I'd venture to say.. Prolog and Shen. Specifically, I am implementing a modern Prolog in Rust that will encompass the results of some more recent (ie. current century) research within the logic programming/Prolog community. It's here:

http://github.com/mthom/rusty-wam

Is this more about new language constructs or a better run-time execution/solver ?

New language constructs. The 'WAM' in the name refers to the Warren Abstract Machine, one of the more efficient ways to implement Prolog in a procedural language.

Also, using the SICStus interface and semantics for attributed variables should make for some very general constraint solvers, yes.

What do you think about answer-set programming vs more general solvers like SICStus.

I don't know enough about answer set programming to give much of an answer. Sorry.

Mercury is a promising logic programming language that's under development https://github.com/Mercury-Language/mercury

Tutorial available here (the "Ralph Becket Murcury tutorial" pdf) http://mercurylang.org/documentation/documentation.html

The Mercury compiler is written in Mercury, which is syntactically derived from Prolog. The original compiler was bootsrapped by stripping the non-Prolog parts from the code and executing it on NU-Prolog. The compiler first successfully compiled itself in 1995, after which NU-Prolog was no longer needed.

And just to tie this back to the original topic, Leon Sterling (first author of The Art of Prolog) was the head of the Department of Computer Science at the University of Melbourne when the development of Mercury began there.

Hmmm... perhaps Prolog and Lisp?

Like in The Reasoned Schemer? https://mitpress.mit.edu/books/reasoned-schemer

As in, "fill-in-the-spaces will soon free us from the tyranny of imperative programming"...

> Hmmm... perhaps Prolog and Lisp?

"when the proof search accidentally returns the trivial case"

"Prolog for the 21st century" is Erlang.

Not at all. All they have in common are comma and dot, and even the semantics is different for these.