< bitcoin-git>
[bitcoin] fanquake opened pull request #22578: [0.21] Backport #22122 to fix macOS CI issues (0.21...0_21_fix_macOS_ci) https://github.com/bitcoin/bitcoin/pull/22578
< bitcoin-git>
[bitcoin] sipa opened pull request #22579: Add references for the generator/constant used in Bech32(m) (master...202107_bech32_doc) https://github.com/bitcoin/bitcoin/pull/22579
< jnewbery>
luke-jr: I think the fix is to remove the std::move from the initializer list for the type and fallback arguments. They're both enum classes and will have underlying fundamental types that don't have move semantics. The m_type{std::move(type)} just copies type into m_type
< bitcoin-git>
bitcoin/master f8f772d Antoine Poinsot: macdeploy: alternative info to download the macOS SDK
< bitcoin-git>
bitcoin/master 19434fa fanquake: Merge bitcoin/bitcoin#22516: macdeploy: alternative info to download the m...
< bitcoin-git>
[bitcoin] fanquake merged pull request #22516: macdeploy: alternative info to download the macOS SDK (master...nit_guix_doc) https://github.com/bitcoin/bitcoin/pull/22516
< bitcoin-git>
[bitcoin] ritickgoenka opened pull request #22582: test: a test to check descendant limits (master...test_descendant_limit_bushy) https://github.com/bitcoin/bitcoin/pull/22582
< laanwj>
jonatack: i'm not sure updating anything prior to -final is a good idea; this means there is a code change (well, data structure change) that is not tested in a rc
< laanwj>
jonatack: for the same reason we don't do a translations update after the last rc and before final
< laanwj>
jonatack: the only exception are the release notes and potentially other documents not included in the code; so if we want to do another I2P hardcoded seeds update, we sould just do it in a rc
< luke-jr>
possibly bumping the UniValue version requirement, assuming jgarzik gets his end done in time? :x
< gleb>
Yeah so after a slight protocol update to Erlay we discussed here at some point, it becomes apparent how sacrificing some latency allows to save bandwidth. The trade-off is somewhat non-trivial...
< gleb>
My estimates show a transaction takes 4-5 seconds to relay across 95% of the nodes
< sipa>
right now?
< gleb>
Yes
< gleb>
What if we increased it to 10 seconds? How we should even think about these numbers
< sipa>
is that a measurement of the network, or something you conclude from simulating something like the current network?
< gleb>
For a good reason, of course.
< gleb>
simulating, but also I saw someone confirming this independently from me.
< sipa>
i think that anything that is "small" compared to the interblock time is fine, but perhaps this is more something to discuss on the ML?
< gleb>
sipa: hm, good point, maybe ML or slow-communication github issue is a better venue.
< gleb>
Also, I hear your point about interblock time.
< gleb>
For now just getting quick feedback.
< gleb>
Okay, I guess no other opinions for now :) Let's move on then.
< sipa>
can also be discussed on the PR i guess
< gleb>
It becomes unreadable :(
< sipa>
yeah...
< gleb>
I already feel somewhat bad for having 100 resolved comments there
< jonatack>
maybe a dedicated issue
< laanwj>
right just make a new issue
< gleb>
yeah gonna do that.
< laanwj>
github's lack of a threading structure makes that necessary
< ariard>
so restrictions in france aren't improving, we're discussing with others orgas to move in another country
< ariard>
top candidate for now is switzerland
< ariard>
dates will stay the same
< ariard>
they do have a nice tool to check the conditions of entries here, like which vaccins are accepted, which countries :https://travelcheck.admin.ch/home
< jonatack>
restrictions certainly are a moving target
< jarolrod>
iceland?
< jarolrod>
although i believe they have restrictions against eastern block countries
< ariard>
jarolrod: note, there is a visa issue, we do have attendees coming from asia/non-first class countries
< ariard>
switzerland is okay for uk/usa/eu, i'm still checking for the other cases
< ariard>
and the good thing it will be easier to accesss for our american friends :)
< jarolrod>
ah yes. Very complicated. Thank you to those coordinating and doing the work :)
< ariard>
so we'll take a decision before august 23rd (or 3 meeting from now)
< jonatack>
(i recently recalled that i know the owner of an amazing chateau here and it would have been pretty amazing to do it there...maybe next time)
< ariard>
meanwhile, gonna send a mail for topics collection/signal group in the coming weeks
< ariard>
stats levels, we do have 33 yes so far, and sponsoring for travel subsidies should be okay
< ariard>
voila, if you have nay question feel free to reach jonatack or schmidty or me, staying available
< ariard>
*any
< jarolrod>
thanks for the update!
< laanwj>
yes, thanks for the update, and for sticking with it, it's definitely not easy this time