Advertisement
Guest User

Untitled

a guest
Feb 2nd, 2017
100
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 109.16 KB | None | 0 0
  1. T 1486091235 22* Topic for 22#BtcNegotiate set by 26Riiume (24Thu Feb 2 20:22:23 2017)
  2. T 1486091255 18<ETH-Contract18> Hard numbers: look at the current mempool stats
  3. T 1486091264 18<runnerman118> Please how can prove the financial impact of something that does not exist
  4. T 1486091265 20<banjomanjo>30 I have arrived, the negotiations may commence/s
  5. T 1486091266 18<Riiume18> I ask the same of the Blockstream side, although they have done a bit in this direction (Testnet, although it's play money so lacking the real financial incentives that govern Bitcoin)
  6. T 1486091286 20<banjomanjo>30 on a serious note though, is there anywhere previous discussions here are posted?
  7. T 1486091305 20<banjomanjo>30 (nm)
  8. T 1486091309 18<Riiume18> runnerman1, by modeling it. I don't have to shoot myself to create a simulation of the effects and see that it doesn't work out well for me
  9. T 1486091319 18<Riiume18> yes, logs
  10. T 1486091322 18<Riiume18> https://notebin.cc/1kdbou2efi8
  11. T 1486091334 20<banjomanjo>30 thanks
  12. T 1486091360 18<runnerman118> I dont even know what to do, I'm not a "professional bitcoiner"
  13. T 1486091428 18<Riiume18> runnerman1, yea it's difficult. Basically the various interest groups need to come together and fund studies proving/disproving their beliefs regarding the financial impact of taking Bitcoin one way or the other
  14. T 1486091450 18<Riiume18> Start a funding campaign to pay some academics to carry out the study, e.g.
  15. T 1486091504 18<psztorc18> Hi Riiume, I actually published some way of measuring something like that: http://www.truthcoin.info/blog/win-win-blocksize/
  16. T 1486091513 18<Riiume18> ahh, sweet! checking it out!
  17. T 1486091515 18<psztorc18> However, I think there is something more important to talk about
  18. T 1486091526 18<psztorc18> which is: "What do you think the miners want?"
  19. T 1486091560 18<Riiume18> Complete control of Bitcoin, without actually destroying Bitcoin's value
  20. T 1486091567 18<psztorc18> because they said they wanted a 2 MB blocksize around ~Dec 2015, and that's what they sort've got with segwit.
  21. T 1486091584 18<psztorc18> Well, are they going to get "Complete control of Bitcoin"?
  22. T 1486091601 23* gwillen (~gwillen@unaffiliated/gwillen23) has joined
  23. T 1486091624 18<Riiume18> No, if they remain persistent, the prediction markets say they get a hard fork, and there hasn't been much analysis of what occurs past that point
  24. T 1486091626 18<runnerman118> Of course miners are in control of bitcoin, it should always be that way
  25. T 1486091636 24* pero (~pero@unaffiliated/pero24) has left ("Leaving")
  26. T 1486091640 18<Riiume18> runnerman1, Full nodes wield substantial influence too
  27. T 1486091642 18<ETH-Contract18> Where is Adam the individual?
  28. T 1486091664 20<banjomanjo>30 alpalp, before doing that, I'm proposing we study it; create a simulation of Bitcoin forking in the near future and attempt to accurately model the economic effects
  29. T 1486091667 20<banjomanjo>30 who said that ^
  30. T 1486091670 18<runnerman118> Nope not as much as miners
  31. T 1486091672 20<banjomanjo>30 log structure is not clear
  32. T 1486091693 20<banjomanjo>30 but that is literally impossible, if you could do that you could model markets accurately, and that would have profound effects for the economy as a whole
  33. T 1486091697 18<psztorc18> Miners will not get "Complete control of Bitcoin", because no one can give that to them (or to themselves).
  34. T 1486091715 19<runnerman1> @banjomanjo exactly
  35. T 1486091719 18<psztorc18> So it seems that, according to you, they want something that they can't have.
  36. T 1486091723 23* jwinterm (~quassel@unaffiliated/jwinterm23) has joined
  37. T 1486091789 23* awfulcrawler (~awfulcraw@122-57-26-40.jetstream.xtra.co.nz23) has joined
  38. T 1486091791 18<runnerman118> @psztorc what are you talking about. Miners pretty much dictate what bitcoin is
  39. T 1486091799 19<Riiume> @banjomanjo, study doesn't have to have perfect predictivity; Correlations and extrapolations are always within some confidence interval
  40. T 1486091813 18<runnerman118> If all miners decide to stop mining we have no bitcoin.
  41. T 1486091822 18<psztorc18> @runnerman1, if "Miners dictate what bitcoin is" and "miners want a 2MB hard fork", then why don't the miners have the 2mb fork they wanted?
  42. T 1486091830 20<banjomanjo>30 [21:41] Point is the whole reason for this blocksize debate is political. If LN gets activated those fees will be going directly to Blockstream.
  43. T 1486091832 20<banjomanjo>30 and who said that
  44. T 1486091836 20<banjomanjo>30 that is also incorrect
  45. T 1486091848 18<runnerman118> If miners decide tomorrow to run a client that give everyone 50 million bitcoin, that what we will have
  46. T 1486091857 20<banjomanjo>30 Riome, you are not going to accurately model a market composed of 4-10 million people anywhere close to accurately
  47. T 1486091862 20<banjomanjo>30 the data from that would be useless/garbage
  48. T 1486091872 18<Riiume18> psztorc, hehe I like where you're going. So what DO the miners want then? Or are they actually irrational?
  49. T 1486091876 20<banjomanjo>30 and runnerman1...no...
  50. T 1486091879 20<banjomanjo>30 thats not how that works
  51. T 1486091897 20<banjomanjo>30 if the exchanges and economy don't support that, all miners are oding is wasting electricity mining worthless coins no one else will buy
  52. T 1486091918 18<psztorc18> @runnerman1, do you consider Litecoin blocks to be Bitcoin blocks?
  53. T 1486091932 18<runnerman118> Oh course...the price will immedially crash lol
  54. T 1486091954 24* dongcarl has quit (Quit: leaving)
  55. T 1486091956 20<banjomanjo>30 thats just your speculation
  56. T 1486091966 20<banjomanjo>30 many would have said that about a BU block being minted at > 1 MB
  57. T 1486091969 18<ETH-Contract18> How can a miner will not be incentivized to settle on-chain the current mempool (and get the reward) with a simple upgrade to a today's technologically-possible 4MB max blocksize limit
  58. T 1486091970 20<banjomanjo>30 that happened, the price didn't crash
  59. T 1486091998 18<psztorc18> @runnerman1, If miners decide tomorrow to run a client that gives everyone 50 million litecoin, will Bitcoin be equal to a client that gives everyone 50 million litecoin?
  60. T 1486092062 18<runnerman118> What?? Litecoin and bitcoin are two different things
  61. T 1486092081 18<psztorc18> By what criterion?
  62. T 1486092109 20<banjomanjo>30 what is with all the repetition of LN fees going to Blockstream?
  63. T 1486092114 20<banjomanjo>30 that is completely untrue
  64. T 1486092139 20<banjomanjo>30 fees go to whoever is facilitating payments, i.e. everyone in the middle of a path through a set of channels
  65. T 1486092142 18<runnerman118> Oh were are the fees going to then?
  66. T 1486092148 20<banjomanjo>30 which requires locking coins up
  67. T 1486092161 20<banjomanjo>30 which if you actually sit and think for a minute, benefits the miners most out of anyone
  68. T 1486092184 20<banjomanjo>30 they as a group have both the largest stash of coins to provide liquidity with, and as miners can use their own blockspace to guarantee smooth channel closures and openings
  69. T 1486092195 20<banjomanjo>30 LN would allow miners to suck up most of those fees, not Blockstream
  70. T 1486092207 18<ETH-Contract18> i don't want my transaction to be processed by a set of middle mans through a set of channel, I just want my transaction to be settle on-chain in the next 10 minutes, simple
  71. T 1486092219 20<banjomanjo>30 then you don't have to use LN
  72. T 1486092219 18<runnerman118> ^
  73. T 1486092223 18<psztorc18> It is possible that LN will decrease on-chain fees. And it is possible that Blockstream will run a LN hub that will attract customers.
  74. T 1486092224 20<banjomanjo>30 have fun
  75. T 1486092235 18<psztorc18> runnerman1, Are you avoiding my questions?
  76. T 1486092238 20<banjomanjo>30 I on the other hand will use it if it makes sense at the time and is cost effective
  77. T 1486092248 18<runnerman118> which question
  78. T 1486092253 18<runnerman118> I might of missed it
  79. T 1486092267 18<psztorc18> I see.
  80. T 1486092325 18<runnerman118> The criterion one? They are different because the clients/node the miners are running are different
  81. T 1486092325 18<psztorc18> @Riiume, I think I know what the miners want. But you can't broker a compromise unless you give them what they want, and you probably can't do that unless you know what they want.
  82. T 1486092328 23* iopools (b53ea498@gateway/web/freenode/ip.181.62.164.15223) has joined
  83. T 1486092335 23* amiller (~socrates1@unaffiliated/socrates102423) has joined
  84. T 1486092351 20<banjomanjo>30 Riiume, what do you mean "sidechains will be able to affect the main blockchain willy nilly" ?
  85. T 1486092353 18<Riiume18> psztorc, please give us your view on what they want!
  86. T 1486092372 18<ETH-Contract18> explain that please
  87. T 1486092374 18<jwinterm18> I bet it's hookers and blow
  88. T 1486092386 18<runnerman118> Who knows, everyone wants something different
  89. T 1486092411 18<gwillen18> I am curious what this channel is expectd to accomplish
  90. T 1486092418 18<gwillen18> and whether anybody is moderating it with that goal in mind
  91. T 1486092430 18<psztorc18> @runnerman1, That isn't true, because sometimes when the miners run a different client we still call it Bitcoin, and other times we don't.
