@@ -1,8 +1,12 @@
11# effective
22
33[](https://github.com/arybczak/effective/actions?query=branch%3Amaster)
4+[](https://hackage.haskell.org/package/effective)
45
5−A simple, yet powerful extensible effects library (in the making).
6+*Note:* this is a pre-release.
7+
8+A performant, easy to reason about extensible effects library with seamless
9+integration with the existing Haskell ecosystem.
610
711Main features:
812
@@ -13,7 +17,7 @@ Main features:
13173. Correct semantics in presence of runtime exceptions (no more lost or
1418 discarded state).
1519
16−4. Seamless integration with existing Haskell ecosystem (`exceptions`,
20+4. Seamless integration with the existing ecosystem (`exceptions`,
1721 `monad-control`, `unliftio-core`, `resourcet`).
1822
19235. Effects can be defined for either
@@ -26,8 +30,7 @@ Main features:
2630
2731## Motivation
2832
29−Let's address the elephant in the room. Do we really need yet another library
30−for handling effects? There's
33+Do we really need yet another library for handling effects? There's
3134[freer-simple](https://hackage.haskell.org/package/freer-simple),
3235[fused-effects](https://hackage.haskell.org/package/fused-effects),
3336[polysemy](https://hackage.haskell.org/package/polysemy),
@@ -36,7 +39,7 @@ for handling effects? There's
3639Unfortunately, of all of them only `eff` is a promising proposition because
3740of reasonable performance characteristics (see the talk "Effects for Less"
3841linked below for more information) and potential for good interoperability with
39−existing ecosystem.
42+the existing ecosystem.
4043
4144The second point is arguably the most important, because it allows focusing on
4245things that matter instead of reinventing all kinds of wheels and is crucial for
@@ -59,7 +62,7 @@ support for them has been accepted).
5962
6063On the other hand, if support for continuations is excluded from the handler
6164monad, then the ability to define effects with non-linear control flow (such as
62−`NonDet`) is lost. I'll argue it's a small price to pay for predictability,
65+`NonDet`) is lost. Arguably it's a small price to pay for predictability,
6366because such specialized effects are needed rarely and locally, at which point a
6467dedicated, well established solution such as
6568[conduit](https://hackage.haskell.org/package/conduit),