JackH 2017-01-10 00:32:57
for that comment alone I wish I was a miner so I could block segwit out of spite for what you are saying
JackH 2017-01-10 00:33:08
you are advocating we do NOTHING? what is wrong with you people
JackH 2017-01-10 00:34:38
and is 50% of nodes not enough?
Xanather 2017-01-10 00:36:49
I think 50% is too low
Xanather 2017-01-10 00:36:57
maybe 60% at minimum
sturles 2017-01-10 00:38:25
JackH: No, we should push segwit, but it can't be activated too early. If you know anyone running an outdated node, please convince them to upgrade.
Xanather 2017-01-10 00:39:20
software compatibility isn't the problem, miners simply don't want to signal support for segwit
JackH 2017-01-10 00:39:36
and that would do what? would 95% of miners magically start signalling?
arubi 2017-01-10 00:40:15
by that logic, the current largest body of hashing power seem to not want to signal for anything
arubi 2017-01-10 00:40:20
and keep things as they are
arubi 2017-01-10 00:41:32
and since it's only been two months, and not even one major release with segwit enabled, I don't think anyone can say that miners don't want segwit
JackH 2017-01-10 00:42:07
I wonder when you guys change your tone? how long and how far do we have to go? 6 months?
arubi 2017-01-10 00:42:17
change the tone to what?
JackH 2017-01-10 00:42:31
9 months maybe? how many excuses will you keep coming up with for why not enough miners are signalling
buZz 2017-01-10 00:42:32
C minor
arubi 2017-01-10 00:42:45
they might never signal, who cares
arubi 2017-01-10 00:42:52
bitcoin stays the same, and that's fine
JackH 2017-01-10 00:42:58
for who
Xanather 2017-01-10 00:43:01
JackH, they'll change their tone if BU hash rate surpasses 30%
nkuttler 2017-01-10 00:43:14
JackH: i think you've made your point. no sense in repeating yourself over and over
arubi 2017-01-10 00:43:38
I thought hash rate doesn't matter for BU at all
arubi 2017-01-10 00:44:01
why don't miners just fork to a new chain where block size is different?
arubi 2017-01-10 00:44:37
it's not like they have to wait for some threshold like segwit has, just press a button
Xanather 2017-01-10 00:44:57
arubi, same logic could be applied to segwit
arubi 2017-01-10 00:45:18
nope, because full nodes check for activation threshold too
buZz 2017-01-10 00:45:28
yeah why not fork away so we dont have to listen to them
buZz 2017-01-10 00:45:40
maybe because that would destroy all value they hold
buZz 2017-01-10 00:45:45
obviously
arubi 2017-01-10 00:48:00
Xanather, in case some miner decides to activate segwit on their own, they'll effectively be mining using stricter rules than any other miner. nodes won't back them up because the threshold isn't crossed yet. it's not the same logic as BU, in which any miner can simply create a bigger block and have all clients sync to its chain eventually
Xanather 2017-01-10 00:48:58
arubi, BU follows the longest chain, so that wouldn't work for BU either
Xanather 2017-01-10 00:49:13
but you indirectly answered you own question so good job :)
arubi 2017-01-10 00:49:23
even if the largest miner suddenly starts mining segwit, that won't cause nodes to follow. the threshold has to cross first. in BU, the largest miner decides the chain
arubi 2017-01-10 00:49:35
and also decides the block size for everyone
arubi 2017-01-10 00:49:59
unfortunately for BU, the longest chain isn't always the valid chain
Xanather 2017-01-10 00:50:28
arubi, your argument falls apart when you try to define what a valid chain is
Lauda 2017-01-10 00:50:43
Doesn't that technically open an attack vector? Longest chain is not a fool-proof method.
arubi 2017-01-10 00:50:50
the rules of validity are found in the client that you run
Lauda 2017-01-10 00:50:51
Chain with most PoW is that should be looked at.
Lauda 2017-01-10 00:51:02
what*
arubi 2017-01-10 00:51:14
Xanather, that's why even if 100% signal for BU, that still wouldn't cause my client to follow their chain if it's invalid
arubi 2017-01-10 00:51:28
so like I said, hashrate doesn't matter for BU at all
arubi 2017-01-10 00:51:49
and no, I won't change my tone if BU gets or surpasses 30%
marijn_ 2017-01-10 00:52:14
Is any up to date core node automatically switching to segwit if it's signalled?
Lauda 2017-01-10 00:52:21
Yes
Lauda 2017-01-10 00:52:37
Same goes for any soft fork.
arubi 2017-01-10 00:52:40
v0.13.1 and above
arubi 2017-01-10 00:53:00
and any older versions will just continue to work as they always did, since, ^, soft fork
Lauda 2017-01-10 00:53:20
'as they always did'? Not exactly true, but yeah :p
arubi 2017-01-10 00:53:29
why not true?
marijn_ 2017-01-10 00:53:34
so the protocol is augmented to get the corresponding signatures seperately?
