wh40ksim VS scryer-prolog

Here's some stuff I've written in Prolog, some for my own enjoyment, one for my degree project.

Most of the benefits I found come down to two things:

a) Prolog, like the various kanrens, is a relational language so a program is effectively a database. There's no need to do anything special to glue together a data layer and a logic layer, because you have both written in Prolog.

b) Prolog's declarative style makes translating rules and directives to code a breeze. The three projects below are all games and benefit heavily from this feature. I

1. Warhammer 40K simulation:

https://github.com/stassa/wh40ksim

Runs simulations of combat between WH40k units.

2. Gleemin, a Magic: the Gathering expert system:

https://github.com/stassa/Gleemin

Doesn't work anymore! Because backwards compatibility. Includes a) a parser for the rules text on M:tG cards written in Prolog's Definite Clause Grammars notation, b) a rules engine and c) a (primitive) AI player. The parser translates rules text from cards into rules engine calls. The cards themselves are Prolog predicates. Your data and your program are one and now you can also do stuff with them.

3. Nests & Insects, a roguelike TTRPG:

https://github.com/stassa/nests-and-insects

WIP! Here I use Prolog to keep the data about my tabletop rpg organised, and also to automatically fill-in the character sheets typeset in the rulebook. The Prolog code runs a character creation process and generates completed character sheets. I plan to do the same for enemies' stat blocks, various procedural generation tables, etc. I also use Prolog to typeset the ASCII-styled rulebook, but that's probably not a good application of Prolog.

You asked about "logic programming" in general and not miniKanren in particular. I haven't actually used miniKanren, so I commented about the logic programming language I've used the most, Prolog. I hope that's not a thread hijack!

All three of the projects above are basically games. I have more "serious" stuff on my github but I feel a certain shortfall of gravitas, I suppose.

thank you! the scryer community deserves much of the credit too. everyone is welcome and encouraged to join us at https://github.com/mthom/scryer-prolog! some exciting plans in the pipe

Djot is the markup syntax that is used for the documentation of Scryer Prolog, using a parser written in Prolog:

https://github.com/aarroyoc/djota

It works well so far. One of the few limitations I noticed so far pertains to the formatting of tables. For instance, consider the table used in library(format) to describe control sequences:

https://github.com/mthom/scryer-prolog/blob/b0566e41503a6c8d...

It contains several entries that span multiple lines, yet are meant to denote only a single row of the table, such as:

    % | `~Nr`    |  where N is an integer between 2 and 36: format the            |