Re: Should we MAY a "curly-write" and "neoteric-write"? Or even "sweet-write"? David A. Wheeler (10 Apr 2013 03:00 UTC)
Draft updated SRFI-110 and reference implementation David A. Wheeler (15 Apr 2013 04:09 UTC)

Re: Should we MAY a "curly-write" and "neoteric-write"? Or even "sweet-write"? David A. Wheeler 10 Apr 2013 03:00 UTC

John Cowan:
> I'd say forget sweet-write and go with curly-write and neoteric-write,
> and go ahead and use MUST modals for them, without overspecifying
> what they output.  R7RS systems MUST provide curly-write{simple,shared}
> and neoteric-write-{simple,shared} as well.
>
> As long as there is a good reference implementation, there is no reason
> not to require these things.

Okay.  A write-simple is, well, simple; I'll start there and see what people think.

Is there a a simple example of an efficient R7RS "write" and "write-shared" implementation?  The "obvious" solution involves hash tables (which are not portable), & I fear there's some corner case or clever simplification I won't realize.

--- David A. Wheeler