  92. T 1486092472 18<Riiume18> Folks, sorry to interject, but I want to move us to a focal point
  93. T 1486092497 18<Riiume18> There are many great side-discussions to be had, and I encourage people to pursue those (possibly by spawning additional chans)
  94. T 1486092525 18<Riiume18> Our goal is to find a compromise palatable to both sides
  95. T 1486092529 18<Riiume18> Segwit and big blockers
  96. T 1486092548 18<Riiume18> psztorc has raised an intersting question: what does each side want?
  97. T 1486092565 20<banjomanjo>30 I think that is a red herring
  98. T 1486092569 18<Riiume18> I say we now attempt to answer this, as it will lead us more directly along our agenda
  99. T 1486092586 20<banjomanjo>30 its quite clear what they what on both sides, a functional, secure, scalable bitcoin
  100. T 1486092593 20<banjomanjo>30 its not the what thats the matter of contension
  101. T 1486092594 20<banjomanjo>30 its the how
  102. T 1486092602 18<runnerman118> I think the best solution is Segwit as a hardfork and BU
  103. T 1486092610 20<banjomanjo>30 why?
  104. T 1486092621 20<banjomanjo>30 segwit as a hardfork changes absolutely nothing except one piece of data
  105. T 1486092630 19<jwinterm> well, considering 50% of the miners don't appear to want either segwit or BU/8 MB blocks, I tend to agree with banjomanjo
  106. T 1486092631 18<ETH-Contract18> One side want a market controlled blocksize, the other want a Core/Blockstream controlled blocksize with high fee policy
  107. T 1486092633 18<jwinterm18> it's a red herring
  108. T 1486092636 20<banjomanjo>30 and BU is untested, and in my opinion fundamentally flawed
  109. T 1486092638 18<runnerman118> let miners decide thier blocksize ,and segwit can fix mallebility issues
  110. T 1486092644 18<jwinterm18> to even consider that there are only two sides is a false dichotomy
  111. T 1486092655 20<banjomanjo>30 ^^
  112. T 1486092660 18<runnerman118> I'm running a BU, it work, not flawed
  113. T 1486092668 18<gwillen18> I think it's worth keeping in mind that there are surely more than two sides, and probably casting it as two sides directly opposed to each other is ... the opposite of a best practice for creating common ground among disagreeing people :-)
  114. T 1486092668 20<banjomanjo>30 it is flawed runnerman
  115. T 1486092668 18<amiller18> how about agreeing on, what are the underlying differences in values or assumptions/premises that lead to such a difference in opinion on how to go forward?
  116. T 1486092673 18<runnerman118> How
  117. T 1486092696 23* wingman2 (~wingman2@web.innestech.net23) has joined
  118. T 1486092704 20<banjomanjo>30 every single time the blocksize is raised will either result in 1) complete chaos as reorganizations happen, and people WILL lose money
  119. T 1486092723 20<banjomanjo>30 or 2) economic activity will grind to a halt everytime miners announce an attempt at raising it to prevent that
  120. T 1486092735 20<banjomanjo>30 thats ridiculous, that is completely throwing out security guarantees and the promise of 100% uptime
  121. T 1486092742 20<banjomanjo>30 and as far as segwit as a hardfork
  122. T 1486092750 18<runnerman118> You know miners can change the blocksize what that want now
  123. T 1486092751 20<banjomanjo>30 that changes _absolutely nothing_ except the witness committment
  124. T 1486092757 20<banjomanjo>30 it moves from the coinbase TX to the blockheader
  125. T 1486092760 19<Riiume> banjomanjo, be that as it may, our job is to arrive at an agreement, distasteful/stupid as it may seem to us
  126. T 1486092765 20<banjomanjo>30 that is a beyond irrelevant matter
  127. T 1486092775 18<psztorc18> I agree with maaku7 that this is unlikely to help. I just thought I would stop by. : )
  128. T 1486092790 18<psztorc18> To compromise, two sides must get what they want.
  129. T 1486092805 18<wingman218> neither side
  130. T 1486092812 20<banjomanjo>30 there are alot more than 2 sides
  131. T 1486092828 18<runnerman118> Classic and XT are pretty mmuch dead
  132. T 1486092830 20<banjomanjo>30 some users do not want _anything_ changed in bitcoin ever
  133. T 1486092831 19<Riiume> banjomanjo the political reality is only 2 sides wield significant capital and public support.
  134. T 1486092834 20<banjomanjo>30 some do not ever want hardforks
  135. T 1486092841 20<banjomanjo>30 some want a hardfork for feature x, some for feature y
  136. T 1486092850 20<banjomanjo>30 saying "two sides" is beyond over simplifying 
  137. T 1486092858 20<banjomanjo>30 its like calling all races of people on Earth "black, or white"
  138. T 1486092869 18<runnerman118> What are your solutions to this blocksize issue?
  139. T 1486092870 18<jwinterm18> Riiume: what of the 45% of miners that signal neither segwit or BU/8MB?
  140. T 1486092872 18<gwillen18> psztorc: I'm curious to see whether it could help, but I agree it is unlikely to solve anything by itself. But I'd like to at least hear what people have to say.
  141. T 1486092874 18<jwinterm18> which side are they on?
  142. T 1486092875 20<banjomanjo>30 Riiume, you can't know that
  143. T 1486092892 20<banjomanjo>30 the amount of objective information available to speculate that with is not even close to make a definitive statement like that
  144. T 1486092894 19<psztorc> banjomanjo, I didn't say there were two sides.
  145. T 1486092894 18<gwillen18> psztorc: but I do think that casting it as 'two sides' is going to obscure things more than enlighten
  146. T 1486092897 18<jwinterm18> or is 45% of the network not significant capital?
  147. T 1486092908 20<banjomanjo>30 (was more a general statement than singling you out psztorc )
  148. T 1486092915 18<psztorc18> Reading comprehension is low.
  149. T 1486092917 18<Riiume18> jwinterm: They might be waiting for one side to assume a position of dominance before casting their vote, to avoid ending up in a conflict with the winner
  150. T 1486092922 18<iopools18> the solution is the hardfork, if you think a 1MB chain is better than a dynamic one, then don't worry, everybody will dismiss the dynamic blocksize chain
  151. T 1486092930 18<psztorc18> Anyways, good luck everyone. I'll be around.
  152. T 1486092956 18<Riiume18> psztorcs, peace, thanks for your time
  153. T 1486092962 18<runnerman118> cya
  154. T 1486093039 18<runnerman118> New season of The 100, gonna head out too
  155. T 1486093050 18<Riiume18> runnerman1, later, thank you for your thoughts!
  156. T 1486093055 18<ETH-Contract18> LAST 1000 BLOCKS: Bitcoin Unlimited blocks: 220 ( 22% ) Bitcoin Classic blocks: 10 ( 1% ) SegWit blocks: 238 ( 23.8% )
  157. T 1486093062 18<runnerman118> I'll stop by later
  158. T 1486093064 18<runnerman118> bye
  159. T 1486093067 24* runnerman1 has quit (Quit: Page closed)
  160. T 1486093086 18<wingman218> geeze thats close
  161. T 1486093091 24* iopools has quit (Quit: Page closed)
  162. T 1486093098 18<Riiume18> Let me put this question forward and everyone please answer...
  163. T 1486093110 18<Riiume18> What do you believe would be a workable compromise that over 95% of miners would accept?
  164. T 1486093130 18<wingman218> nothing
  165. T 1486093149 18<jwinterm18> not sure it exists, but maybe something along the lines of your proposal on reddit
  166. T 1486093164 18<jwinterm18> segwit plus hf to 2,4,8 MB block
  167. T 1486093209 23* BobBarker (~null@2607:5300:60:1c3f::23) has joined
  168. T 1486093227 18<BobBarker18> I'm negotiating on behalf on monero
  169. T 1486093250 18<gwillen18> Riiume: I'm not sure you're asking the right question
  170. T 1486093254 18<amiller18> do you think miners would be more active in signalling, without necessarily a full upgrade or a complicated code change, if they were asked more questions and had some way of responding to a survey? like reaching 95% to a non-binding resolution expressing support for a compromise of some kind?
  171. T 1486093255 23* elusive_vxn (8f9ff401@gateway/web/freenode/ip.143.159.244.123) has joined
  172. T 1486093286 18<ETH-Contract18> what about BU+FlexibleTransactions+XThinBlocks (plus no devs will interfere in the market decision for the max blocksize limit)
  173. T 1486093287 18<gwillen18> I think there are a lot of hard feelings right now that are going to make it difficult to get people to agree to proposals that might have been acceptable to them before this turned into a big fight
  174. T 1486093312 18<gwillen18> and that the hard feelings that the people who feel they have been disrespected are going to have to be tackled by engaging with people and getting their input
  175. T 1486093320 18<jwinterm18> ETH-Contract: seems unlikely
  176. T 1486093361 18<Riiume18> amiller, gwillen So an intermediate step, a commitment to enter a compromise process of some sort might be beneficial? Well, it's actionable at least.
  177. T 1486093381 18<gwillen18> well
  178. T 1486093390 18<wingman218> it don't like this 95% thing
  179. T 1486093391 23* NLNico (~NLNico@unaffiliated/nlnico23) has joined
  180. T 1486093395 18<gwillen18> as I think someone mentioned on the reddit thread, there have been attempts before
  181. T 1486093396 18<ETH-Contract18> who knows, the competition is open, no miner is obligated to use a particular software of any company
  182. T 1486093408 18<gwillen18> there was scaling, and some other stuff, where people were going to come to the table and talk, and they did
  183. T 1486093409 20<banjomanjo>30 ETH-Contract, flextrans doesn't even have working code
  184. T 1486093422 18<gwillen18> I was at the first scaling, there were some good conversations while people were face-to-face
  185. T 1486093431 18<gwillen18> most of which didn't seem to survive once people were back home
  186. T 1486093471 18<gwillen18> which suggests to me, among other things, that if you really want people to see eye-to-eye you need to get them in the same room rather than talking to each other in text through computers.
  187. T 1486093480 18<gwillen18> (This may be extra challenging when some of them don't speak the same language.)
  188. T 1486093499 18<elusive_vxn18> we all speak the universal language of shoe on head
  189. T 1486093506 20<banjomanjo>30 there are alot more than two sides here, compromising between two sides doesn't matter at all if you cannot accomplish the compromise without using a hardfork
  190. T 1486093512 18<gwillen18> There was supposedly some kind of "agreement" about scaling once before
  191. T 1486093524 20<banjomanjo>30 use of a hardfork requires unanimity, not just agreement between two parties
  192. T 1486093525 18<gwillen18> except immediately afterwards various people disagreed about who had agreed to what
  193. T 1486093535 20<banjomanjo>30 I have seen multiple people here just continue proposing hardforks
  194. T 1486093537 18<Riiume18> Would it be possible to make a new agreement that's binding, for instance...
  195. T 1486093551 18<Riiume18> All parties must put a substantial sum of money on deposit with some trusted counterparty
  196. T 1486093556 18<gwillen18> I'm going to have to duck out, I'm missing dinner
  197. T 1486093557 20<banjomanjo>30 no
  198. T 1486093560 20<banjomanjo>30 Riiume, just no
  199. T 1486093564 18<Riiume18> and they don't get it back until a consensus is achieved and executed on the Bitcoin network
  200. T 1486093581 18<gwillen18> I think this is interesting and will continue to keep an eye in this general direction
  201. T 1486093621 18<wingman218> So what number of hardforks are you personally willing to accept?
  202. T 1486093631 20<banjomanjo>30 doesn't matter
  203. T 1486093637 20<banjomanjo>30 hardforks require EVERYONE to agree
  204. T 1486093645 20<banjomanjo>30 you can't just ultimatum me with a hardfork
  205. T 1486093649 20<banjomanjo>30 or ultimatum developers
  206. T 1486093654 20<banjomanjo>30 that is completely out of any individual's hands
  207. T 1486093657 18<wingman218> no they don't
  208. T 1486093661 20<banjomanjo>30 yes they do
  209. T 1486093671 18<amiller18> i'm happy to see any nicely moderated discussion with a pleasant/open tone, thanks for the effort and good luck!
