For a thorough and accessible introduction to objective Bayesian hypothesis testing, we encourage learners to explore @RyanBurnML's explainer, which balances theory and practical considerations. https://t.co/MYYYHMIFW2
@crascit This is what statically linking together with an export map solves. The export map forces the SO to use the symbol versions that are statically linked into the binary. Here's an example of how it works: https://t.co/Ivff5mjL20
@crascit Statically linking libstdc++ largely solves the problem.
If the binary is an executable, you can even statically link the standard c library if you use libmusl.
If it's a SO, you can use an export map and remap glibc symbols to portable versions: https://t.co/39n99KszMq
@seanbax How would `auto&& res = f(some_val)` get translated using this new syntax if it was in a generic context where you don't know if f(some_val) returns and int or an int&?
@seanbax If I write template <class T> void f(forward T x), what does T get deduced as in f(MyType{}) and MyType t; f(t)? Does it follow 708 and deduce T as MyType in both cases? or does it deduce as MyType and MyType&?
@seanbax@Love2Code Why are coroutines unusable? https://t.co/SOZTJssDVm, the async framework https://t.co/HZqb58nwp8 and https://t.co/EVRu0dRaYr are built on, makes extensive use of coroutines: https://t.co/8RB6ztzfRA
@Windows - What's going on with https://t.co/eEgJ9cRe5F? I switched from mac to windows hoping to be more productive with WSL2, and I definitely don't want to be spending time debugging DNS issues.
@seanbax Yeah, I don't have a very deep understanding of it. But I thought a forward template might deduce T without the reference qualifiers ... which seem like it could help with the forwarding issue