From: Robert Ramey (ramey_at_[hidden])
Date: 2021-03-15 17:57:34
On 3/15/21 9:48 AM, Richard Hodges via Boost wrote:
> Boost Describe Review Summary
Very good summary. Thanks for investing the effort to be review
manager. This job is the backbone of Boost and I'm sure a lot more
effort that it would first appear.
I do have a minor nit to point out.
>
> The remaining two reviews were marked CONDITIONAL ACCEPT.
>
> This seems to be common practice, as committing firmly to accept or reject
> can seem like a bold position either way. However the Boost review process
> has no provision for attaching conditions to library acceptance. The
> verdict must be either that the library is sufficiently well written and
> documented, as well as providing sufficient utility, to be included in the
> Boost offering, or it is not.
> Conditional Acceptances
I see the review manager's job as having wide latitude. That is, I see
the review manager having the authority to do any of the following.
a) consider CONDITIONAL as "reject"
b) consider CONDITIONAL as "accept"
c) consider CONDITIONAL as "accept" if the author agrees to make
changes, "reject" otherwise.
The actual number of "votes" is not relevant. The review manager treats
these as recommendations - but the final decision is his and his alone.
So he can interpret each "CONDITIONAL" in any way he things is most
useful and/or appropriate. I see him having the authority to
"negotiate" with the author to implement changes in exchange for
acceptance.
To me, this is a fascinating thing about Boost. It's blatantly
undemocratic. There is no one man/one vote here! It is the essence
meritocracy. There is not reverence for democracy for it's own sake. But
still it attempts to cast a wide net in getting varied input for library
acceptance. It's out of sync with the times - and I think that is the
(not so) secret to it's success.
Exercise for the reader:Contrast the Boost procedures for achieving
approval with that of the C++ committee. What does this teach us - if
anything?
Robert Ramey