[R6RS] Timeline for R6RS SRFIs
Michael Sperber
sperber
Thu Jun 2 10:15:31 EDT 2005
Marc Feeley <feeley at iro.umontreal.ca> writes:
> On 2-Jun-05, at 8:30 AM, Michael Sperber wrote:
>
>> That's why I said we can just drop all but the imperative layer for
>> consideration for R6RS, if that's what everybody wants. The nice
>> thing about it is that it isn't monolithic, after all.
>>
>
> As far as I can tell the imperative section of SRFI 68 goes beyond
> the requirements we agreed to in Boston.
Sure, but note that we never said we'd *only* cover the minimal
requirements. This was mainly in the interest of saving time and
work. Given that a lot of the work has already been done, it's not
clear to me that we should just upfront zap everything that isn't in
the minimal requirements. (Even though, as I said, I'm certainly
amenable to leaving stuff out we have trouble agreeing on.)
> It also refers to the stream layer (e.g. input-port-stream,
> set-input-port-stream!), so strictly speaking that section is not
> self contained.
Yes, and that needs to be fixed.
> I'm also concerned about backward compatibility (e.g. read-char
> seems to return an integer
That's a bug---I'll fix it.
> and the procedure eof-object? no longer exists).
I'll be happy to put it back if that's what people want, but this
seems like an aspect of the SRFI that would obviously improve the I/O
substrate (also, CALL-WITH-xxPUT-PORT), so I suggest having that
discussion on the SRFI mailing list. (Generally, we might do for the
I/O what we envision doing for R5RS---offering a
backwards-compatibility module, which would be almost trivial in this
instance.)
Why don't the R6RS people raise their concerns about the imperative
layer only at this point---once we'll get that done, we can decide
how to move on.
> Could you make a specific self contained proposal that would address
> the requirements?
Sure. It'll take me a little bit, as I'm still travelling and am
scheduled to help Will do the arithmetic SRFI next week, but I'll do
it.
--
Cheers =8-} Mike
Friede, V?lkerverst?ndigung und ?berhaupt blabla
More information about the R6RS
mailing list