[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Scheme-reports] Formal comment: The denotational semantics
On Thu, Jul 5, 2012 at 4:25 PM, John Cowan <cowan@x> wrote:
> Michael Sperber scripsit:
>> PS: This is a general issue with the draft - the R6RS document contains
>> many improvements of the description of the language that are not about
>> changes in the language. (To name two other prominent examples: the
>> distinction and relationship between numbers and the objects that
>> represent them,
> I don't disagree in general, though it is very hard to locate these
> improvements to incorporate them into R7RS-small. But though I admit
> that reasonable people can reasonably disagree, I find the pervasive
> use of "number object" in R6RS verbose, pedantic, and wearying in the
> extreme during multiple re-readings: I think it adds nothing. I also
> note that the analogous phrases "vector object", "procedure object",
> and "pair object" are not used in R6RS, although vectors, procedures,
> and (ordered) pairs are just as much mathematical objects as numbers are.
>> Especially embarrassing is the pointer to the Indiana Scheme
>> repository, which has not been maintained for something like 14 years.)
> That objection is well-taken: I just removed it from the trunk yesterday.
I'd prefer to keep it in. It has a lot of useful code
and code of historical interest. Many classic papers
describe or use code that is only easily available in
the Indiana or CMU repositories (for example, I know
of nowhere else to download the Rabbit compiler).
Furthermore, the code tends to actually work in any
modern Scheme, which is more than I can say for any
of the implementation-specific repositories.
schemers.org is a more important site, but it
doesn't actually host any code.
We can move it to the end of the list and mention
it is largely of historical interest, but let's keep it.
Scheme-reports mailing list