> On 20 Dec 2023, at 23:27, Arthur A. Gleckler <a...@speechcode.com> wrote: > > Hi, Daphne. I hope that the holiday season is treating you well. > > Today marks the end of the extended last call for SRFI 245. What would you > like to do? It doesn't seem like the debate between 245 and 251 has been > resolved, or has even been fully fleshed out. But I will respect whatever > decision you make about about moving forward with this SRFI. Would you like > to finalize it, extend the last call until you have a chance to make changes > or extend the sample implementation, or do something else? > > Thanks.
Dear Arthur, Apologies for the delay here. I hope you don’t mind me sending this to the public SRFI 245 list. It’s a tricky decision. I haven’t heard anything more from SRFI 251 supporters on how the SRFI 245 spec could be made clearer. I also haven’t managed to come up with a sample implementation as requested by MNW, but this could be supplied later. (cpletrec is harder to implement than I initially thought.) I am stuck between uncertainty about whether we should finalize as-is; withdraw on the basis that there are trickier issues here than I initially thought; or even re-enter the draft period in the hope of finding a solution to the underlying tensions between the clearly widespread desire to relax this restriction, the safety of returning to a continuation which would ultimately cause re-evaluation of a definition, optimizability, and overall intuitive semantics. (I note that SRFI 251’s supporters actually object to the latter point in re SRFI 245.) I tentatively suggest finalizing the SRFI as it currently stands and returning to the question in a subsequent SRFI. In terms of the implications for R7RS Large, I will consider if the ballot question on this issue should be rephrased to ask people about general support for the *idea* of relaxing the restriction without necessarily endorsing the strict R6RS program-style (as requested by MNW), SRFI 245, or SRFI 251 approaches, so that we can try to find a solution that satisfies everyone without binding ourselves (no pun intended) to a proposal which still seems incomplete. If as SRFI Editor you have an objection to finalizing when issues such as these are still open, let me know and we can consider an alternative approach. Daphne
