On Fri, Nov 16 2012, Alex Shinn wrote:

> On Fri, Nov 16, 2012 at 5:28 PM, Helmut Eller <eller.helmut@gmail.com> wrote:
>
>     The phrase "it is an error" that is used so often in the report is
>     rather misleading.  Usually, when I read "it is an error" I expect that
>     the problem will be detected either at compile-time or at runtime.
>     However the meaning of "it is an error" in R7RS is the same as "it is
>     unspecified".  I don't understand why the word "error" should be used in
>     such cases as it has nothing to do with the 'error procedure or
>     error-objects.
>
>     Especially bizarre is the interpretation that implementations can
>     provide extra functionality by exploiting these "error situations"
>     without actually raising an exception.
>
> This is the convention for all the Scheme reports.

R5RS does not have exceptions and in particular did not have an error
procedure.  A quick search reveals that R6RS does not use the phrase "it
is an error" once.  I don't know the situation of previous documents.

> It is documented in the overview section 1.3.2.

That doesn't necessarily make it a good idea.

> If you're unfamiliar with previous reports, it would
> be wise to read the entire R7RS report before
> criticizing it.

Is that your way to shut down criticism?  Good.  It worked.

Helmut

_______________________________________________
Scheme-reports mailing list
Scheme-reports@scheme-reports.org
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports