New draft (#3) of SRFI 172: Two Safer Subsets of R7RS Arthur A. Gleckler (21 Oct 2019 02:56 UTC)
Re: New draft (#3) of SRFI 172: Two Safer Subsets of R7RS Lassi Kortela (21 Oct 2019 08:18 UTC)
Re: New draft (#3) of SRFI 172: Two Safer Subsets of R7RS Marc Nieper-Wißkirchen (21 Oct 2019 17:24 UTC)

Re: New draft (#3) of SRFI 172: Two Safer Subsets of R7RS Marc Nieper-Wißkirchen 21 Oct 2019 17:24 UTC

Am Mo., 21. Okt. 2019 um 19:19 Uhr schrieb John Cowan <xxxxxx@ccil.org>:
>
> Symbols often consume permanent resources in the outer program, depending on implementation.  If Scheme had gensym, I'd allow it.

I would call it a bug if a Scheme implementation doesn't GC interned
symbols. Having said that, there are implementation that still have to
fix this bug. :)

>
> On Mon, Oct 21, 2019 at 4:18 AM Lassi Kortela <xxxxxx@lassi.io> wrote:
>>
>> > I've just published draft #3 of SRFI 172.
>>
>> LGTM! The "alternative versions" provision sounds good.
>>
>> Is string->symbol gone on purpose? symbol->string is still there.