  210. T 1486093679 19<Riiume> banjomanjo, hard forks occur when one group of people packs up their toys and leaves the blockchain
  211. T 1486093683 20<banjomanjo>30 without that, you are not hardforking an upgrade, you are making an altcoin that is predistributed
  212. T 1486093690 18<wingman218> anyone can hardfork
  213. T 1486093701 19<Riiume> banjomanjo hardfork/altcoin is semantics
  214. T 1486093703 20<banjomanjo>30 yes, and without everyone, EVERYONE, going along, that is not an upgrade
  215. T 1486093708 20<banjomanjo>30 that is creating an alternative chain
  216. T 1486093715 18<wingman218> at any time
  217. T 1486093742 19<wingman2> banjomanjo: oh sorry I wasn't talking about an upgrade
  218. T 1486093745 20<banjomanjo>30 yes, and like I said, without the entire rest of the network following you, that irrelevant
  219. T 1486093748 20<banjomanjo>30 its creating an altcoin
  220. T 1486093760 20<banjomanjo>30 that will completely undermine Bitcoins function as a stable store of value
  221. T 1486093769 20<banjomanjo>30 its essentially counterfeiting the entire supply of Bitcoin
  222. T 1486093770 18<elusive_vxn18> excuse me, is this chat for roleplaying?
  223. T 1486093775 23* fvcxza (68c89a4b@gateway/web/freenode/ip.104.200.154.7523) has joined
  224. T 1486093778 18<wingman218> maybe but it is a hardfork
  225. T 1486093787 23* boob_barker (62d400f4@gateway/web/freenode/ip.98.212.0.24423) has joined
  226. T 1486093789 18<boob_barker18> hello
  227. T 1486093797 18<boob_barker18> im boob barker
  228. T 1486093821 22* 26ChanServ sets ban on 18boob_barker*!*@*
  229. T 1486093821 22* 26ChanServ has kicked 18boob_barker from 22#BtcNegotiate (24User is banned from this channel)
  230. T 1486093826 18<BobBarker18> that man is an imposter
  231. T 1486093845 20<banjomanjo>30 wingman2, but again, how does that in anyway help or offer something constructive?
  232. T 1486093847 23* bobbiest_barker (8f9ff401@gateway/web/freenode/ip.143.159.244.123) has joined
  233. T 1486093880 18<bobbiest_barker18> I am the real Bob Barker, BobBarker is the real imposter
  234. T 1486093957 20<banjomanjo>30 wingman2, the entire point of this is to avoid a contentious hardfork yes/
  235. T 1486093965 20<banjomanjo>30 how does pointing out that one is possible help achieve that?
  236. T 1486094012 18<awfulcrawler18> you can't stop people from forking if they want to
  237. T 1486094019 18<awfulcrawler18> it's not an aggressive act
  238. T 1486094019 23* BobLivesMatter (8f9ff401@gateway/web/freenode/ip.143.159.244.123) has joined
  239. T 1486094024 18<ETH-Contract18> what if a majority wants a contentious hardfork, just to settle more faster their transactions currently stuck in the mempool??
  240. T 1486094041 19<wingman2> banjomanjo: It doesn't, I was just saying you were wrong about a hard fork needing everyone
  241. T 1486094042 20<banjomanjo>30 awfulcrawler, 1) yes depending on the size and intent, it very well can be aggressive
  242. T 1486094048 18<awfulcrawler18> it isn't
  243. T 1486094052 20<banjomanjo>30 again, *how does that help avoid that, which is the whole purpose of this channel?*
  244. T 1486094065 19<wingman2> banjomanjo: It doesn't, I was just saying you were wrong about a hard fork needing everyone
  245. T 1486094065 20<banjomanjo>30 it does to be a successful upgrade wingman2
  246. T 1486094069 20<banjomanjo>30 otherwise its just breaking things
  247. T 1486094080 18<awfulcrawler18> it leaves the original project unchanged and branches off and does its own thing
  248. T 1486094116 18<BobLivesMatter18> Are Bobs welcome here?
  249. T 1486094149 18<ETH-Contract18> Only you Bob send Alice some btc in the Lightning Network
  250. T 1486094185 22* 26ChanServ sets ban on 18BobLivesMatter!*@*
  251. T 1486094185 22* 26ChanServ has kicked 18BobLivesMatter from 22#BtcNegotiate (24User is banned from this channel)
  252. T 1486094191 22* 26ChanServ sets ban on 18bobbiest_barker!*@*
  253. T 1486094191 22* 26ChanServ has kicked 18bobbiest_barker from 22#BtcNegotiate (24User is banned from this channel)
  254. T 1486094206 18<elusive_vxn18> that was brutal..
  255. T 1486094224 20<banjomanjo>30 awfulcrawler, its not that simple
  256. T 1486094241 18<awfulcrawler18> it is exactly that simple.
  257. T 1486094245 20<banjomanjo>30 it forks off, counterfeiting the distribution of coin supply, and introduces an incentive to attack or defend one or the other
  258. T 1486094258 20<banjomanjo>30 that is most definitely a hostile situation/action in the majority of cases
  259. T 1486094269 20<banjomanjo>30 ETH/ETC was the only example of a large chain having a contentious forks
  260. T 1486094273 18<elusive_vxn18> >hostile
  261. T 1486094276 18<awfulcrawler18> it doesn't counterfeit anything... it's a copy of a ledger
  262. T 1486094276 20<banjomanjo>30 there were speculative attacks, mining attacks, PR attacks
  263. T 1486094284 20<banjomanjo>30 i.e. a counterfeit
  264. T 1486094305 19<Riiume> Whether banjomanjo is exactly correct in his description of hard forks doesn't matter:
  265. T 1486094310 18<awfulcrawler18> not a counterfeit at all
  266. T 1486094311 20<banjomanjo>30 it is duplicating coins that cost money and energy to produce and essentially producing them for zero cost
  267. T 1486094311 18<Riiume18> our group is by definition opposed to a hard fork
  268. T 1486094314 20<banjomanjo>30 that is counterfeiting
  269. T 1486094331 18<elusive_vxn18> is anyone here in a position of negotiating, or is this roleplay?
  270. T 1486094337 18<awfulcrawler18> it's not duplicating coins but rather saying your 'score' at company B is the same as at company A
  271. T 1486094340 18<Riiume18> paul sztorc was here
  272. T 1486094344 18<Riiume18> also, Maaku7 chimed in
  273. T 1486094348 18<Riiume18> (bitcoin core dev)
  274. T 1486094359 20<banjomanjo>30 awfulcrawler, no, its duplicating tokens for free, that cost money to initially produce
  275. T 1486094369 20<banjomanjo>30 but again, lets move on
  276. T 1486094378 19<Riiume> banjomanjo, awfulcrawler, sorry to interrupt, but yea, you can continue in a private chan if you wish
  277. T 1486094382 18<Riiume18> the point is
  278. T 1486094383 18<awfulcrawler18> it's not dupicating tokens because they aren't being represented as the same thing
  279. T 1486094383 20<banjomanjo>30 someone said earlier they want Segwit as a hardfork
  280. T 1486094386 18<elusive_vxn18> so ka.
  281. T 1486094387 20<banjomanjo>30 why?
  282. T 1486094391 18<Riiume18> we don't want a hardfork, that's the point of our group
  283. T 1486094394 18<Riiume18> sorry
  284. T 1486094397 18<wingman218> Riiume: thats fine, you can choose not to hardfork
  285. T 1486094399 18<Riiume18> we don't want a DISPUTED hard fork
  286. T 1486094404 20<banjomanjo>30 ^^
  287. T 1486094412 18<Riiume18> right, so back to the main thread...
  288. T 1486094418 18<Riiume18> several ideas have been floated
  289. T 1486094424 18<Riiume18> 1) doing nothing (status quo)
  290. T 1486094450 18<Riiume18> 2) Soliciting a commitment from all miners to enter a "process of consensus", and somehow make it binding
  291. T 1486094467 18<Riiume18> (e.g. by placing their money in escrow until they implement their consensus)
  292. T 1486094500 18<Riiume18> 3) segwit + hf to either 2, 4, or 8 MB blocks
  293. T 1486094502 18<elusive_vxn18> > implying miners care about anything other than profit + security
  294. T 1486094537 18<Riiume18> elusive_vxn, we might use incentives and blackmail to compel them to enter the consensus process
  295. T 1486094539 20<banjomanjo>30 yeah they do elusive_vxn 
  296. T 1486094546 20<banjomanjo>30 keeping in line with the desires of users
  297. T 1486094553 20<banjomanjo>30 otherwise they don't make profit
  298. T 1486094573 18<ETH-Contract18> @Riiume how do you know the network can handle 2, 4 o 8MB blocks?
  299. T 1486094586 18<Riiume18> Eth, I don't, that was someone else's idea\
  300. T 1486094605 18<ETH-Contract18> Do you think the miners and the full nodes can know that?
  301. T 1486094612 18<wingman218> Riiume: any more?
  302. T 1486094627 18<awfulcrawler18> both sides agree that the network can handle 4MB blocks
  303. T 1486094629 20<banjomanjo>30 Cornell put the upper bounds of safe blocksize at 4 MB a yearish ago
  304. T 1486094636 20<banjomanjo>30 thats not both sides agreeing awfulcrawler 
  305. T 1486094641 20<banjomanjo>30 thats the result of an academic study
  306. T 1486094641 18<Riiume18> ETH-Contract, jwinterm proposed it, I believe
  307. T 1486094647 18<awfulcrawler18> segwit results in max 4MB blocksize
  308. T 1486094650 19<Riiume> banjomanjo, link to study?
  309. T 1486094654 20<banjomanjo>30 max _potential_ blocksize
  310. T 1486094657 18<awfulcrawler18> so both sides agree
  311. T 1486094686 18<awfulcrawler18> if 4MB were not ok then there would be extra restrictions in segwit
  312. T 1486094686 18<jwinterm18> I believe the study estimated that at 4 MB around 10% of nodes would drop off the network
  313. T 1486094714 20<banjomanjo>30 http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf
  314. T 1486094733 18<Riiume18> So then we need more information, i.e. what are the effects of a 10% loss of nodes on long term market cap prospects for Bitcoin?
