Am Fr., 18. März 2022 um 01:24 Uhr schrieb Vincent Manis <
vman...@gmail.com>:
> A few days ago, I posted emails with titles such as “I'm still worried”.
> I'm delighted to see how much things have advanced since then.
I'm glad to see all this new enthusiasm!
> Can I ask that someone create a web page with a consolidated list of
> such things as the mailing lists for each committee, the charters for
> each, a pointer to the (yet-to-be-created?) issues manager, and the
> like? That would not only serve as a handy reference, but would also be
> a great place to point newcomers (if there are any) to. Since almost all
> of this has been discussed in recent emails, creating the web page would
> be mostly a matter of cut-and-paste.
That would be great! If this page could be served from a git repo (our
centralized one), it would be even better because its history would be
preserved as well.
> One thing that we haven't discussed explicitly is the intended final
> product of the committee. I do think that it's important to establish
> that now, and include it in the webpage. My proposal: “The final product
> of the four committees will be a report in two volumes, the first
> entitled ‘R7RS-Large, Volume 1: Foundation Language and Libraries’ and
> the second, ‘R7RS-Large, Volume 2: External Libraries’. The content,
> organization, and format of these reports are TBD. Volume 1 will be
> prepared by Subcommittee F, with some content from Subcommittee E;
> Volume 2 by Subcommittee B (, with some content from Subcommittee E?).
> Both volumes will be prepared with collaboration from Subcommittee C.” I
> think that this can be substantially improved, but it might serve as a
> stalking horse. Since I'm not completely clear on Subcommittee E, maybe
> the intention is that it would prepare Volume 3 instead of submitting
> content to the other two volumes?
One complaint about R6RS was that the language described in its first
volume was not self-contained but that the second volume (Standard
Libraries) added extra semantic constraints to the language presented
in the first volume.
I would like to prevent this situation for R7RS so I think it is
important for the Foundations (Volume 1 in your list) to be
self-contained. Thus, it is probably easier and better if the E stuff
is not spread over the different volumes but in its own volume. Volume
2 (the batteries) should be independent of the procedures targeting
the environment, so an extra Volume 3 makes sense to me. The
environment APIs are those that are highly system dependent and will
be those that will likely have to be revised first (because systems
change), so organizing them in their own volume sounds like a good
idea.