andrewtoth has quit [Remote host closed the connection]
pigeons has joined #bitcoin-core-dev
abubakarsadiq has joined #bitcoin-core-dev
<Murch[m]>
The Testnet 4 PR, #29775, has nine stale ACKs 😅, would be cool if maybe one or two more regular contributors could take a look—it seems close to RTM to me
<gmaxwell>
Murch[m]: I'm not involved enough in the project to even consider suggesting not moving forward with that PR as is, but ISTM that it might be necessary to include a zillion coin premine in testnets going forward to deal with that stuff.
<sipa>
gmaxwell: i'm not sure how that would help?
<Murch[m]>
My understanding is that a lot of the blocks so far were mined by Wiz and Portlandhodl, and there are three faucets already. A few people trying to collect testnet coins already for (allegedly) testing doesn’t seem like a big issue
<Murch[m]>
sipa: I think he means as a deterrent for them getting value
<sipa>
i think the deterrent is "we switched from testnet3 to testnet4, and if testnet4 becomes unusable, we'll do it again"
<sipa>
some people trying to acquire them is probably inevitable
<sipa>
but if the goal is a more controlled distribution of coins, there is signet
<gmaxwell>
sipa: if you trade these for value you'll just be making bitcoin core project wealthy, not yourself.
<Murch[m]>
gmaxwell: I read the thread, it seems like there is really just one person that asks where to get testnet coins "because they wanted to test without having to sync Testnet3". How did you get from there to "people trying to amass tn4 coins"?
<gmaxwell>
Murch[m]: I didn't actually read the thread, as it was just a nussance begging post. In that case it seems I might be false positiving there, ... historically many of those threads have looked like "I need coins for testing" "okay here is a testnet cent" "NO NO I NEED HUNDREDS OF COINS" "but it's divisible" "but my TESTING requires more coins!" ... and so on. :P
<gmaxwell>
so apologies for the likely false alarm there.
<sipa>
it's something we can consider for testnet5 i think, but at this point it's a question of getting tn4 in 28.0, or delaying things further
<Murch[m]>
Yeah sure, that’s been very annoying lately especially with the Babylon testnets etc.
<Murch[m]>
That’s why we are working on Testnet 4
<sipa>
and i think getting tn4 in is likely much more harmful to the trading-testnet crowd than delaying
<Murch[m]>
Yeah 💯
<gmaxwell>
yeah, point withdrawn. Though I'm not entirely sure how much they care what core is shipping but who knows. I think what matters to the degens is what the exchanges have up, maybe they'll happily stay on 3 and its problem solved.
<Murch[m]>
I mean, if they continue to assign Testnet 3 value, they can have it 😁
<gmaxwell>
yeah not your problem.
<gmaxwell>
one could also just circulate to tn5 before its a problem. The pattern I think with tn3 was that some people first quietly amassed a lot of coins, and then did things like setup an exchange. So periodically rotating may help.
<Murch[m]>
Yeah, it would have probably been better if we had sunset Testnet 3 a bit sooner, but now that we have exercised that muscle, it should also be easier to do it again
<Murch[m]>
What would be a better cadence? Maybe a new testnet every three years or so?
<sipa>
could even have a preprogrammed end date (a consensus rule that makes blocks with a timestamp after X invalid)
<Murch[m]>
That might be annoying, because it forces someone to work on it (although that might also be the point). Gives me "Ethereum Difficulty Bomb" jeebies
<instagibbs>
integration into and uptake of a new testnet takes time, I wouldn't be too glib about testnet creation, though N years sunsetting may be something to consider
<gmaxwell>
instagibbs: I think testnet has largely been failed for a couple years now.. but fortunately regtest and the signet stuff gave people many other options, so it didn't matter much.
<instagibbs>
Oh I agree, just saying it should probably be more than 3, I guess :)
<gmaxwell>
or more effort should be put into fixing things so people can use alternatives.
<gmaxwell>
fundimentally TN vs alternatives I think is only really required for testing that involves mining, and the diff1 rule (even with fix) kinda degrades it for that purpose.
andrewtoth_ has quit [Remote host closed the connection]
andrewtoth has joined #bitcoin-core-dev
<gmaxwell>
otherwise people should be able to use regtest (for local tests) or some signet thing, and should have a much better testing expirence with either than any TN like thing could give.
<bitcoin-git>
bitcoin/master a3cb309 Cory Fields: refactor: use recommended type hiding on multi_index types
<bitcoin-git>
bitcoin/master 0f68a05 glozow: Merge bitcoin/bitcoin#30194: refactor: use recommended type hiding on mult...
<bitcoin-git>
[bitcoin] glozow merged pull request #30194: refactor: use recommended type hiding on multi_index types (master...multi_index_types) https://github.com/bitcoin/bitcoin/pull/30194
pigeons has quit [Read error: Connection reset by peer]
pigeons has joined #bitcoin-core-dev
pigeons has joined #bitcoin-core-dev
emcy__ has joined #bitcoin-core-dev
mcey_ has quit [Ping timeout: 265 seconds]
<bitcoin-git>
[bitcoin] justinvforvendetta closed pull request #30564: get miniupnpc from ssl, this reverts commit 21b8a14d37c19ce292d5529597e0d52338db48a9 (master...master) https://github.com/bitcoin/bitcoin/pull/30564
<bitcoin-git>
[bitcoin] justinvforvendetta opened pull request #30602: Revert "depends: Fetch miniupnpc sources from an alternative website" (master...master) https://github.com/bitcoin/bitcoin/pull/30602
<bitcoin-git>
[bitcoin] justinvforvendetta closed pull request #30602: Revert "depends: Fetch miniupnpc sources from an alternative website" (master...master) https://github.com/bitcoin/bitcoin/pull/30602
<bitcoin-git>
[bitcoin] justinvforvendetta opened pull request #30603: Revert "depends: Fetch miniupnpc sources from an alternative website" (master...master) https://github.com/bitcoin/bitcoin/pull/30603
<Murch[m]>
From what I understand, distributing coins in Signet has also been a bit of a pain point
nanotube has joined #bitcoin-core-dev
Talkless has quit [Remote host closed the connection]
<bitcoin-git>
[bitcoin] sipa opened pull request #30605: Cluster linearization: separate tests from tests-of-tests (master...202408_cluster_tests) https://github.com/bitcoin/bitcoin/pull/30605