<bitcoin-git>
[bitcoin] laanwj opened pull request #32447: [RFC] dbwrapper: Set global leveldb mmap limit (master...2025-05-leveldb-mmap-file-limit) https://github.com/bitcoin/bitcoin/pull/32447
Cory37 has quit [Quit: Client closed]
<bitcoin-git>
[bitcoin] laanwj closed pull request #32447: [RFC] dbwrapper: Set global leveldb mmap limit (master...2025-05-leveldb-mmap-file-limit) https://github.com/bitcoin/bitcoin/pull/32447
Cory37 has joined #bitcoin-core-dev
<bitcoin-git>
[leveldb-subtree] laanwj opened pull request #52: Revert "Increase maximum read-only mmap()s used from 1000 to 4096 on 64-bit systems" (bitcoin-fork...2025-05-revert-mmap-increase) https://github.com/bitcoin-core/leveldb-subtree/pull/52
Chris_Stewart_5 has quit [Ping timeout: 252 seconds]
vysn has quit [Quit: WeeChat 4.6.1]
Chris_Stewart_5 has joined #bitcoin-core-dev
<laanwj>
well that went from "maybe a good idea" to "that's horrible i would wish this on my worst enemies" really fast... but at least we found another probably redundant custom leveldb patch
brunoerg has quit [Read error: Connection reset by peer]
brunoerg has joined #bitcoin-core-dev
Guyver2 has quit [Quit: Going offline, see ya! (www.adiirc.com)]
<achow101>
While there are still several additional cleanups that can be done, I think this project has reached its completion point and it can be removed from the updates list
<darosior>
achow101: gg
<achow101>
#topic orphan resolution WG Update (glozow)
<glozow>
no updates. I hope to get back to it soon.
<achow101>
#topic QML GUI WG Update (jarolrod, johnny9dev)
<glozow>
achow101: \o/
<johnny9dev>
bitcoin-core/gui-qml#448 is complete and just waiting for one last validation and merge
<johnny9dev>
I opened up the first draft of sending to multiple recipients at bitcoin-core/gui-qml#450
<johnny9dev>
Contributor Gee has un-drafted AssumeUTXO snapshot loading bitcoin-core/gui-qml#424 and is looking for review. Related to that, he is also looking to contribute this work to the Qt widgets gui at bitcoin-core/gui#870
<achow101>
#topic Script Validation WG Update (fjahr)
<fjahr>
Received some feedback and new benchmarks on the batch validation PR, I am addressing these shortly, nothing else from me
ProbablyTidwell has quit [Quit: Client closed]
<achow101>
#topic moving the repo to bitcoin-core (take 3) (achow101)
<achow101>
Last week, a couple people said they wanted more time to formulate opinions, so we had punted another week
<achow101>
Any new opinions on moving the repo?
<Murch[m]>
Still seems like a good idea to me. Maybe we should check with Bitcoin Twitter, though. They seem to think that we should run everything by them. :p
zeropoint has joined #bitcoin-core-dev
<glozow>
lol
<laanwj>
hehe
<glozow>
I mean, I do think public perception is relevant for this
<fjahr>
The danger of moving the repo seems extremely limited but I guess since we discussed them at length last time I am leaning to just moving the bips because that's the safest option. But not a strong opinion.
jonatack has joined #bitcoin-core-dev
<achow101>
I still think we should move both
<jonatack>
hi
jespada_ has joined #bitcoin-core-dev
<laanwj>
same
<furszy>
I'm concerned about moving the repo, giving up ownership of /bitcoin due to some public drama in the future, and then the new /bitcoin owners recreating the repo with different source code.
<furszy>
that's my only concern..
<achow101>
why would bitcoin/ be given up?
<laanwj>
having all code repositories under one org would be great and has been the plan pretty much since bitcoin-core org was created
Emc53 has joined #bitcoin-core-dev
<kanzure>
if it is purely a perception problem then by what metric would we determine whether there is any perceptual advantage or improvement by moving or not moving
<Murch[m]>
I thought it was pretty unanimous that control over the bitcoin org would not be given up
<sr_gi[m]>
I think given the latest drama, we should no be making a move like right now, it may just add more fuel to the fire
<laanwj>
no, we're not giving up the bitcoin org
<glozow>
first pwuille is given up, then bitcoin... chaos
<sr_gi[m]>
Independently of what we decide
<glozow>
yeah seems like very bad timing
<laanwj>
i mean maybe, on the other hand it's always bad timing, there is always some drama in bitcoin
<abubakarsadiq>
leanwj I always wonder why do we have the two orgs
<achow101>
kanzure: it's not purely about perception, there are some tangible organizational and moderation benefits that moving would give us
jespada has quit [Ping timeout: 276 seconds]
<sipa>
abubakarsadiq: bitcoin-core was always intended to move everything bitcoin-core-specific to, but the bitcoin/bitcoin repo just never got around to it
<laanwj>
abubakarsadiq: good question, it's just how it's historically grew, there is no deep reason behind it
<kanzure>
not urgent at the moment but is there a way i could subscribe to bitcoin-core/ gh bans to auto-ban on the bips org to not duplicate that work
<achow101>
sr_gi[m]: otoh, if there's already drama, what could a little more do to make it worse?
