TypeSynonymInstances.md

+
+
+
+# Type Synonym Instances
+
+
+## Brief Explanation
+
+
+
+Haskell 98 permits only type constructors defined using `data` or `newtype` in instance heads.
+
+
+
+The proposal is to allow type synonyms (fully applied, as ever) in instance heads. These would be fully expanded before any other restrictions on instance heads were checked (see [FlexibleInstances](flexible-instances) and [OverlappingInstances](overlapping-instances)).
+
+
+
+Not very useful without either [FlexibleInstances](flexible-instances) or [UndecidableInstances](undecidable-instances).
+
+
+## References
+
+
+- [ Instance declarations](http://www.haskell.org/onlinereport/decls.html#instance-decls) in the Haskell 98 Report
+- [ Instance declarations](http://www.haskell.org/ghc/docs/latest/html/users_guide/type-extensions.html#instance-decls) in the GHC's User's Guide.
+
+## Tickets
+
+
+
+<table><tr><th>[\#70](https://gitlab.haskell.org//haskell/prime/issues/70)</th>
+<td>allow TypeSynonymInstances</td></tr></table>
+
+
+
+## Pros
+
+
+- Offered by GHC and Hugs for several years.
+- Synonyms can cut large instance heads to a manageable size, e.g. when building a monad using monad transformers.
+- Allows a library author to generalize types while retaining source compatibility with old client instances. e.g. replacing
+
+  ```wiki
+  data T a = ...
+  ```
+
+  with
+
+  ```wiki
+  data GenT param a = ...
+  type T = GenT Default
+  ```
+
+## Cons
+
+
+- Since constraints on the instance head refer to the expanded type, errors can be more obscure.
+
+## Comment
+
+
+
+This change would make type synonyms a bit more like newtypes. [NewtypeDeriving](newtype-deriving) makes newtypes a bit more like
+type synonyms since they can be made to inherit properties of the underlying type, albeit (usefully) selectively.
+Having both type synonyms and newtype is a bit confusing, and the more alike they are, the more confusing it is.
+I guess looking for a single replacement is not an option for Haskell', but at least we should consider if
+automatic derivation for newtypes does not supersede this one as it would make it significantly less tiresome to
+introduce a newtype for purposes of abbreviation.
+
+