I'm still worried

[Vincent Manis]

Here are the things that still furrow my brow.

1. I'm glad that Marc has agreed to be the editor of Medium. However,
language design is still best done by committee (notwithstanding the
remark once quoted by Arthur C Clarke, “A Board is long and thin; it is
made of wood.”) to represent the diverse interests of users and
implementers. 3 or 4 people would be a minimal size. I strongly
recommend that there be at least one implementer on this committee.

2. The result of this committee should be a Report (whether or not in
the ALGOL 60 format, I leave to the committee). I think it presumptuous
to entitle this report “R8RS”: “R7RS Medium” is perfectly satisfactory.

3. This effort must not slow down the work of Large in any way,
including deadlines for voting and the like. The Medium docket should be
a subset (proper or not) of the Large features that have already been
approved, or are likely to be approved (tested via straw poll).

4. What about the hypothetical case where Medium adopts a SRFI that
Large has rejected, or rejects in the future? I would consider that a
disaster.

Viewed this way, the Medium effort would consist of (a) bringing chosen
R6RS facilities into the R7RS world (either currently finalized as SRFIs
or proposed as new SRFIs where necessary, using SRFI 181 as an example);
(b) tightening up as many of the unspecified or “it is an error” cases
in R7RS as possible; (c) tightening up those SRFIs (including PFNs) that
are to be included, and bringing them to a consistent style and
notation; and (d) producing interim and final reports.

This approach actually aids the Large effort, as most if not all of
these activities are actually needed for Large.

I suggest that once the committee is selected, it should produce a
charter that expresses its action plan, along with a brief description
of its process, both for development of Medium itself and for
interaction with Large. Much of this material has already been proposed
in various emails on this list, so this should not be too difficult.

-- vincent

[Marc Nieper-Wißkirchen]

Am Mi., 9. März 2022 um 21:52 Uhr schrieb Vincent Manis <vman...@gmail.com>:
>
> Here are the things that still furrow my brow.
>
> 1. I'm glad that Marc has agreed to be the editor of Medium. However,
> language design is still best done by committee (notwithstanding the
> remark once quoted by Arthur C Clarke, “A Board is long and thin; it is
> made of wood.”) to represent the diverse interests of users and
> implementers. 3 or 4 people would be a minimal size. I strongly
> recommend that there be at least one implementer on this committee.

Just to clarify this point: I would and should only volunteer as an
editor for something like Medium if I can subscribe to whatever will
be in the charter for Medium. While I have some ideas of what Medium
should be, WG 2 may see it very differently. But this is probably
self-evident.

In any case, it has always been meant to be a work of a group of
people; small enough to be flexible and fast but large enough to have
enough shoulders on which to build the project.

[Arthur A. Gleckler]

On Wed, Mar 9, 2022 at 2:16 PM Marc Nieper-Wißkirchen <marc....@gmail.com> wrote:

 

Just to clarify this point: I would and should only volunteer as an
editor for something like Medium if I can subscribe to whatever will
be in the charter for Medium. While I have some ideas of what Medium
should be, WG 2 may see it very differently. But this is probably
self-evident. 

Just to be clear, WG2 stands for Working Group 2, which is the group of people who are working on R7RS Large, per the Steering Committee's charter.  While there may be a lot of overlap, we should use a different term for whatever group works on something else. 

[Lassi Kortela]

> Just to clarify this point: I would and should only volunteer as an
> editor for something like Medium if I can subscribe to whatever will
> be in the charter for Medium. While I have some ideas of what Medium
> should be, WG 2 may see it very differently. But this is probably
> self-evident.
>
> In any case, it has always been meant to be a work of a group of
> people; small enough to be flexible and fast but large enough to have
> enough shoulders on which to build the project.

I keep agreeing with everything Marc says.

[Marc Nieper-Wißkirchen]

Am Mi., 9. März 2022 um 23:32 Uhr schrieb Arthur A. Gleckler
<a...@speechcode.com>:

I actually meant WG2 when I wrote WG2. My point was that when the
raison d'être of the Medium project is to facilitate the development
of R7RS Large, the goals for the Medium project should be endorsed by
those working on R7RS Large, thus WG2.

[Arthur A. Gleckler]

On Wed, Mar 9, 2022 at 10:47 PM Marc Nieper-Wißkirchen <marc....@gmail.com> wrote:

 

I actually meant WG2 when I wrote WG2. My point was that when the
raison d'être of the Medium project is to facilitate the development
of R7RS Large, the goals for the Medium project should be endorsed by
those working on R7RS Large, thus WG2. 

If it's a subcommittee of WG2, then that makes sense.