  315. T 1486094747 18<Riiume18> I suggest somebody (maybe me?) fund that study
  316. T 1486094748 20<banjomanjo>30 4 MB is dependent on certai conditions awfulcrawler 
  317. T 1486094764 20<banjomanjo>30 to achieve a 4 MB block the entire block would have to be stuffed with just multisig transactions
  318. T 1486094782 18<awfulcrawler18> And both sides agree that this is allowable
  319. T 1486094798 20<banjomanjo>30 block weight alots 1 MB of space for TX data(inputs, outputs, legacy TX), and 3 MB for witness data(segwit signatures)
  320. T 1486094827 20<banjomanjo>30 Riiume, I linked the study: http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf
  321. T 1486094841 18<Riiume18> banjomajo, thx
  322. T 1486094845 18<Riiume18> reading
  323. T 1486094899 23* BobBorker (8f9ff401@gateway/web/freenode/ip.143.159.244.123) has joined
  324. T 1486094953 24* elusive_vxn has quit (Quit: Page closed)
  325. T 1486094953 18<ETH-Contract18> which chain will have a greater market cap: a 1MB chain with high confirmation times or a 2/4/8MB chain with lower confirmation times?
  326. T 1486094965 24* BobBorker has quit (Client Quit)
  327. T 1486094981 18<wingman218> Riiume: so why is the BU style limit not one of the options
  328. T 1486095022 20<banjomanjo>30 it is contentious, completely untested, and unstable
  329. T 1486095040 20<banjomanjo>30 it has absolutely zero defense against sybil attacks as well
  330. T 1486095055 18<Riiume18> wingman2, nobody has suggested it, but I think because they are of the belieft that there is no way to convince BitFury+BTCC+FullNodeOperators to accept it
  331. T 1486095066 18<wingman218> so you will never agree to it?
  332. T 1486095070 20<banjomanjo>30 no
  333. T 1486095082 18<Riiume18> wingman2, I would, but the groups I mentioned would not
  334. T 1486095085 20<banjomanjo>30 it removes any promise of predictably in the growth curve of costs for operaitng my node
  335. T 1486095092 20<banjomanjo>30 I would not accept that under any circumstances at all
  336. T 1486095101 20<banjomanjo>30 unless free quantum computers start raining from the sky
  337. T 1486095143 18<Riiume18> Interesting sidenote, BTC is more resistant to quantum annealing methods than Cryptonight
  338. T 1486095214 19<wingman2> banjomanjo: so you want to be able to run a node above all else?
  339. T 1486095242 20<banjomanjo>30 yes, that is non-negotiable
  340. T 1486095251 18<jwinterm18> I like the idea of a dynamic blocksize, but I think BU is not a good idea
  341. T 1486095257 20<banjomanjo>30 if I am not verifying the blockchain, bitcoin is not trustless(or trustminimized, whatever you want to say)
  342. T 1486095260 18<ETH-Contract18> LAST 1000 BLOCKS === Bitcoin Unlimited blocks: 220 ( 22% ) Bitcoin Classic blocks: 10 ( 1% ) SegWit blocks: 238 ( 23.8% ) ===
  343. T 1486095273 20<banjomanjo>30 I will not support any form of bitcoin that loses that trustless property
  344. T 1486095292 20<banjomanjo>30 that makes it no different than Paypal or a network of Banks
  345. T 1486095323 18<ETH-Contract18> interesting debate good night everyone
  346. T 1486095326 24* ETH-Contract has quit (Quit: Page closed)
  347. T 1486095432 18<wingman218> interesting i want to be able to send a transaction above all else
  348. T 1486095447 20<banjomanjo>30 well, I'm the one who has to pay to store it
  349. T 1486095457 18<wingman218> so do i
  350. T 1486095460 20<banjomanjo>30 I actually pay for your transaction forever
  351. T 1486095462 20<banjomanjo>30 you pay for it once
  352. T 1486095478 18<wingman218> i run a node
  353. T 1486095519 20<banjomanjo>30 and you have clearly just stated doing that is low on your priority list
  354. T 1486095529 20<banjomanjo>30 I came here for trustless, not a new Paypal
  355. T 1486095539 20<banjomanjo>30 as did from what I see most of the people in Bitcoin
  356. T 1486095570 18<awfulcrawler18> you run a node of your own free will you don't get to guilt btc users about it
  357. T 1486095581 18<wingman218> i wanted machine to machine transactions
  358. T 1486095661 20<banjomanjo>30 my running a node is what guarantees the rules of the network are enforced
  359. T 1486095667 20<banjomanjo>30 its not just an optional thing I do because I want to
  360. T 1486095685 20<banjomanjo>30 it is the only fully secure way to use Bitcoin, and it reinforces the lessened security for those who do not run a node
  361. T 1486095694 20<banjomanjo>30 it is not just a "thing I do", so do not trivialize it as such
  362. T 1486095716 18<awfulcrawler18> you aren't a hero so stop portaying yourself as such. Lots of people run nodes dude
  363. T 1486095722 20<banjomanjo>30 Bitcoin doesn't just magically keep its rules in force just cause
  364. T 1486095727 20<banjomanjo>30 it does because of the large number of nodes running
  365. T 1486095743 20<banjomanjo>30 the less nodes running, the more the guarantee of core rules staying the same is lessened
  366. T 1486095752 23* adam3us (~adam3us@unaffiliated/adam3us23) has joined
  367. T 1486095773 20<banjomanjo>30 I didn't portray myself as anything, I stated why I run a node and why it is important for the security of my money
  368. T 1486095780 20<banjomanjo>30 and how it helps reinforce that security for others
  369. T 1486095876 23* pero (~pero@unaffiliated/pero23) has joined
  370. T 1486095905 18<adam3us18> here's a peterotodd concept: bitcoin security depends on censor-resistant bandwidth, not raw bandwidth.
  371. T 1486095922 20<banjomanjo>30 hmm...
  372. T 1486095934 18<wingman218> That's interesting
  373. T 1486096013 20<banjomanjo>30 Yeah, I think thats a very succinct way of pointing out throughput alone is not the main value prospective 
  374. T 1486096129 20<banjomanjo>30 wingman2, you said you wanted machine to machine transactions as if they are not possible?
  375. T 1486096168 19<wingman2> 12:18 < banjomanjo> I came here for trustless, not a new Paypal
  376. T 1486096185 20<banjomanjo>30 yes, but you said that as if machine to machine payments are not possible
  377. T 1486096187 20<banjomanjo>30 they are
  378. T 1486096310 18<wingman218> its what i was interest in bitcoin for
  379. T 1486096330 18<wingman218> *interested
  380. T 1486096335 20<banjomanjo>30 well then it seems the two chief reasons we are here are not mutually exclusive by anymeans
  381. T 1486096351 20<banjomanjo>30 and in fact I would argue machine to machine payments it dependent on bitcoin staying trustless as well
  382. T 1486096372 18<wingman218> not really
  383. T 1486096387 20<banjomanjo>30 well thats good to have established, a tiny step of progress :)
  384. T 1486096447 18<wingman218> also i don't do them
  385. T 1486096509 18<wingman218> The idea of driving down the highway and negotiating with all of the access points for internet is not really possible
  386. T 1486096549 20<banjomanjo>30 O.o
  387. T 1486096552 20<banjomanjo>30 why not?
  388. T 1486096572 20<banjomanjo>30 your phone is pinging back and forth with every WAP in range of it all day
  389. T 1486096662 18<wingman218> I mean it wasn't possible until the idea of lightning networks came out
  390. T 1486096693 20<banjomanjo>30 yes it was, just not as efficiently
  391. T 1486096702 20<banjomanjo>30 Satoshi has capacity for one way payment channels built in since day one
  392. T 1486096722 20<banjomanjo>30 and even discussed how the economics of the system would make it unviable for microtransactions to stay on chain forever
  393. T 1486096748 20<banjomanjo>30 had*
  394. T 1486096839 24* pero (~pero@unaffiliated/pero24) has left ("Leaving")
  395. T 1486096890 19<wingman2> banjomanjo: It would be super inefficient for each car to create a new transaction every hundred meters with a node it has never seen before and may never see again
  396. T 1486096994 18<wingman218> Anyway it's why I was originally excited about bitcoin, but that was dashed pretty early
  397. T 1486097173 18<wingman218> i will never get what i orginally wanted from bitcoin
  398. T 1486097183 20<banjomanjo>30 yes you will :)
  399. T 1486097228 18<wingman218> i might get it from lightning
  400. T 1486097254 18<wingman218> which i support
  401. T 1486097291 18<Riiume18> wingman2, we have our preferences but... remember our task is to achieve consensus, regardless of what we personally would want
  402. T 1486097304 18<wingman218> and i support segwit
  403. T 1486097328 18<Riiume18> Then you must present a strategy by which SegWit can obtain consensus
  404. T 1486097358 18<Riiume18> I want our group to stay focused on that
  405. T 1486097365 18<wingman218> Riiume: we don't acheive consensus
  406. T 1486097384 18<wingman218> bitcoin acheives consensus
  407. T 1486097399 18<Riiume18> wingman2, correct... well let me be more precise...
  408. T 1486097418 18<Riiume18> We want over 95% of the hashpower on the existing protocol to accept the consensus
  409. T 1486097431 18<wingman218> we can't
  410. T 1486097448 18<wingman218> won't happen
  411. T 1486097453 18<Riiume18> i disagree
  412. T 1486097479 18<Riiume18> if I agreed with that, I wouldn't have formed this channel to begin with
  413. T 1486097510 18<wingman218> "12:50 < Riiume> i disagree "and thats why it won't happen
  414. T 1486097524 18<wingman218> "i disagree"
  415. T 1486097623 18<wingman218> you can't have consensus when "i disagree"
  416. T 1486097625 20<banjomanjo>30 guys, I think you are severely underestimating the support Segwit has
  417. T 1486097640 18<jwinterm18> consensus != unanimous consensus
  418. T 1486097643 20<banjomanjo>30 60% of the visible broadcasting nodes are upgraded to enforce it
  419. T 1486097664 20<banjomanjo>30 almost every major business in the space is onboard, either having completed code updates, in process with them, or intending to
  420. T 1486097688 24* fvcxza has quit (Ping timeout: 260 seconds)
  421. T 1486097689 20<banjomanjo>30 do you really think it is rational in the long term for miners to ignore pretty much the vast majority of the entire rest of the ecosystem?
  422. T 1486097701 20<banjomanjo>30 that eventually will directly conflict with their motive to turn a profit
  423. T 1486097724 20<banjomanjo>30 the network is not just the miners, and the miners don't make money without everyone else
  424. T 1486097728 18<wingman218> we have 10 months
  425. T 1486097734 20* banjomanjo 30shrugs
  426. T 1486097743 20<banjomanjo>30 softforks have taken longer than this to roll out before
  427. T 1486097801 18<wingman218> We have 10 months and then everyone has to upgrade to start signaling segwit again
  428. T 1486097901 18<wingman218> So that we don't run out of flags because of people not upgrading their software
  429. T 1486097910 18<Riiume18> <Hey all, I'm going afk, please keep the conversation directed towards the stated goal with an eye towards actionable plans.>
  430. T 1486097923 18* Riiume goes afk
  431. T 1486097952 18<awfulcrawler18> Would a 2MB hardfork be an acceptable compromise?
  432. T 1486097998 20<banjomanjo>30 instead of segwit, no
  433. T 1486098012 20<banjomanjo>30 potentially down the line if there is consensus for it, I wouldn't be opposed in practice
  434. T 1486098015 18<awfulcrawler18> 2MB + segwit?
  435. T 1486098018 20<banjomanjo>30 but that is entirely dependent on segwit
  436. T 1486098022 20<banjomanjo>30 not immediately after, no
  437. T 1486098025 20<banjomanjo>30 that wouldn't be safe
  438. T 1486098039 20<banjomanjo>30 I'd want Schnorr signatures + OWAS first
  439. T 1486098046 20<banjomanjo>30 and again, it would be entirely dependent on consensus
  440. T 1486098056 20<banjomanjo>30 if everyone is onboard, fine, but if there is contention over it no
  441. T 1486098080 20<banjomanjo>30 a hardfork without unanimity over a puny 1 MB increase that does nothing in the longrun is an insane risk to me
  442. T 1486098104 18<wingman218> Actually, about that, core should change the end date for segwit activation to five years in the future.
