The choice of Scheme compiler

[Necr0x0Der]

Chicken Scheme looks ok, but other choices can be considered carefully in order to proceed further. There are other choices such as Dr.Racket or Guile with different approaches (e.g. compilation to a VM rather than to a shared library). We need to create a list of suitable Scheme compilers and analyze them in accordance with a number of properties, e.g.

• speed of execution of compiled code (can be measured without integrating to MeTTa)
• speed of compilation
• flawless integration to MeTTa, i.e. will MeTTa code with compiled functions and grounded atoms be executable (like it is done for Chicken Scheme through textual representations and eval), or are there pitfalls
• platform support and easiness of deployment (in particular, possibility to install by metta-morph itself without user actions; or possibility to install as a Python package)
• availability of libraries and possibility to install them and make accessible from MeTTa without user actions
• availability of API (besides CLI) for compilation either in Scheme itself or in one of C++/Rust/Python
• compilation control and other features: compilation to shared libraries or/and RAM, detection of the necessity to recompile stuff between different runs, compilation error handling, etc.
• support for type-checking [optional?]
• ...

[patham9]

A first summary regarding Ambivalence and Matchable (the major workhorses of MeTTamorph) in other Scheme implementations:

Ambilavence in other Scheme implementations: In October it was introduced to Gerbil (std/amb): https://github.com/mighty-gerbils/gerbil/blob/master/CHANGELOG.md In Racket: https://planet.racket-lang.org/package-source/murphy/amb.plt/1/0/planet-docs/amb/index.html In Guile, usable through a library on Gitlab: https://gitlab.com/tojoqk/guile-amb I will soon try whether the superpose macro for Chicken Scheme also works (or can adapted to work) with these amb implementations, and whether the depth-first search strategy matches.

Regarding Matchable: that one initially was on track to get standardized: https://srfi.schemers.org/srfi-204/srfi-204.html This however did not happen, but the library is written in a portable way, written with the hygienic syntax-rules macro system, so will also work with the other Schemes with just minor modifications. The source for Chicken Scheme is here: https://code.call-cc.org/svn/chicken-eggs/release/5/matchable/trunk/ and the package description: http://wiki.call-cc.org/eggref/5/matchable

[patham9]

Some findings since last time:

• Gerbil Scheme's amb implementation is a bit non-standard in Lisp/Scheme world as it demands an amb-begin application before each amb/amb-collect usage, which makes superpose implementation more difficult as it demands additional syntactic transformations which could unfortunately become non-trivial.
• Gerbil Scheme does not allow calling POSIX fork and pipe functions directly, which means a parallel superpose would need a different approach than currently in Mettamorph.
• Guile's amb misses a few functions such as amb1 to create ambivalence for each list item, but this should be fixable.
• C interop with Guile and Racket seems like it will be slightly more complex than with Chicken Scheme and Gerbil.

I guess considering these points we can at least already eliminate Gerbil from the options?

[Necr0x0Der]

Yes, Gerbil can be discarded.