It is starting to look like the Capy and Corosio review outcome is not going to materialize. What are our options at this point?
Would it be acceptable to just bring in an new review manager to assess the already submitted reviews, rather than start from scratch? Rob On Fri, Sep 4, 2026 at 8:18 AM Glen Fernandes via Boost < boost@lists.boost.org> wrote:
Vinnie Falco wrote:
It is starting to look like the Capy and Corosio review outcome is not going to materialize.
What are our options at this point?
A brand new formal review is always an option.
Glen _______________________________________________ Boost mailing list -- boost@lists.boost.org To unsubscribe send an email to boost-leave@lists.boost.org https://lists.boost.org/mailman3/lists/boost.lists.boost.org/ Archived at: https://lists.boost.org/archives/list/boost@lists.boost.org/message/XVHNAUFH...
On 9/4/26 8:42 AM, Robert Beeston via Boost wrote:
Would it be acceptable to just bring in an new review manager to assess the already submitted reviews, rather than start from scratch?
Rob
On Fri, Sep 4, 2026 at 8:18 AM Glen Fernandes via Boost < boost@lists.boost.org> wrote:
Vinnie Falco wrote:
It is starting to look like the Capy and Corosio review outcome is not going to materialize.
What are our options at this point?
A brand new formal review is always an option.
Glen _______________________________________________
Why not to both? New formal review which would accept new reviews and add them to the set of current ones. One of the problems with the formal review process has been the two week window for accepting reviews. Not everyone who might review can fit into a specific time window. The boost library incubator was an attempt to remedy this but was not successful. Robert Ramey
sob., 5 wrz 2026 o 18:39 Robert Ramey via Boost <boost@lists.boost.org> napisał(a):
On 9/4/26 8:42 AM, Robert Beeston via Boost wrote:
Would it be acceptable to just bring in an new review manager to assess the already submitted reviews, rather than start from scratch?
Rob
On Fri, Sep 4, 2026 at 8:18 AM Glen Fernandes via Boost < boost@lists.boost.org> wrote:
Vinnie Falco wrote:
It is starting to look like the Capy and Corosio review outcome is not going to materialize.
What are our options at this point?
A brand new formal review is always an option.
Glen _______________________________________________
Why not to both? New formal review which would accept new reviews and add them to the set of current ones.
One of the problems with the formal review process has been the two week window for accepting reviews. Not everyone who might review can fit into a specific time window. The boost library incubator was an attempt to remedy this but was not successful.
It is true that two weeks is too short a time. On the other hand, if one has an infinite amount of time and gets relaxed, one will do a review even less likely. Regardless of the result, Capy/Corosio authors received a bunch of feedback. The first step would be to incorporate all of it (but not necessarily apply every suggestion). I think that one of the reasons for the diminishing numbers of reviews, even if not the most important one, is that all the easy/small libraries got accepted. Now we are left with big ones, requiring a lot of specialist knowledge. One way to mitigate this, is to have super clear and super polished documentation, so that some questions need not be asked and some confusion need not arise. Regards, &rzej;
On Fri, Sep 4, 2026 at 8:17 AM Glen Fernandes <glen.fernandes@gmail.com> wrote:
A brand new formal review is always an option.
I am concerned about review volume. The average number of reviews per library has been on a yearly decline. And I would imagine there would be less enthusiasm for a re-review of the same library set. We have invested significant resources in promoting Capy and Corosio at CppCon, and I was expecting a review verdict (either acceptance, with conditions, or rejection) ahead of the conference. My original thinking for asking Jeff Garland was that 1. It would a fitting reprisal of his role acting as review manager for Boost.Asio 2. Jeff's deep institutional connections in WG21 and Beman project would bring in more new reviewers Unfortunately, my decision backfired spectacularly. I am not aware of any reviewers that were brought in specifically by the review manager. I asked Dietmar Kuhl on the LEWG reflector if he would participate, since he has taken ownership of standardizing networking on std::execution. He declined. This speaks to the larger, ongoing problem of declining Boost participation. The Boost Formal Review process is one of the two valuable social technologies that Boost delivers to the C++ ecosystem, and this technology depends on reviewers and review managers. If the formal review process becomes increasingly viewed as a process that does not deliver its verdict reliably, I fear this will cause permanent damage to the Boost Libraries. Thanks
Andrey Semashev wrote:
On 4 Sep 2026 18:06, Vinnie Falco via Boost wrote:
It is starting to look like the Capy and Corosio review outcome is not going to materialize.
What are our options at this point?
Was anyone able to reach Jeff directly?
I have, he said "I'm working on it."
Many apologies for the delay - the results are imminent. On Sat, Sep 5, 2026 at 11:24 AM Peter Dimov via Boost <boost@lists.boost.org> wrote:
Andrey Semashev wrote:
On 4 Sep 2026 18:06, Vinnie Falco via Boost wrote:
It is starting to look like the Capy and Corosio review outcome is not going to materialize.
What are our options at this point?
Was anyone able to reach Jeff directly?
I have, he said "I'm working on it."
_______________________________________________ Boost mailing list -- boost@lists.boost.org To unsubscribe send an email to boost-leave@lists.boost.org https://lists.boost.org/mailman3/lists/boost.lists.boost.org/ Archived at: https://lists.boost.org/archives/list/boost@lists.boost.org/message/HNLOLJWT...
On Sat, Sep 5, 2026 at 11:24 AM Peter Dimov via Boost <boost@lists.boost.org> wrote:
...he said "I'm working on it."
There are essentially three options: 1. Waiting longer. This has the highest variance. If the review is delivered, then closure. If not, the withdrawal will be silent (like the bdt2 withdrawal which was followed by 10 years of mailing list silence). 2. A replacement review manager to assess existing reviews. This preserves Jeff's investment in the review process and also gives us an institutional outcome. 3. A brand new review (Glen's option). This is the worst possible choice. The review manager's work to date is implicitly voided. The appetite to review the same libraries twice would likely be less than before. If a review outcome is to materialize it would have to happen in the next 6 days because CppCon is unlikely to yield reviews, and the wg21 workload post-conference is high (there's a mailing plus telecons). A sensible path would be to start soliciting a new review manager now, since the earliest we could get a new review going would probably be early October if that.
participants (8)
-
Andrey Semashev -
Andrzej Krzemienski -
Glen Fernandes -
Jeff Garland -
Peter Dimov -
Robert Beeston -
Robert Ramey -
Vinnie Falco