  443. T 1486098116 18<awfulcrawler18> 1MB increase either does nothing or is an insane risk...
  444. T 1486098129 18<awfulcrawler18> if it 'does nothing' how can it be risky?
  445. T 1486098143 24* NLNico has quit (Ping timeout: 248 seconds)
  446. T 1486098160 20<banjomanjo>30 because of the risk of consensus failure if its not unanimous
  447. T 1486098175 20<banjomanjo>30 1 MB increase in the long term is just a bandaid that will not last long at all
  448. T 1486098197 20<banjomanjo>30 not worth risking a failed hardfork over a bandaid that is in no way even close to a permenant solution
  449. T 1486098249 18<wingman218> I don't like the idea of 2mb, But I don't have a problem with a failed hard fork
  450. T 1486098268 24* Riiume (446a9f30@gateway/web/freenode/ip.68.106.159.4824) has left
  451. T 1486098273 18<awfulcrawler18> consensus isn't 100% of people agreeing to do the same thing
  452. T 1486098274 23* Riiume (446a9f30@gateway/web/freenode/ip.68.106.159.4823) has joined
  453. T 1486098276 20<banjomanjo>30 well then I think you are miscalculating the security model of the network enormously wingman2 
  454. T 1486098284 24* Riiume (446a9f30@gateway/web/freenode/ip.68.106.159.4824) has left
  455. T 1486098286 18<awfulcrawler18> the ETH / ETC fork was a form of consensus as well
  456. T 1486098290 20<banjomanjo>30 when it comes to a hardfork, yes it is awfulcrawler 
  457. T 1486098299 20<banjomanjo>30 ETH/ETC was not a form of consensus
  458. T 1486098302 20<banjomanjo>30 it was a consensus failure
  459. T 1486098312 18<awfulcrawler18> no it is an emergent property of the decentralized system
  460. T 1486098316 20<banjomanjo>30 and it resulted in two separate chains to the economic detriment of both
  461. T 1486098324 20<banjomanjo>30 no it wasn't, it was a consensus failure
  462. T 1486098329 18<awfulcrawler18> there is no 'failure', there is just the state of the system
  463. T 1486098330 20<banjomanjo>30 "emergent property" is a vague buzzword
  464. T 1486098336 20<banjomanjo>30 that means "anything that happens" essentially
  465. T 1486098340 18<awfulcrawler18> just because you don't like the outcome doesn't mean it's a failure
  466. T 1486098340 20<banjomanjo>30 thats nonsense
  467. T 1486098350 18<awfulcrawler18> 'consensus' as you use it is a vague buzzword
  468. T 1486098367 20<banjomanjo>30 no it is not, it is the network maintaining enforcement of the same consensus rules
  469. T 1486098376 20<banjomanjo>30 the network diverging is a failure of consensus
  470. T 1486098393 18<awfulcrawler18> Network A ----> Network A + network B
  471. T 1486098397 18<awfulcrawler18> Network A still exists
  472. T 1486098402 18<awfulcrawler18> diverging isfine
  473. T 1486098408 20<banjomanjo>30 no it is not
  474. T 1486098414 20<banjomanjo>30 it resulted in a loss of value for both chains
  475. T 1486098427 18<wingman218> i don't have a problem with that
  476. T 1486098428 20<banjomanjo>30 the loss of money on both sides due to the technical shortcomings not modeling diverging forks
  477. T 1486098430 20<banjomanjo>30 double spends
  478. T 1486098439 20<banjomanjo>30 well guess what wingman2, almost everyone else does
  479. T 1486098441 20<banjomanjo>30 this is money
  480. T 1486098442 18<awfulcrawler18> that's something you personally have a problem with. Code doesn't care
  481. T 1486098450 20<banjomanjo>30 if you don't care about your money maintaining its value, guess what, you are in the wrong place
  482. T 1486098473 18<wingman218> no i'm not
  483. T 1486098480 18<awfulcrawler18> If you want to compromise you have to accept that you'll have less than 100% agreement
  484. T 1486098482 20<banjomanjo>30 I came here in good faith to discuss
  485. T 1486098496 20<banjomanjo>30 and so far have been met with nothing but ultimatums for hardforks
  486. T 1486098505 20<banjomanjo>30 and a complete dismissal of factual aspects of how this system functions
  487. T 1486098525 20<banjomanjo>30 no I don't awfulcrawler 
  488. T 1486098530 20<banjomanjo>30 if we don't get agree, status quo it is
  489. T 1486098545 18<wingman218> nothing but ultimatums for hardforks no hardforks
  490. T 1486098564 18<awfulcrawler18> yeah you don't want to compromise. Topic is 'compromise to avoid contentious hardfork'
  491. T 1486098572 18<jwinterm18> no ultimatum, no ultimatum, you're the ultimatum
  492. T 1486098573 18<awfulcrawler18> good luck
  493. T 1486098588 20<banjomanjo>30 no where did I say no hardforks
  494. T 1486098597 20<banjomanjo>30 and I literally *just* laid out the conditions under which I would accept one
  495. T 1486098623 20<banjomanjo>30 that was unbelievably disingenuous of you to say
  496. T 1486098656 18<jwinterm18> if every node and 100% of hash power agrees with hardfork?
  497. T 1486098663 18<jwinterm18> that seems...unlikely
  498. T 1486098668 20* banjomanjo 30shrugs
  499. T 1486098673 18<awfulcrawler18> did I say you said 'no hardforks'...(because I didn't)
  500. T 1486098690 20<banjomanjo>30 kicking people off the network they bought into is a complete betrayal of the social contract of bitcoin in my mind
  501. T 1486098720 18<awfulcrawler18> forking doesn't kick people off though
  502. T 1486098736 20<banjomanjo>30 yes it does, again, you act like the value of a token is irrelevant
  503. T 1486098737 20<banjomanjo>30 its not
  504. T 1486098754 18<awfulcrawler18> uh...it doesn't. Fork = network A -> network A + network B
  505. T 1486098758 18<awfulcrawler18> network A still exists
  506. T 1486098762 18<awfulcrawler18> stay on network A
  507. T 1486098773 20<banjomanjo>30 you are kicking people off the network in terms of economic network effect
  508. T 1486098783 20<banjomanjo>30 effectively the same thing
  509. T 1486098785 18<wingman218> yes
  510. T 1486098785 18<awfulcrawler18> no people optionally move from A to B
  511. T 1486098816 20<banjomanjo>30 and again, I've been over this, its not that simple
  512. T 1486098824 20<banjomanjo>30 you ridiculously oversimplify things
  513. T 1486098835 20<banjomanjo>30 a diverging fork is counterfeiting a token and its distribution
  514. T 1486098842 20<banjomanjo>30 those tokens cost electricity and energy to produce
  515. T 1486098851 20<banjomanjo>30 and they are being replicated for essentially free to speculate on
  516. T 1486098854 20<banjomanjo>30 i.e. counterfeiting
  517. T 1486098901 18<awfulcrawler18> I will launch a new blockchain called dollaridoodle and give all btc holders one dollaridoodle for every bitcoin they have. Have I counterfeited anything?
  518. T 1486098969 20<banjomanjo>30 that is not a rational argument, that is a description of a totally different thing
  519. T 1486098982 20<banjomanjo>30 its a new token, it is not a diverging fork claiming to be the original network and token
  520. T 1486098991 20<banjomanjo>30 that is a wild conflation of two different things
  521. T 1486098994 18<awfulcrawler18> it's a thought experiment which describes the same thing as a PoW fork
  522. T 1486099000 20<banjomanjo>30 no it doesn't
  523. T 1486099001 20<banjomanjo>30 at all
  524. T 1486099009 18<awfulcrawler18> it does...exactly
  525. T 1486099013 20<banjomanjo>30 no it doesn't 
  526. T 1486099019 18<awfulcrawler18> this is brick-wall stuff
  527. T 1486099024 18<awfulcrawler18> I'll shut up now don't worry
  528. T 1486099026 20<banjomanjo>30 its describes a network that results in the same distribution 
  529. T 1486099028 20<banjomanjo>30 it is not the same thing
  530. T 1486099038 20<banjomanjo>30 dude, you are the brick wall here
  531. T 1486099046 20<banjomanjo>30 I am actually explaining reasoning behind what I say
  532. T 1486099051 20<banjomanjo>30 you just keep asserting things
  533. T 1486099058 20<banjomanjo>30 and do not provide any rationalization for it
  534. T 1486099085 18<wingman218> anyway
  535. T 1486099106 18<wingman218> 12:58 < Riiume> <Hey all, I'm going afk, please keep the conversation directed towards the stated goal with an eye towards
  536. T 1486099109 18<wingman218> actionable plans.>
  537. T 1486099114 18<wingman218> so
  538. T 1486099217 19<wingman2> you like hard forks or not awfulcrawler and banjomanjo doesn't matter. what can we do to make this go smoothly
  539. T 1486099238 18<wingman218> core should change the end date for segwit activation to five years in the future.
  540. T 1486099247 20<banjomanjo>30 thats not how that works
  541. T 1486099253 18<wingman218> what do you think
  542. T 1486099255 20<banjomanjo>30 that in itself would require a softfork I believe
  543. T 1486099340 20<banjomanjo>30 core can't just "change things"
  544. T 1486099347 20<banjomanjo>30 people have to choose to run their updated software
  545. T 1486099353 20<banjomanjo>30 they have no control whatsoever beyond writing code
  546. T 1486099358 20<banjomanjo>30 the rest is entirely up to the users
  547. T 1486099372 18<wingman218> core should change the end date for the possibility of a segwit activation to five years in the future. not the ~10 months it is now.
  548. T 1486099380 20<banjomanjo>30 they can't do that!
  549. T 1486099389 20<banjomanjo>30 1) people would have to update their clients
  550. T 1486099398 20<banjomanjo>30 2) miners would have to softfork to modify BIP9
  551. T 1486099405 20<banjomanjo>30 they can't just change shit
  552. T 1486099447 18<wingman218> core should change the end date for the possibility of a segwit activation to five years in the future *with a new flag*. not the ~10 months it is now.
