Call for proposals: Stygian Blue and Reddish Green ballots

[Daphne Preston-Kendal]

Dear WG2 members,

The macrological fascicle is now in principle ready to be released for public comment; I am waiting on some changes to the infrastructure behind the scheme-reports.org <http://scheme-reports.org/> domain, which involve working with the future mailing list host and others before they can be finalized.

However, in the meanwhile, please submit proposals for ballot questions to be voted on by the Scheme community as we move forward into the next stage of the Foundations, and continue work on the Batteries.

An initial set of proposals I have made, and instructions for the format of proposals, can be found under:
<https://codeberg.org/scheme/r7rs/wiki/Stygian-Blue-and-Reddish-Green-ballot-proposals>

You can either send proposals to this list for me to add, or edit the wiki directly to add them — if you edit the wiki, please let me know about the question you have added.

Best wishes


Daphne Preston-Kendal

[Dr. Arne Babenhauserheide]

Dear Daphne,

thank you for your work on moving R7RS large forward!

> An initial set of proposals I have made, and instructions for the format of proposals, can be found under:
> <https://codeberg.org/scheme/r7rs/wiki/Stygian-Blue-and-Reddish-Green-ballot-proposals>

The information provided for the list-queues question doesn’t yet enable
me to take an informed decision, because the options seem different
enough that I cannot judge how they differ in performance and memory
properties, but these would be an important deciding factor for me.

Are there already tests which show their absolute cost and algorithmic
complexity (or proofs that give the minimum bounds for those)?

Best wishes,
Arne
--
Unpolitisch sein
heißt politisch sein,
ohne es zu merken.
draketo.de

[Daphne Preston-Kendal]

On 12 Feb 2024, at 12:51, 'Dr. Arne Babenhauserheide' via scheme-reports-wg2 <scheme-re...@googlegroups.com> wrote:

> Dear Daphne,
>
> thank you for your work on moving R7RS large forward!
>
>> An initial set of proposals I have made, and instructions for the format of proposals, can be found under:
>> <https://codeberg.org/scheme/r7rs/wiki/Stygian-Blue-and-Reddish-Green-ballot-proposals>
>
> The information provided for the list-queues question doesn’t yet enable
> me to take an informed decision, because the options seem different
> enough that I cannot judge how they differ in performance and memory
> properties, but these would be an important deciding factor for me.
>
> Are there already tests which show their absolute cost and algorithmic
> complexity (or proofs that give the minimum bounds for those)?

The question is proposing that we should ultimately provide only one of these on principle and remove all the others. It isn’t intended to ask *yet* which one specifically. We don’t have specs or complete implementations for finger trees and pvectors yet, for example.

I tried to find a way to reword the question or the rationale to make this clearer, but couldn’t find a good way yet. Perhaps this sort of question is (procedurally) the wrong way to go about this entirely, though …

John did a brief summary of the advantages of each kind of sequence at <https://codeberg.org/scheme/r7rs/issues/146#issuecomment-1260251>.


Daphne