<achow101>
kanzure: afaik, you can't
<pinheadmz>
isnt the whole point to separate bans between bips and bitcoin core ?
<achow101>
yeah
<sipa>
yes, they're different organizations, they should make independent conclusions
Emc99 has quit [Ping timeout: 240 seconds]
<sipa>
that doesn't mean they can't come to the same conclusion :p
rkrux has quit [Quit: Client closed]
<_aj_>
if bips is going to auto-ban everyone bitcoin-core does, what's the point?
<jonatack>
moving bitcoin/bitcoin to bitcoin-core/bitcoin or bitcoin-core/bitcoin-core seems the most philosophically correct
<kanzure>
my esteemed bip editor colleagues may have different opinions, but i don't see why i should want to duplicate any gh ban work; any gh-banned user can simply email an editor or any other BIP contributor anyway.
<jonatack>
probably best not waiting for the stars to align to do it
<jonatack>
i don't have a strong opinion about moving the bips repo
<laanwj>
jonatack: right; "bitcoin core" is a particular implementation, "bitcoin" is the network and block chain
<jonatack>
it may indeed be a good idea
<jonatack>
laanwj: yes
<achow101>
kanzure: and currently it is duplicated work for bitcoin core to have the bans be the same on both orgs, if we actually did that
<jonatack>
"may indeed be a good idea" -> bips migration to bitcoin-bips
<achow101>
jonatack: shall we bikeshed the name of the repo? I prefer bitcoin-core/bitcoin-core
<kanzure>
there are already two orgs anyway, so the duplicative work is an issue as-is
<fanquake>
The tangible benefits are certainly minor. The team syncing thing is negligable. i.e a handful of actions a year, across multiple people (think I've added a couple people in the last few months). Essentially the same in regards to ban management across repos. Aside from obvious spam, which requires no coordination, I think we've banned a handful of people?
<sr_gi[m]>
achow101: some people love to see hidden agendas in all we do, so why give them more stuff to keep making noise for longer? Given there is no real rush for this, I think we should exercise some caution
ProbablyTidwell has joined #bitcoin-core-dev
<sipa>
i agree there is no rush
<kanzure>
important to remind non-github users that github is simply a platform that some of us choose to use to work together. before github and sourceforge people used to actually email each other to discuss code or send patches. you can email anyone you want in the whole world! any contributor, as maintainers do not have a monopoly on email address namespace.
<furszy>
yeah.. I just don't see any value in jumping straight into another public debate right now. There's risk involved in moving the main repo, but no risk in moving only the BIPs repo. Maybe we could start with that for now?
<jonatack>
achow101: bitcoin-core/bitcoin-core sgtm too
<pinheadmz>
agreed, punt
<_aj_>
"bitcoin org is the network, not an implementation" doesn't really match the "but we'll keep bitcoin/bitcoin pointing at bitcoin-core/whatever forever"
<laanwj>
in a way moving bips is more controversial than moving bitcoin core, as bips does belong with the global network/blockchain and is shared between implementations
<darosior>
I did not think it through more since last time, in the middle of a conference currently. I still believe it makes sense ideologically but in practice it's unnecessary risk and burden. In addition right now it is pretty bad timing and i don't see the rush.
<achow101>
_aj_: sure, but that's more a practical thing because there's 15 years of links that would be broken or could end up pointing to a scam
<jonatack>
FWIW, I don't see how anti-core sentiment would be further fueled by moving to bitcoin-core
<stickies-v>
exponential backoff? discussing this every week might be a bit inefficient
<hodlinator>
I think it might help perception to "concede" that bitcoin-core is not bitcoin, but it's hard to predict.
<fjahr>
I think if we explain that this is actually a move that means people get less widely banned it should not cause too much drama, but unsure if people listen...
<kanzure>
_aj_: unfortunately gh namespace 'bitcoin' owners have essentially a moral obligation to not abandon it to the wolves due to long-term security issues, such as release spoofing, tag spoofing, issue rewriting, comment rewriting... etc.
<laanwj>
in any case, i don't think bringing this back every week is going to help, if there is no broad agreement to do anything then let's just not do it, or maybe in another 10 years
<darosior>
stickies-v: +1 i don't really see what's the rush here
<abubakarsadiq>
+1 darosior
<jonatack>
the current anti sentiment seems to be as much or more fueled in any case by *how* than by *what*
ProbablyTidwell has quit [Client Quit]
<achow101>
alright, we can punt for another decade lol
<darosior>
:)
<achow101>
any other topics to discuss?
<fjahr>
why not move the bips then?
<stickies-v>
kanzure: we could purposefully break the redirect in x years to ensure any people still relying on bitcoin/bitcoin move, almost entirely eliminating the hostile takeover risk
<Murch[m]>
Maybe bring it up in a month?