  553. T 1486099528 19<wingman2> banjomanjo: https://github.com/bitcoin/bitcoin/blob/812714fd80e96e28cd288c553c83838cecbfc2d9/src/chainparams.cpp#L97
  554. T 1486099599 18<wingman218> We have to start everything again after November 15th, 2017.
  555. T 1486099615 18<wingman218> Everyone has to start signaling again.
  556. T 1486099726 20<banjomanjo>30 this is ridiculous
  557. T 1486099734 20<banjomanjo>30 you want a "compromise" to avoid a contentious hardfork
  558. T 1486099750 20<banjomanjo>30 but you want to take the SF solution to things right now, and set it up to hopefully take longer?
  559. T 1486099805 20<banjomanjo>30 thats effing ridiculous
  560. T 1486099808 18<wingman218> no That's not what that
  561. T 1486099831 18<wingman218> It's literally the date segwit dies
  562. T 1486099834 20<banjomanjo>30 no its not
  563. T 1486099842 20<banjomanjo>30 its the date the bit has to be reset
  564. T 1486099852 18<wingman218> well yes
  565. T 1486099868 18<wingman218> And then it has to be done all over again
  566. T 1486099897 18<wingman218> everyone has to upgrade again
  567. T 1486100023 20<banjomanjo>30 so?
  568. T 1486100032 18<wingman218> Without that end date, the slow march of time will get us segwit
  569. T 1486100033 20<banjomanjo>30 you're suggestion requires updating again anyway
  570. T 1486100036 20<banjomanjo>30 wait til then
  571. T 1486100093 18<wingman218> Just by everyone and eventually having to upgrade
  572. T 1486100104 18<wingman218> *Just by everyone eventually having to upgrade
  573. T 1486100220 18<jwinterm18> thread got scrubbed: https://www.reddit.com/r/Bitcoin/comments/5rqvwc/btcnegotiate_on_freenode_a_working_group_to/
  574. T 1486100231 18<jwinterm18> imagine that
  575. T 1486100241 18<wingman218> bu will die over time if they start having problems keeping up with security fixes and such
  576. T 1486100348 18<adam3us18> bitcoin is going through updates anyway, 0.14 is coming in a few months.
  577. T 1486100448 18<wingman218> i really don't like that end date
  578. T 1486100811 23* vogelito (~Adium@fixed-190-149-187-190-149-151.iusacell.net23) has joined
  579. T 1486100899 18<vogelito18> Good night. It would be nice if the logs included the nick, otherwise pretty hard to readT 1486091235 22* Topic for 22#BtcNegotiate set by 26Riiume (24Thu Feb 2 20:22:23 2017)
  580. T 1486091255 18<ETH-Contract18> Hard numbers: look at the current mempool stats
  581. T 1486091264 18<runnerman118> Please how can prove the financial impact of something that does not exist
  582. T 1486091265 20<banjomanjo>30 I have arrived, the negotiations may commence/s
  583. T 1486091266 18<Riiume18> I ask the same of the Blockstream side, although they have done a bit in this direction (Testnet, although it's play money so lacking the real financial incentives that govern Bitcoin)
  584. T 1486091286 20<banjomanjo>30 on a serious note though, is there anywhere previous discussions here are posted?
  585. T 1486091305 20<banjomanjo>30 (nm)
  586. T 1486091309 18<Riiume18> runnerman1, by modeling it. I don't have to shoot myself to create a simulation of the effects and see that it doesn't work out well for me
  587. T 1486091319 18<Riiume18> yes, logs
  588. T 1486091322 18<Riiume18> https://notebin.cc/1kdbou2efi8
  589. T 1486091334 20<banjomanjo>30 thanks
  590. T 1486091360 18<runnerman118> I dont even know what to do, I'm not a "professional bitcoiner"
  591. T 1486091428 18<Riiume18> runnerman1, yea it's difficult. Basically the various interest groups need to come together and fund studies proving/disproving their beliefs regarding the financial impact of taking Bitcoin one way or the other
  592. T 1486091450 18<Riiume18> Start a funding campaign to pay some academics to carry out the study, e.g.
  593. T 1486091504 18<psztorc18> Hi Riiume, I actually published some way of measuring something like that: http://www.truthcoin.info/blog/win-win-blocksize/
  594. T 1486091513 18<Riiume18> ahh, sweet! checking it out!
  595. T 1486091515 18<psztorc18> However, I think there is something more important to talk about
  596. T 1486091526 18<psztorc18> which is: "What do you think the miners want?"
  597. T 1486091560 18<Riiume18> Complete control of Bitcoin, without actually destroying Bitcoin's value
  598. T 1486091567 18<psztorc18> because they said they wanted a 2 MB blocksize around ~Dec 2015, and that's what they sort've got with segwit.
  599. T 1486091584 18<psztorc18> Well, are they going to get "Complete control of Bitcoin"?
  600. T 1486091601 23* gwillen (~gwillen@unaffiliated/gwillen23) has joined
  601. T 1486091624 18<Riiume18> No, if they remain persistent, the prediction markets say they get a hard fork, and there hasn't been much analysis of what occurs past that point
  602. T 1486091626 18<runnerman118> Of course miners are in control of bitcoin, it should always be that way
  603. T 1486091636 24* pero (~pero@unaffiliated/pero24) has left ("Leaving")
  604. T 1486091640 18<Riiume18> runnerman1, Full nodes wield substantial influence too
  605. T 1486091642 18<ETH-Contract18> Where is Adam the individual?
  606. T 1486091664 20<banjomanjo>30 alpalp, before doing that, I'm proposing we study it; create a simulation of Bitcoin forking in the near future and attempt to accurately model the economic effects
  607. T 1486091667 20<banjomanjo>30 who said that ^
  608. T 1486091670 18<runnerman118> Nope not as much as miners
  609. T 1486091672 20<banjomanjo>30 log structure is not clear
  610. T 1486091693 20<banjomanjo>30 but that is literally impossible, if you could do that you could model markets accurately, and that would have profound effects for the economy as a whole
  611. T 1486091697 18<psztorc18> Miners will not get "Complete control of Bitcoin", because no one can give that to them (or to themselves).
  612. T 1486091715 19<runnerman1> @banjomanjo exactly
  613. T 1486091719 18<psztorc18> So it seems that, according to you, they want something that they can't have.
  614. T 1486091723 23* jwinterm (~quassel@unaffiliated/jwinterm23) has joined
  615. T 1486091789 23* awfulcrawler (~awfulcraw@122-57-26-40.jetstream.xtra.co.nz23) has joined
  616. T 1486091791 18<runnerman118> @psztorc what are you talking about. Miners pretty much dictate what bitcoin is
  617. T 1486091799 19<Riiume> @banjomanjo, study doesn't have to have perfect predictivity; Correlations and extrapolations are always within some confidence interval
  618. T 1486091813 18<runnerman118> If all miners decide to stop mining we have no bitcoin.
  619. T 1486091822 18<psztorc18> @runnerman1, if "Miners dictate what bitcoin is" and "miners want a 2MB hard fork", then why don't the miners have the 2mb fork they wanted?
  620. T 1486091830 20<banjomanjo>30 [21:41] Point is the whole reason for this blocksize debate is political. If LN gets activated those fees will be going directly to Blockstream.
  621. T 1486091832 20<banjomanjo>30 and who said that
  622. T 1486091836 20<banjomanjo>30 that is also incorrect
  623. T 1486091848 18<runnerman118> If miners decide tomorrow to run a client that give everyone 50 million bitcoin, that what we will have
  624. T 1486091857 20<banjomanjo>30 Riome, you are not going to accurately model a market composed of 4-10 million people anywhere close to accurately
  625. T 1486091862 20<banjomanjo>30 the data from that would be useless/garbage
  626. T 1486091872 18<Riiume18> psztorc, hehe I like where you're going. So what DO the miners want then? Or are they actually irrational?
  627. T 1486091876 20<banjomanjo>30 and runnerman1...no...
  628. T 1486091879 20<banjomanjo>30 thats not how that works
  629. T 1486091897 20<banjomanjo>30 if the exchanges and economy don't support that, all miners are oding is wasting electricity mining worthless coins no one else will buy
  630. T 1486091918 18<psztorc18> @runnerman1, do you consider Litecoin blocks to be Bitcoin blocks?
  631. T 1486091932 18<runnerman118> Oh course...the price will immedially crash lol
  632. T 1486091954 24* dongcarl has quit (Quit: leaving)
  633. T 1486091956 20<banjomanjo>30 thats just your speculation
  634. T 1486091966 20<banjomanjo>30 many would have said that about a BU block being minted at > 1 MB
  635. T 1486091969 18<ETH-Contract18> How can a miner will not be incentivized to settle on-chain the current mempool (and get the reward) with a simple upgrade to a today's technologically-possible 4MB max blocksize limit
  636. T 1486091970 20<banjomanjo>30 that happened, the price didn't crash
  637. T 1486091998 18<psztorc18> @runnerman1, If miners decide tomorrow to run a client that gives everyone 50 million litecoin, will Bitcoin be equal to a client that gives everyone 50 million litecoin?
  638. T 1486092062 18<runnerman118> What?? Litecoin and bitcoin are two different things
  639. T 1486092081 18<psztorc18> By what criterion?
  640. T 1486092109 20<banjomanjo>30 what is with all the repetition of LN fees going to Blockstream?
  641. T 1486092114 20<banjomanjo>30 that is completely untrue
  642. T 1486092139 20<banjomanjo>30 fees go to whoever is facilitating payments, i.e. everyone in the middle of a path through a set of channels
  643. T 1486092142 18<runnerman118> Oh were are the fees going to then?
  644. T 1486092148 20<banjomanjo>30 which requires locking coins up
  645. T 1486092161 20<banjomanjo>30 which if you actually sit and think for a minute, benefits the miners most out of anyone
  646. T 1486092184 20<banjomanjo>30 they as a group have both the largest stash of coins to provide liquidity with, and as miners can use their own blockspace to guarantee smooth channel closures and openings
  647. T 1486092195 20<banjomanjo>30 LN would allow miners to suck up most of those fees, not Blockstream
  648. T 1486092207 18<ETH-Contract18> i don't want my transaction to be processed by a set of middle mans through a set of channel, I just want my transaction to be settle on-chain in the next 10 minutes, simple
  649. T 1486092219 20<banjomanjo>30 then you don't have to use LN
  650. T 1486092219 18<runnerman118> ^
  651. T 1486092223 18<psztorc18> It is possible that LN will decrease on-chain fees. And it is possible that Blockstream will run a LN hub that will attract customers.
