thoughts on the record update problem

Greg Weber greg at
Thu Mar 8 22:52:22 CET 2012

Since it is impossible for anyone to have read and kept in their mind
all the discussion that has gone I will fill you in as I know it. We
would like the syntax of record updates to be the same as they are
today. That is separate from this proposal. This proposal is the only
solution so far that works for updates with the desugaring hack and it
requires a special form of type annotations instead of using the
syntax we would like. Does that make sense? I take it that you agree
that we should separate the discussion of semantics from
implementation: this is a perfect example of why.

On Thu, Mar 8, 2012 at 1:14 PM, Chris Smith <cdsmith at> wrote:
> On Thu, Mar 8, 2012 at 2:09 PM, Greg Weber <greg at> wrote:
>> The semantics that will be exposed to users have already been largely
>> decide upon.
> Admittedly I haven't had time to carefully read some parts of this
> thread, and if that claim is true, then of course implementation
> should be the major concern.  But it seems unlikely that claim is
> true, since in the very same email you express what looks like a
> pretty serious concern about the semantics that will be exposed to
> users (namely, the need for a new kind of type annotation).
> --
> Chris Smith

More information about the Glasgow-haskell-users mailing list