Comment by AnonHP
Comment by AnonHP 20 hours ago
This is from 2003 (December, so it's about 21 years ago). Is it still relevant?
Comment by AnonHP 20 hours ago
This is from 2003 (December, so it's about 21 years ago). Is it still relevant?
Mike Pall (LuaJIT author/maintainer) is not happy with some of the design decisions and (breaking) changes:
https://www.freelists.org/post/luajit/Port-bitop-to-53,1
The biggest issue certainly is the addition of integer types.
There is design tension between the language designer and the VM implementation. AFAIK some features post 5.1 make performant VM implementations difficult. Lua is originally a configuration language where performance didn't matter all that much. Because Luajit is so fast much more of the application code can be done in the scripting environment so now the performance matters much more than it did.
So the spectrum is from newer language and slower to older language and faster. If performance is an issue the cost of the newer language features could be that more of the application code has to be written in C++ instead of Lua - in that context the Lua language shouldn't be considered independent of the host language. Performance matters to me so increase in C++ code would not be worth the newer language features. Using Rust instead of C++ as the host language might change the landscape again since Rust is so much more ergonomic than C++.
> Their defence is always that they do not follow semver.
"Defence"? Seems like the correct answer and a fact, Lua doesn't do semver, so when there are major changes between 5.3 and 5.4, it's not illogical or actually breaking semver, as they don't follow it.
Plenty of projects don't follow semver, and that's perfectly fine.
Yes. The differences between 5.0 and 5.1 are small and I do not recommend using newer versions, regardless.