  652. T 1486092224 20<banjomanjo>30 have fun
  653. T 1486092235 18<psztorc18> runnerman1, Are you avoiding my questions?
  654. T 1486092238 20<banjomanjo>30 I on the other hand will use it if it makes sense at the time and is cost effective
  655. T 1486092248 18<runnerman118> which question
  656. T 1486092253 18<runnerman118> I might of missed it
  657. T 1486092267 18<psztorc18> I see.
  658. T 1486092325 18<runnerman118> The criterion one? They are different because the clients/node the miners are running are different
  659. T 1486092325 18<psztorc18> @Riiume, I think I know what the miners want. But you can't broker a compromise unless you give them what they want, and you probably can't do that unless you know what they want.
  660. T 1486092328 23* iopools (b53ea498@gateway/web/freenode/ip.181.62.164.15223) has joined
  661. T 1486092335 23* amiller (~socrates1@unaffiliated/socrates102423) has joined
  662. T 1486092351 20<banjomanjo>30 Riiume, what do you mean "sidechains will be able to affect the main blockchain willy nilly" ?
  663. T 1486092353 18<Riiume18> psztorc, please give us your view on what they want!
  664. T 1486092372 18<ETH-Contract18> explain that please
  665. T 1486092374 18<jwinterm18> I bet it's hookers and blow
  666. T 1486092386 18<runnerman118> Who knows, everyone wants something different
  667. T 1486092411 18<gwillen18> I am curious what this channel is expectd to accomplish
  668. T 1486092418 18<gwillen18> and whether anybody is moderating it with that goal in mind
  669. T 1486092430 18<psztorc18> @runnerman1, That isn't true, because sometimes when the miners run a different client we still call it Bitcoin, and other times we don't.
  670. T 1486092472 18<Riiume18> Folks, sorry to interject, but I want to move us to a focal point
  671. T 1486092497 18<Riiume18> There are many great side-discussions to be had, and I encourage people to pursue those (possibly by spawning additional chans)
  672. T 1486092525 18<Riiume18> Our goal is to find a compromise palatable to both sides
  673. T 1486092529 18<Riiume18> Segwit and big blockers
  674. T 1486092548 18<Riiume18> psztorc has raised an intersting question: what does each side want?
  675. T 1486092565 20<banjomanjo>30 I think that is a red herring
  676. T 1486092569 18<Riiume18> I say we now attempt to answer this, as it will lead us more directly along our agenda
  677. T 1486092586 20<banjomanjo>30 its quite clear what they what on both sides, a functional, secure, scalable bitcoin
  678. T 1486092593 20<banjomanjo>30 its not the what thats the matter of contension
  679. T 1486092594 20<banjomanjo>30 its the how
  680. T 1486092602 18<runnerman118> I think the best solution is Segwit as a hardfork and BU
  681. T 1486092610 20<banjomanjo>30 why?
  682. T 1486092621 20<banjomanjo>30 segwit as a hardfork changes absolutely nothing except one piece of data
  683. T 1486092630 19<jwinterm> well, considering 50% of the miners don't appear to want either segwit or BU/8 MB blocks, I tend to agree with banjomanjo
  684. T 1486092631 18<ETH-Contract18> One side want a market controlled blocksize, the other want a Core/Blockstream controlled blocksize with high fee policy
  685. T 1486092633 18<jwinterm18> it's a red herring
  686. T 1486092636 20<banjomanjo>30 and BU is untested, and in my opinion fundamentally flawed
  687. T 1486092638 18<runnerman118> let miners decide thier blocksize ,and segwit can fix mallebility issues
  688. T 1486092644 18<jwinterm18> to even consider that there are only two sides is a false dichotomy
  689. T 1486092655 20<banjomanjo>30 ^^
  690. T 1486092660 18<runnerman118> I'm running a BU, it work, not flawed
  691. T 1486092668 18<gwillen18> I think it's worth keeping in mind that there are surely more than two sides, and probably casting it as two sides directly opposed to each other is ... the opposite of a best practice for creating common ground among disagreeing people :-)
  692. T 1486092668 20<banjomanjo>30 it is flawed runnerman
  693. T 1486092668 18<amiller18> how about agreeing on, what are the underlying differences in values or assumptions/premises that lead to such a difference in opinion on how to go forward?
  694. T 1486092673 18<runnerman118> How
  695. T 1486092696 23* wingman2 (~wingman2@web.innestech.net23) has joined
  696. T 1486092704 20<banjomanjo>30 every single time the blocksize is raised will either result in 1) complete chaos as reorganizations happen, and people WILL lose money
  697. T 1486092723 20<banjomanjo>30 or 2) economic activity will grind to a halt everytime miners announce an attempt at raising it to prevent that
  698. T 1486092735 20<banjomanjo>30 thats ridiculous, that is completely throwing out security guarantees and the promise of 100% uptime
  699. T 1486092742 20<banjomanjo>30 and as far as segwit as a hardfork
  700. T 1486092750 18<runnerman118> You know miners can change the blocksize what that want now
  701. T 1486092751 20<banjomanjo>30 that changes _absolutely nothing_ except the witness committment
  702. T 1486092757 20<banjomanjo>30 it moves from the coinbase TX to the blockheader
  703. T 1486092760 19<Riiume> banjomanjo, be that as it may, our job is to arrive at an agreement, distasteful/stupid as it may seem to us
  704. T 1486092765 20<banjomanjo>30 that is a beyond irrelevant matter
  705. T 1486092775 18<psztorc18> I agree with maaku7 that this is unlikely to help. I just thought I would stop by. : )
  706. T 1486092790 18<psztorc18> To compromise, two sides must get what they want.
  707. T 1486092805 18<wingman218> neither side
  708. T 1486092812 20<banjomanjo>30 there are alot more than 2 sides
  709. T 1486092828 18<runnerman118> Classic and XT are pretty mmuch dead
  710. T 1486092830 20<banjomanjo>30 some users do not want _anything_ changed in bitcoin ever
  711. T 1486092831 19<Riiume> banjomanjo the political reality is only 2 sides wield significant capital and public support.
  712. T 1486092834 20<banjomanjo>30 some do not ever want hardforks
  713. T 1486092841 20<banjomanjo>30 some want a hardfork for feature x, some for feature y
  714. T 1486092850 20<banjomanjo>30 saying "two sides" is beyond over simplifying 
  715. T 1486092858 20<banjomanjo>30 its like calling all races of people on Earth "black, or white"
  716. T 1486092869 18<runnerman118> What are your solutions to this blocksize issue?
  717. T 1486092870 18<jwinterm18> Riiume: what of the 45% of miners that signal neither segwit or BU/8MB?
  718. T 1486092872 18<gwillen18> psztorc: I'm curious to see whether it could help, but I agree it is unlikely to solve anything by itself. But I'd like to at least hear what people have to say.
  719. T 1486092874 18<jwinterm18> which side are they on?
  720. T 1486092875 20<banjomanjo>30 Riiume, you can't know that
  721. T 1486092892 20<banjomanjo>30 the amount of objective information available to speculate that with is not even close to make a definitive statement like that
  722. T 1486092894 19<psztorc> banjomanjo, I didn't say there were two sides.
  723. T 1486092894 18<gwillen18> psztorc: but I do think that casting it as 'two sides' is going to obscure things more than enlighten
  724. T 1486092897 18<jwinterm18> or is 45% of the network not significant capital?
  725. T 1486092908 20<banjomanjo>30 (was more a general statement than singling you out psztorc )
  726. T 1486092915 18<psztorc18> Reading comprehension is low.
  727. T 1486092917 18<Riiume18> jwinterm: They might be waiting for one side to assume a position of dominance before casting their vote, to avoid ending up in a conflict with the winner
  728. T 1486092922 18<iopools18> the solution is the hardfork, if you think a 1MB chain is better than a dynamic one, then don't worry, everybody will dismiss the dynamic blocksize chain
  729. T 1486092930 18<psztorc18> Anyways, good luck everyone. I'll be around.
  730. T 1486092956 18<Riiume18> psztorcs, peace, thanks for your time
  731. T 1486092962 18<runnerman118> cya
  732. T 1486093039 18<runnerman118> New season of The 100, gonna head out too
  733. T 1486093050 18<Riiume18> runnerman1, later, thank you for your thoughts!
  734. T 1486093055 18<ETH-Contract18> LAST 1000 BLOCKS: Bitcoin Unlimited blocks: 220 ( 22% ) Bitcoin Classic blocks: 10 ( 1% ) SegWit blocks: 238 ( 23.8% )
  735. T 1486093062 18<runnerman118> I'll stop by later
  736. T 1486093064 18<runnerman118> bye
  737. T 1486093067 24* runnerman1 has quit (Quit: Page closed)
  738. T 1486093086 18<wingman218> geeze thats close
  739. T 1486093091 24* iopools has quit (Quit: Page closed)
  740. T 1486093098 18<Riiume18> Let me put this question forward and everyone please answer...
  741. T 1486093110 18<Riiume18> What do you believe would be a workable compromise that over 95% of miners would accept?
  742. T 1486093130 18<wingman218> nothing
  743. T 1486093149 18<jwinterm18> not sure it exists, but maybe something along the lines of your proposal on reddit
  744. T 1486093164 18<jwinterm18> segwit plus hf to 2,4,8 MB block
  745. T 1486093209 23* BobBarker (~null@2607:5300:60:1c3f::23) has joined
  746. T 1486093227 18<BobBarker18> I'm negotiating on behalf on monero
  747. T 1486093250 18<gwillen18> Riiume: I'm not sure you're asking the right question
  748. T 1486093254 18<amiller18> do you think miners would be more active in signalling, without necessarily a full upgrade or a complicated code change, if they were asked more questions and had some way of responding to a survey? like reaching 95% to a non-binding resolution expressing support for a compromise of some kind?
  749. T 1486093255 23* elusive_vxn (8f9ff401@gateway/web/freenode/ip.143.159.244.123) has joined
  750. T 1486093286 18<ETH-Contract18> what about BU+FlexibleTransactions+XThinBlocks (plus no devs will interfere in the market decision for the max blocksize limit)
  751. T 1486093287 18<gwillen18> I think there are a lot of hard feelings right now that are going to make it difficult to get people to agree to proposals that might have been acceptable to them before this turned into a big fight
  752. T 1486093312 18<gwillen18> and that the hard feelings that the people who feel they have been disrespected are going to have to be tackled by engaging with people and getting their input
  753. T 1486093320 18<jwinterm18> ETH-Contract: seems unlikely
  754. T 1486093361 18<Riiume18> amiller, gwillen So an intermediate step, a commitment to enter a compromise process of some sort might be beneficial? Well, it's actionable at least.
  755. T 1486093381 18<gwillen18> well
  756. T 1486093390 18<wingman218> it don't like this 95% thing
  757. T 1486093391 23* NLNico (~NLNico@unaffiliated/nlnico23) has joined
  758. T 1486093395 18<gwillen18> as I think someone mentioned on the reddit thread, there have been attempts before
  759. T 1486093396 18<ETH-Contract18> who knows, the competition is open, no miner is obligated to use a particular software of any company
  760. T 1486093408 18<gwillen18> there was scaling, and some other stuff, where people were going to come to the table and talk, and they did
  761. T 1486093409 20<banjomanjo>30 ETH-Contract, flextrans doesn't even have working code
  762. T 1486093422 18<gwillen18> I was at the first scaling, there were some good conversations while people were face-to-face
  763. T 1486093431 18<gwillen18> most of which didn't seem to survive once people were back home
  764. T 1486093471 18<gwillen18> which suggests to me, among other things, that if you really want people to see eye-to-eye you need to get them in the same room rather than talking to each other in text through computers.
