[Haskell-cafe] Equality constraints in type families

[Hugo Pacheco]

The reason for the braces in type families is because type indices are
treated differently than normal parameters. I don't think this should be
adopted for type synonyms either.

Cheers,
hugo

On Thu, Mar 27, 2008 at 9:48 PM, Wolfgang Jeltsch <
g9ks157k at acme.softbase.org> wrote:

> Am Donnerstag, 27. März 2008 22:43 schrieb Wolfgang Jeltsch:
> > Am Mittwoch, 26. März 2008 03:07 schrieb Hugo Pacheco:
> > > > The extra syntax has its advantages (more local information) and
> > > > disadvantages (more clutter).  We weren't convinced that we need the
> > > > extra syntax, so left it out for the moment. However, this is
> > > > something that can always be changed if experience shows that
> programs
> > > > are easier to understand with extra syntax.  It doesn't affect the
> > > > type theory and is really a pure language design question.  I'd be
> > > > glad to hear some more opinions about this matter.
> > >
> > > I would go for the braces as Claus suggested,
> >
> > I would do so, too.
>
> Hmm, but then we should also introduce braces for ordinary type synonyms:
>
>    type ReaderWriterT m = ReaderT (WriterT m)
>
>    x :: {ReaderWriterT IO} Char
>
> > […]
>
> Best wishes,
> Wolfgang
> _______________________________________________
> Haskell-Cafe mailing list
> Haskell-Cafe at haskell.org
> http://www.haskell.org/mailman/listinfo/haskell-cafe
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.haskell.org/pipermail/haskell-cafe/attachments/20080327/3aaa311d/attachment.htm