Lauda 2017-01-10 00:53:39
They can't validate Segwit stuff
arubi 2017-01-10 00:53:47
they never could?
arubi 2017-01-10 00:53:52
so.. like they always did
Lauda 2017-01-10 00:54:04
They were fully validating clients. After the activation they no longer are.
arubi 2017-01-10 00:54:23
that's not true, they're validating the rules of bitcoin
Lauda 2017-01-10 00:54:28
...
arubi 2017-01-10 00:54:30
and segwit doesn't break any rules
arubi 2017-01-10 00:54:47
fully validating doesn't mean validating signatures
arubi 2017-01-10 00:54:58
it means validating the chain and all the rules that make up consensus
Xanather 2017-01-10 00:55:01
arubi, yeah you wouldn't accept the chain, no transactions would be confirmed either if BU was 100%
Lauda 2017-01-10 00:55:30
I disagree and now I know why some people do too.
arubi 2017-01-10 00:55:44
Xanather, sure, if all miners go on to mine an invalid chain then there will be no blocks on the valid bitcoin chain
Xanather 2017-01-10 00:58:41
arubi, so you would accept a chain with no hash power behind it relative to its hard-forked counterpart as still valid?
arubi 2017-01-10 00:58:42
Lauda, can you explain what you disagree with? validating signatures is a small part of validating scripts\transactions\blocks\chains. validating signatures only matters in a script when there is an op_checksig
arubi 2017-01-10 00:59:29
Xanather, my client will follow the rules it's programmed to follow. if all miners decide they want to mine an invalid chain, that's their problem. the client in fact protects me from that
Lauda 2017-01-10 00:59:31
I disagree with play of semantics. Even though I support Segwit and have updated my clients, I refuse to say 'older clients continue to work as they always did'.
Lauda 2017-01-10 00:59:46
I find that to be a half-truth at most. That's just me.
arubi 2017-01-10 00:59:56
so Lauda , was that the case for previous soft forks too?
Lauda 2017-01-10 01:00:24
Maybe.
arubi 2017-01-10 01:00:43
I guess it has to be right? p2sh, cltv, csv..
Lauda 2017-01-10 01:01:15
I don't care about the past in this regard. I have stated my opinion.
Xanather 2017-01-10 01:01:33
arubi, its much easier to define a "bitcoin" aka the valid chain as the one with the most PoW put into it, other chains become vulnerable, every other definition becomes opinionated
arubi 2017-01-10 01:02:04
Xanather, so if 95% of miners suddenly decide to print 1 bil. bitcoins, and mine that chain, will you accept it?
Lauda 2017-01-10 01:02:26
A chain without widespread consensus is not Bitcoin.
arubi 2017-01-10 01:02:26
maybe if they start to steal money by redeeming transactions with invalid signatures?
atroxes 2017-01-10 01:02:38
arubi: Noone would, and that's not going to happen due to game theory
arubi 2017-01-10 01:02:55
atroxes, that's not what we're arguing about
arubi 2017-01-10 01:03:04
purely "what if"
atroxes 2017-01-10 01:03:32
Sorry for interrupting :) Just thought using hypotheticals that will never happen seemed weird :)
atroxes 2017-01-10 01:03:50
but yeah, then that's "Bitcoin" because 95% says so.
arubi 2017-01-10 01:04:00
I'm saying, if everyone follows the most pow, then miners are free to print money
Xanather 2017-01-10 01:04:07
arubi, I think thats another problem (mining centralization), obviously I would be against that and in that senario the PoW algorithm would have to change
arubi 2017-01-10 01:04:21
Xanather, so wait, why change it?
Xanather 2017-01-10 01:04:22
otherwise the chain without 1 billion pointed coins would be attacked repeatedly
arubi 2017-01-10 01:04:35
huh? so most pow or not?
Lauda 2017-01-10 01:04:41
Yeah ecosystem would emergency fork into another Pow algo.
arubi 2017-01-10 01:05:08
why fork?? Xanather, you say the proper chain to follow is the one with most PoW.
arubi 2017-01-10 01:05:33
now we have a chain with the most PoW where 1 bil. bitcoins are printed by some miner to itself. why fork?
arubi 2017-01-10 01:06:12
obviously there are more rules for validity than "most PoW". that's what conensus is
Xanather 2017-01-10 01:08:14
arubi, well your idea assumes that miners want to kill themselves
arubi 2017-01-10 01:08:36
not really, they like money
Xanather 2017-01-10 01:08:55
arubi, haha you think if they created 1 billion bitcoins they would make a few trillion dollars?
arubi 2017-01-10 01:09:21
if the software most nodes run is just programmed to follow the most pow chain, then they can be sure that nodes will follow them
arubi 2017-01-10 01:09:52
I mean, someone will have to actively search for those 1 bil. bitcoins right? not like the software is going to alert them
arubi 2017-01-10 01:12:45
maybe some miner just steals the coins from some old, dormant address. do these "longest pow" nodes even check if the script was redeemed correctly? probably not, right? they just follow PoW
Xanather 2017-01-10 01:15:08
if any malicious actor had control of that much hash power well even "valid" chains for a given PoW algorithm would cease to work properly
arubi 2017-01-10 01:15:19
that's not true at all
arubi 2017-01-10 01:15:58
no matter the hash power, a miner can't make fully validating nodes accept invalid blocks, or transactions
Xanather 2017-01-10 01:16:05
arubi, except it is true, the malicious actor could keep pumping out empty blocks
Xanather 2017-01-10 01:16:24
transactions stop verifying, network becomes halted
arubi 2017-01-10 01:17:19
that's nothing like the issues introduced by "longest pow" nodes like you suggested before
arubi 2017-01-10 01:17:33
and that's also true for the longest pow chain..
arubi 2017-01-10 01:18:31
miners decide which transactions to include in a block. that's always been the case. they have to include one (with coinbase), but probably also not a must for "longest pow" as the block could simply be invalid and nodes will still follow it
delboy1978uk 2017-01-10 01:28:45
hi guys o/
