aboutsummaryrefslogtreecommitdiff
path: root/readme.md
diff options
context:
space:
mode:
Diffstat (limited to 'readme.md')
-rw-r--r--readme.md16
1 files changed, 3 insertions, 13 deletions
diff --git a/readme.md b/readme.md
index 4c3d25d..ee92e61 100644
--- a/readme.md
+++ b/readme.md
@@ -11,26 +11,16 @@
of functional programming languages
- Based on bleeding-edge research results
- Exponentially big normal forms of the family $e_n=λx.c_nωx$, where
- $ω:=λx.xx$ and $c_n$ denotes the $n$th Church numeral, consume only
+ $ω:=λx.xx$ and $c_n$ denotes the $n$-th Church numeral, consume only
a linear in $n$ amount of memory and are computed in linear
time\[0\]
-## Why C?
-
-You might realize that the reduction transitions are of a very
-functional nature and aren't that obvious to implement in non-functional
-languages.
-
-I used C because **(1)** most functional languages use automatic memory
-management (often using periodic garbage collecting) which unnecessarily
-slows down the transitions, **(2)** there aren't that many efficient
-lambda calculus reducers in C out there, **(3)** it's a fun challenge
-and **(4)** I like C.
-
## Libraries
- [CHAMP](https://github.com/ammut/immutable-c-ollections) \[MIT\]:
Underrated efficient hash array mapped trie
+- [BDWGC](https://github.com/ivmai/bdwgc) \[MIT\]: Boehm-Demers-Weiser
+ Garbage Collector
## Research