<sipa>
fjahr: that's out of scope for this meeting, imo
<achow101>
Murch[m]: I was thinking coredev actually
<sipa>
bip editor can decide to do that, if they want
<fjahr>
it would fix the problem we have...
<achow101>
fjahr: that's for the bip editors to decide
<darosior>
Yeah Coredev seems like a good fit to discuss this
<laanwj>
tbf i dono't htink sentiment is ever going to be better, there's always new drama
<Murch[m]>
fjahr: Only one of three problems by my count
<stickies-v>
thank you to everyone who’s hosted review clubs in recent weeks! next week we have a kernel review club coming up (https://bitcoincore.reviews/32317, notes up tomorrow)
<Murch[m]>
It would fix that bans by Bitcoin Core affect BIPs, but it would not fix the org split that Bitcoin Core has currently, nor disclaim that Bitcoin Core is Bitcoin
<stickies-v>
we have free slots available in the next weeks, so if you’re down to represent your WG pls dm myself or glozow
<laanwj>
stickies-v: good to know!
<bitcoin-git>
[bitcoin] furszy opened pull request #32449: wallet: init, don't error out when loading legacy wallets (master...2025_init_skip_legacy_wallets) https://github.com/bitcoin/bitcoin/pull/32449
<achow101>
#endmeeting
<corebot>
achow101: Meeting ended at 2025-05-08T16:27+0000
<kanzure>
for a future topic perhaps consideration of a private login-only issue tracker (but open registration) to downregulate public social media linkbrigading effects
<_aj_>
Murch[m]: (we've coped pretty well with bitcoin.com, bitcoin.org, twitter.com/bitcoin, reddit.com/bitcoin not necessarily being "bitcoin", why should github.com/bitcoin be that different?)
<jonatack>
Murch[m]: right, if anything reckon the move would be seen as positive. maybe some trial balloons can be lofted to check.
<kanzure>
we have absolutely not coped well with bitcoin.com are you kidding me
<fjahr>
Murch[m]: we still have headaches to not loose bitcoin/bitcoin so the org split isn't really going away imo and the third is philosophical. Seemed to me that the first issue was what started this discussion and bothered people the most. That's why it seemed like the highest priority to me.
<kanzure>
tons of users have been completely fooled by bitcoin.com
<stickies-v>
ah yes thank you abubakarsadiq - avenues aplenty!
<Murch[m]>
fjahr: Fair point
<laanwj>
we could also move everything from bitcoin-core to bitcoin instead lol
<achow101>
lol
<fjahr>
Just putting it here then: Would be great if bip editors could discuss it :)
<jonatack>
heh stoke the flames higher
<jonatack>
fjahr: we began discussing right away, no clear rough consensus for the moment iirc
<fjahr>
ok, thanks for the update!
<jonatack>
(when achow101 opened the issue i brought it up)
<_aj_>
laanwj: change the display name to "Bitcoin Core", add links to other resources/clients in the org description
<_aj_>
laanwj: work out whether the bitcoin core logo is black (x.com/bitcoincoreorg, github.com/bitcoin) or orange (github.com/bitcoin-core, bitcoincore.org)...
<sipa>
haha
<laanwj>
looooolll
enochazariah has quit [Quit: Client closed]
<laanwj>
black is the general logo, orange is specifically the main chain
<dzxzg>
next testnet reset should get a new color :)
Guest27 has joined #bitcoin-core-dev
Guest27 has quit [Client Quit]
dzxzg has quit []
<Murch[m]>
Next testnet reset should have a substantial premine for any dev that signs up in advance to get one
<Murch[m]>
That should make it easier to get coins to anyone that needs them and provide ample coin to sell into any exchanges that might start trading it
<Murch[m]>
Like maybe 1'000'000 coins per party that wants any
Talkless has joined #bitcoin-core-dev
<_aj_>
10.5M coins divided by how many people sign up, and block reward halves every 105k blocks would make the maths easier maybe
<laanwj>
dzxzg: a fun idea but i'm afraid giving them unique colors will only encourage people to collect them all :)
<bitcoin-git>
[bitcoin] fanquake opened pull request #32450: randomenv: remove some `/proc/` accesses (master...remove_randomenv_proc) https://github.com/bitcoin/bitcoin/pull/32450
<laanwj>
Murch: yes you'd say if the premine is that large it will discourage people from trading them for real money, on the other hand, crypto people...
<Murch[m]>
Given that Testnet 4 was immediately monopolized and traded after launch, it seems like another thing that could be tried
<Murch[m]>
Giving out half the coins in advance and halving the reward schedule like _aj_ suggests sounds good
adys has joined #bitcoin-core-dev
kevkevin has quit [Remote host closed the connection]
saikasyap has joined #bitcoin-core-dev
<bitcoin-git>
[bitcoin] andrewtoth opened pull request #32451: contrib: add xor-blocks tool to obfuscate blocks directory (master...xor-blocks) https://github.com/bitcoin/bitcoin/pull/32451