> When this SRFI was first published, it was more than a preliminary
> idea.  However, the more this discussion continues, the clearer it
> becomes that the idea isn't stable enough for SRFI-ification.

That's fair enough. I don't think any of us saw this coming.

> However, since the authors seem to be divided into two camps, and each
> camp seems to have a coherent idea, I'm wondering whether it would be
> better to have two competing SRFIs rather than one unfinished SRFI.
> Then reviewers, implementers, and perhaps R7RS Large voters could
> evaluate two coherent proposals on their merits, and choose one.

That may make sense. However, the point of contention is a very small
part of the API surface -- not even a Scheme procedure, but what
properties to have.

The properties should mostly be the remit of the Scheme Registry anyway,
but we need to enshrine the most essential ones like 'message in the
SRFI as well.

> I'm not going to insist upon this.  My role as SRFI editor is primarily
> to be an impartial facilitator.  But I hope that you will consider this
> approach if you can't come to agreement soon.

We're definitely much closer than last time, and no longer talking past
each other. But we should abide by the SRFI process.