5 ms·

Everything is based on composition, which makes it easy to build computations up like building blocks. Stacks are also efficient. The JVM bytecode is actually a

by cultus 8y ago

Everything is based on composition, which makes it easy to build computations up like building blocks. Stacks are also efficient. The JVM bytecode is actually a stack language, for example. There's a newer one called Kitten that is statically typed and uses term rewriting to allow for more normal syntax when you want. It's pretty cool. Too bad none of these languages will ever take off.

bgongfu 8y ago
And since composition is seamless, and they're relatively trivial to implement; they make great glue/DSL/process description languages. Take off as in replace C++ or Java, probably not. I see them more as a complement to existing languages, a more convenient way to glue the pieces together. Even wrote my own [0] to see how far it's possible to push that idea in C. [0] https://github.com/basic-gongfu/cixl https://github.com/basic-gongfu/cixl
- rurban 8y ago
  Interesting. But does cixl really do refcounting on stack values? It should only be done on aggregates or objects reaching out into the heap, not on int or float primitives on the stack.
  - bgongfu 8y ago
    Primitive types such as booleans, int, floats, times etc. are passed by value. Reference counting is only used for heap allocated values and types that themselves reference values, like pairs and tables. https://github.com/basic-gongfu/cixl/blob/master/src/cixl/box.h https://github.com/basic-gongfu/cixl/blob/master/src/cixl/bo...
Someone 8y ago
”Stacks are also efficient. The JVM bytecode is actually a stack language, for example.” The advantage of stack-based is that instructions need not tell where there arguments live, making them smaller (if you have 16 integer registers, for example, a simple “add register I to register j, store result in register k” instruction needs 12 bits to encode the values of i, j, and k. That’s quite a lot if your instructions are 16 bits) The disadvantage is that you have to move data to where instructions expect to see it before you can do computations. That makes your code larger and slower. Last I know needing more instructions to do the same is more of a problem than having smaller instructions solves (https://www.usenix.org/events/vee05/full_papers/p153-yunhe.pdf https://www.usenix.org/events/vee05/full_papers/p153-yunhe.p...), but that’s over 10 years old.
vanderZwan 8y ago
> Too bad none of these languages will ever take off. I'm sure they will at least inspire some good features in others.
munificent 8y ago
> Stacks are also efficient. The JVM bytecode is actually a stack language, for example. Being stack-based doesn't help JVM efficiency. Modern JVM's take that bytecode and immediately translate it to some other non-stack-based form and then compile it. JVM bytecode is effectively a serialization format for syntax trees. (Stacks are a nice notation for serializing a tree.)

by cultus 8y ago

bgongfu 8y ago

And since composition is seamless, and they're relatively trivial to implement; they make great glue/DSL/process description languages. Take off as in replace C++ or Java, probably not. I see them more as a complement to existing languages, a more convenient way to glue the pieces together. Even wrote my own [0] to see how far it's possible to push that idea in C. [0] https://github.com/basic-gongfu/cixl https://github.com/basic-gongfu/cixl

rurban 8y ago
Interesting. But does cixl really do refcounting on stack values? It should only be done on aggregates or objects reaching out into the heap, not on int or float primitives on the stack.
- bgongfu 8y ago
  Primitive types such as booleans, int, floats, times etc. are passed by value. Reference counting is only used for heap allocated values and types that themselves reference values, like pairs and tables. https://github.com/basic-gongfu/cixl/blob/master/src/cixl/box.h https://github.com/basic-gongfu/cixl/blob/master/src/cixl/bo...

Someone 8y ago

”Stacks are also efficient. The JVM bytecode is actually a stack language, for example.” The advantage of stack-based is that instructions need not tell where there arguments live, making them smaller (if you have 16 integer registers, for example, a simple “add register I to register j, store result in register k” instruction needs 12 bits to encode the values of i, j, and k. That’s quite a lot if your instructions are 16 bits) The disadvantage is that you have to move data to where instructions expect to see it before you can do computations. That makes your code larger and slower. Last I know needing more instructions to do the same is more of a problem than having smaller instructions solves (https://www.usenix.org/events/vee05/full_papers/p153-yunhe.pdf https://www.usenix.org/events/vee05/full_papers/p153-yunhe.p...), but that’s over 10 years old.

vanderZwan 8y ago

> Too bad none of these languages will ever take off. I'm sure they will at least inspire some good features in others.

munificent 8y ago

> Stacks are also efficient. The JVM bytecode is actually a stack language, for example. Being stack-based doesn't help JVM efficiency. Modern JVM's take that bytecode and immediately translate it to some other non-stack-based form and then compile it. JVM bytecode is effectively a serialization format for syntax trees. (Stacks are a nice notation for serializing a tree.)