Boost logo

Boost :

From: John Maddock (jz.maddock_at_[hidden])
Date: 2024-08-21 11:32:01


> The situation is becoming more confusing every day. 20 days ago the
> Foundation posted a message (signed by the Board) stating that "we
> feel the most appropriate thing to do at this juncture would be to let
> the developers make a decision on how they would like to proceed
> regarding what level of ownership the C++ Alliance should have on
> Boost Library assets"
>
> Note the phrase "let the developers make a decision".
>
> The Foundation clearly expressed that there are two available options
> (Alliance or Foundation) and a third option was not needed ("given
> that there are already many Boost Developers on the Boost Foundation
> Board of Directors, we don’t see this as a meaningful deviation from
> the status quo").
>
> 20 days later members of the Foundation say there will be more options
> because they know at least one additional option is in development.
> But according to the previous logic, this additional option should be
> a "meaningful deviation from the status quo" or it should not be
> considered. But we do not know anything about it and the Foundation or
> its members have not disclosed any information to the developers.
>
> Some days ago we knew that "the Boost Foundation is currently voting
> on a motion to utilize the library review process to decide the
> governance question which includes process suggestions. We should wait
> until that outcome".
>
> But we have no information about this outcome and the board signed the
> original two option proposal. It's at least surprising that the
> Foundation made such a strong decision proposal and 20 days later
> members of the board think that "allowing them to put forward
> proposals is part of facilitating that agency trying". Is that a new
> position from the board or a personal opinion from board members?
>
> Minutes from the Foundation are not updated since June so we don't
> know if there is new information that justifies any new position from
> the board or its members.
>
> Additionally, Foundation board members considered that the decision
> should have "participation from the greater C++ community" and "the
> review manager should take into consideration feedback from all
> sources." which contradict the earlier "let the developers make a
> decision" from the board.
>
> It was announced that the Boost Foundation was having a meeting (past
> Thursday?) about the review process for the governance, but we have no
> information about the decisions taken on that meeting.
>
> This simply is not sustainable. It's not productive.
>
> We already have a trusted review manager (Glen), which is a Foundation
> board member, and a very active Boost developer. We have some initial
> information about the operational perspective both from David
> (https://lists.boost.org/Archives/boost/2024/08/257346.php) and Vinnie
> (https://lists.boost.org/Archives/boost/2024/08/257347.php), which
> must be completed before the review.
>
> The review was proposed initially for the next week (pull
> https://github.com/boostorg/website/pull/871/commits/55010d4fdcf87db98332da39b90ab5ada2511c3b).
> So it seems that the C++ Alliance should already have a mature
> proposal (otherwise, C++ Alliance's proposal will be simply rejected).
>
> We also had some clarifications ("Misunderstandings about the Boost
> Foundation",
> https://lists.boost.org/Archives/boost/2024/08/257463.php) that the
> Foundation thinks it should be considered from developers during this
> process.
>
> I think it's evident that we have no new information that justifies
> additional delays. If anyone has an interesting proposal in a mature
> state, great, boost developers should know about it just now because
> it's time to decide and move to the next thing.
>
> According to our established review process, the review manager should
> finalize the schedule with the Review Wizard and the submitter. Please
> let Glen, Mateusz and Vinnie agree on that.
>
> I expect the usual 10 days period will be enough so that active boost
> developers (which are the ones the Foundation requested a decision
> from) can emit their opinions. In any case, according to the formal
> review process the review manager can ask the review wizard for
> permission to extend the review schedule if there are too few reviews
> or some new important information must be considered, so the review
> can be extended using the already established process if necessary.
>
> IMO the Foundation or the Alliance should not try to condition this
> developer decision that the same Foundation proposed 20 days ago and
> the Alliance agreed on. That ship has sailed.
>
> The review manager "shall check the submission to make sure it really
> is complete enough to warrant formal review"
> (https://www.boost.org/community/reviews.html).
>
> If so, we should schedule the review on the following weeks and take a
> decision with the information we have, as no decision is the worst
> decision. If 6 months later we have newer information that requires a
> new review and a new change, so be it.
>
Thank you Ion, this is very well written, and sums up the situation
nicely.  I wholeheartedly agree.  We need to move along with the this.

John.


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk