[r6rs-discuss] Confusing taxonomy of identifiers

From: R. Kent Dybvig <dyb>
Date: Fri, 29 Jun 2007 11:34:56 -0400

> > ... an identifer may name a kind of syntax, or it may name a
> > location ...
>
> Is that meant to be an exhaustive or exclusive list?

Exhaustive, in that context. Identifiers also serve as library names.

> So there are two kinds of identifier. But then:
>
> > A variable that names a subform in a syntactic abstraction is called
> > a pattern variable
>
> Substituting the definition of "variable" we have:

> ...

Pattern variables are variables that happen to be bound to pieces of
syntax and are referenced in certain peculiar ways (i.e., within templates
and possibly followed by ellipses). The text you quote doesn't get this
across and is both confusing and misleading. It's also unnecessary:
there's really no reason to mention pattern variables where that text
appears, since they are adequately covered in the section on syntax-rules
and the chapter on syntax-case.

Thanks for pointing out the problem.

Kent
Received on Fri Jun 29 2007 - 11:34:56 UTC