On Wed, Aug 31, 2011 at 9:03 AM, John Cowan <cowan@mercury.ccil.org> wrote:
> Arthur A. Gleckler scripsit:
>
>> If some implementation includes a compatible superset of
>> `define-record-type' that uses the same name, it would be useful to
>> be able to distinguish it from the standard one by using a different
>> module name.  Putting the standard one in a module makes it easier to
>> manage this.
>
> Only trivially so.  Generative types are something really new in R7RS
> compared to R5RS (though old in implementations), and implementations
> shouldn't be licensed to omit them.

We've already agreed that this does not constitute an
argument.  The draft still needs to be updated to state
that modules in the report may only be omitted in
extreme circumstances (I thought I had done this).

>> Sure it's a revision of R6RS.  It's just that the edits are major.
>
> Yeesh.  And the U.S. Constitution is a revision of the Articles of
> Confederation, by the same token.

R7RS is a revision of both R5RS and R6RS (we've taken text
directly from both).  To say it is closer to one or the other is
irrelevant - it is the seventh revision of the Scheme reports,
and to call it otherwise is confusing.

--
Alex