it is swiftly replaced with something that is "not as difficult to use", and the cycle continues
i recently got a new toy! it's a (civilian) miniature drone helicopter directly inspired by the (military) flir black hornet drone helicopter
it doesn't have a FLIR sensor but otherwise seems (from declared characteristics) very similar in design and handling. also it costs $100 instead of $100,000
did you know: maximum-sized USB HS BULK packets where the payload is all-ones are about 16% longer to transmit than the ones where the payload is all-zeroes or random?
this is easily measurable using the #GlasgowInterfaceExplorer with a slightly modified benchmark applet:
It looks like the Wasm GC extension has been finalized and is starting to be widely implemented. It seems pretty good from the spec--it's really more of an extension for graphs of minimally structured objects (in an opaque, managed heap outside of Wasm linear memory) that so happens to also be what you need to implement GC in a Wasm engine.
@pervognsen reasonable but unfortunate. interior pointers lighten the load on the gc a lot. i guess the source language could have by-value-only types like C# 'struct'
i remember being a high schooler who was like "i wanna make a pcb owo" and then running headfirst into CAD software after having never encountered one before
inspired by @stargirl, a "no stupid questions" thread!
ask me any question about embedded development, FPGAs, programming languages, life, angel girls, or anything else you think I know about, and I'll do my best to answer them
the reason i am excited about WebAssembly is because it's the first interesting computer architecture designed in the last 30 to 40 years (unlike e.g. RISC-V)
people who have not looked at how Wasm works usually don't get it, which makes me sad