From 4a3a6ebd7044533d675173fde3562285331ab958 Mon Sep 17 00:00:00 2001 From: hauntsaninja Date: Tue, 26 Sep 2023 00:11:26 -0700 Subject: [PATCH] ecosystem --- docs/source/typing_anti_pitch.rst | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/docs/source/typing_anti_pitch.rst b/docs/source/typing_anti_pitch.rst index 645c4b6e0..f15f6a3de 100644 --- a/docs/source/typing_anti_pitch.rst +++ b/docs/source/typing_anti_pitch.rst @@ -47,10 +47,11 @@ in Python:: `allow gradual adoption of static type checking `_, the total cost of adding type annotations to a large extant codebase can be prohibitive. -* Your library does enough dynamic things that type checking would be unlikely to help your users. - Either you would have to spend a lot of effort redesigning your library in ways that static type - checkers could better understand, or you'd have to invest a lot of effort into figuring out clever - type annotations to twist the arms of type checkers. +* Your application uses a particularly dynamic framework or your library does enough dynamic things + that type checking would be unlikely to help your developers and users. Migrating application + frameworks could be costly. Either a) redesigning your library in ways that static type checkers + could better understand or b) figuring out clever type annotations to twist the arms of type + checkers would take a lot of effort. * Your codebase has suffered at the hands of `Hyrum's Law `_ and all possible observable behaviour is depended on. In order to avoid false positives for your