  765. T 1486093480 18<gwillen18> (This may be extra challenging when some of them don't speak the same language.)
  766. T 1486093499 18<elusive_vxn18> we all speak the universal language of shoe on head
  767. T 1486093506 20<banjomanjo>30 there are alot more than two sides here, compromising between two sides doesn't matter at all if you cannot accomplish the compromise without using a hardfork
  768. T 1486093512 18<gwillen18> There was supposedly some kind of "agreement" about scaling once before
  769. T 1486093524 20<banjomanjo>30 use of a hardfork requires unanimity, not just agreement between two parties
  770. T 1486093525 18<gwillen18> except immediately afterwards various people disagreed about who had agreed to what
  771. T 1486093535 20<banjomanjo>30 I have seen multiple people here just continue proposing hardforks
  772. T 1486093537 18<Riiume18> Would it be possible to make a new agreement that's binding, for instance...
  773. T 1486093551 18<Riiume18> All parties must put a substantial sum of money on deposit with some trusted counterparty
  774. T 1486093556 18<gwillen18> I'm going to have to duck out, I'm missing dinner
  775. T 1486093557 20<banjomanjo>30 no
  776. T 1486093560 20<banjomanjo>30 Riiume, just no
  777. T 1486093564 18<Riiume18> and they don't get it back until a consensus is achieved and executed on the Bitcoin network
  778. T 1486093581 18<gwillen18> I think this is interesting and will continue to keep an eye in this general direction
  779. T 1486093621 18<wingman218> So what number of hardforks are you personally willing to accept?
  780. T 1486093631 20<banjomanjo>30 doesn't matter
  781. T 1486093637 20<banjomanjo>30 hardforks require EVERYONE to agree
  782. T 1486093645 20<banjomanjo>30 you can't just ultimatum me with a hardfork
  783. T 1486093649 20<banjomanjo>30 or ultimatum developers
  784. T 1486093654 20<banjomanjo>30 that is completely out of any individual's hands
  785. T 1486093657 18<wingman218> no they don't
  786. T 1486093661 20<banjomanjo>30 yes they do
  787. T 1486093671 18<amiller18> i'm happy to see any nicely moderated discussion with a pleasant/open tone, thanks for the effort and good luck!
  788. T 1486093679 19<Riiume> banjomanjo, hard forks occur when one group of people packs up their toys and leaves the blockchain
  789. T 1486093683 20<banjomanjo>30 without that, you are not hardforking an upgrade, you are making an altcoin that is predistributed
  790. T 1486093690 18<wingman218> anyone can hardfork
  791. T 1486093701 19<Riiume> banjomanjo hardfork/altcoin is semantics
  792. T 1486093703 20<banjomanjo>30 yes, and without everyone, EVERYONE, going along, that is not an upgrade
  793. T 1486093708 20<banjomanjo>30 that is creating an alternative chain
  794. T 1486093715 18<wingman218> at any time
  795. T 1486093742 19<wingman2> banjomanjo: oh sorry I wasn't talking about an upgrade
  796. T 1486093745 20<banjomanjo>30 yes, and like I said, without the entire rest of the network following you, that irrelevant
  797. T 1486093748 20<banjomanjo>30 its creating an altcoin
  798. T 1486093760 20<banjomanjo>30 that will completely undermine Bitcoins function as a stable store of value
  799. T 1486093769 20<banjomanjo>30 its essentially counterfeiting the entire supply of Bitcoin
  800. T 1486093770 18<elusive_vxn18> excuse me, is this chat for roleplaying?
  801. T 1486093775 23* fvcxza (68c89a4b@gateway/web/freenode/ip.104.200.154.7523) has joined
  802. T 1486093778 18<wingman218> maybe but it is a hardfork
  803. T 1486093787 23* boob_barker (62d400f4@gateway/web/freenode/ip.98.212.0.24423) has joined
  804. T 1486093789 18<boob_barker18> hello
  805. T 1486093797 18<boob_barker18> im boob barker
  806. T 1486093821 22* 26ChanServ sets ban on 18boob_barker*!*@*
  807. T 1486093821 22* 26ChanServ has kicked 18boob_barker from 22#BtcNegotiate (24User is banned from this channel)
  808. T 1486093826 18<BobBarker18> that man is an imposter
  809. T 1486093845 20<banjomanjo>30 wingman2, but again, how does that in anyway help or offer something constructive?
  810. T 1486093847 23* bobbiest_barker (8f9ff401@gateway/web/freenode/ip.143.159.244.123) has joined
  811. T 1486093880 18<bobbiest_barker18> I am the real Bob Barker, BobBarker is the real imposter
  812. T 1486093957 20<banjomanjo>30 wingman2, the entire point of this is to avoid a contentious hardfork yes/
  813. T 1486093965 20<banjomanjo>30 how does pointing out that one is possible help achieve that?
  814. T 1486094012 18<awfulcrawler18> you can't stop people from forking if they want to
  815. T 1486094019 18<awfulcrawler18> it's not an aggressive act
  816. T 1486094019 23* BobLivesMatter (8f9ff401@gateway/web/freenode/ip.143.159.244.123) has joined
  817. T 1486094024 18<ETH-Contract18> what if a majority wants a contentious hardfork, just to settle more faster their transactions currently stuck in the mempool??
  818. T 1486094041 19<wingman2> banjomanjo: It doesn't, I was just saying you were wrong about a hard fork needing everyone
  819. T 1486094042 20<banjomanjo>30 awfulcrawler, 1) yes depending on the size and intent, it very well can be aggressive
  820. T 1486094048 18<awfulcrawler18> it isn't
  821. T 1486094052 20<banjomanjo>30 again, *how does that help avoid that, which is the whole purpose of this channel?*
  822. T 1486094065 19<wingman2> banjomanjo: It doesn't, I was just saying you were wrong about a hard fork needing everyone
  823. T 1486094065 20<banjomanjo>30 it does to be a successful upgrade wingman2
  824. T 1486094069 20<banjomanjo>30 otherwise its just breaking things
  825. T 1486094080 18<awfulcrawler18> it leaves the original project unchanged and branches off and does its own thing
  826. T 1486094116 18<BobLivesMatter18> Are Bobs welcome here?
  827. T 1486094149 18<ETH-Contract18> Only you Bob send Alice some btc in the Lightning Network
  828. T 1486094185 22* 26ChanServ sets ban on 18BobLivesMatter!*@*
  829. T 1486094185 22* 26ChanServ has kicked 18BobLivesMatter from 22#BtcNegotiate (24User is banned from this channel)
  830. T 1486094191 22* 26ChanServ sets ban on 18bobbiest_barker!*@*
  831. T 1486094191 22* 26ChanServ has kicked 18bobbiest_barker from 22#BtcNegotiate (24User is banned from this channel)
  832. T 1486094206 18<elusive_vxn18> that was brutal..
  833. T 1486094224 20<banjomanjo>30 awfulcrawler, its not that simple
  834. T 1486094241 18<awfulcrawler18> it is exactly that simple.
  835. T 1486094245 20<banjomanjo>30 it forks off, counterfeiting the distribution of coin supply, and introduces an incentive to attack or defend one or the other
  836. T 1486094258 20<banjomanjo>30 that is most definitely a hostile situation/action in the majority of cases
  837. T 1486094269 20<banjomanjo>30 ETH/ETC was the only example of a large chain having a contentious forks
  838. T 1486094273 18<elusive_vxn18> >hostile
  839. T 1486094276 18<awfulcrawler18> it doesn't counterfeit anything... it's a copy of a ledger
  840. T 1486094276 20<banjomanjo>30 there were speculative attacks, mining attacks, PR attacks
  841. T 1486094284 20<banjomanjo>30 i.e. a counterfeit
  842. T 1486094305 19<Riiume> Whether banjomanjo is exactly correct in his description of hard forks doesn't matter:
  843. T 1486094310 18<awfulcrawler18> not a counterfeit at all
  844. T 1486094311 20<banjomanjo>30 it is duplicating coins that cost money and energy to produce and essentially producing them for zero cost
  845. T 1486094311 18<Riiume18> our group is by definition opposed to a hard fork
  846. T 1486094314 20<banjomanjo>30 that is counterfeiting
  847. T 1486094331 18<elusive_vxn18> is anyone here in a position of negotiating, or is this roleplay?
  848. T 1486094337 18<awfulcrawler18> it's not duplicating coins but rather saying your 'score' at company B is the same as at company A
  849. T 1486094340 18<Riiume18> paul sztorc was here
  850. T 1486094344 18<Riiume18> also, Maaku7 chimed in
  851. T 1486094348 18<Riiume18> (bitcoin core dev)
  852. T 1486094359 20<banjomanjo>30 awfulcrawler, no, its duplicating tokens for free, that cost money to initially produce
  853. T 1486094369 20<banjomanjo>30 but again, lets move on
  854. T 1486094378 19<Riiume> banjomanjo, awfulcrawler, sorry to interrupt, but yea, you can continue in a private chan if you wish
  855. T 1486094382 18<Riiume18> the point is
  856. T 1486094383 18<awfulcrawler18> it's not dupicating tokens because they aren't being represented as the same thing
  857. T 1486094383 20<banjomanjo>30 someone said earlier they want Segwit as a hardfork
  858. T 1486094386 18<elusive_vxn18> so ka.
  859. T 1486094387 20<banjomanjo>30 why?
  860. T 1486094391 18<Riiume18> we don't want a hardfork, that's the point of our group
  861. T 1486094394 18<Riiume18> sorry
  862. T 1486094397 18<wingman218> Riiume: thats fine, you can choose not to hardfork
  863. T 1486094399 18<Riiume18> we don't want a DISPUTED hard fork
  864. T 1486094404 20<banjomanjo>30 ^^
  865. T 1486094412 18<Riiume18> right, so back to the main thread...
  866. T 1486094418 18<Riiume18> several ideas have been floated
  867. T 1486094424 18<Riiume18> 1) doing nothing (status quo)
  868. T 1486094450 18<Riiume18> 2) Soliciting a commitment from all miners to enter a "process of consensus", and somehow make it binding
  869. T 1486094467 18<Riiume18> (e.g. by placing their money in escrow until they implement their consensus)
  870. T 1486094500 18<Riiume18> 3) segwit + hf to either 2, 4, or 8 MB blocks
  871. T 1486094502 18<elusive_vxn18> > implying miners care about anything other than profit + security
  872. T 1486094537 18<Riiume18> elusive_vxn, we might use incentives and blackmail to compel them to enter the consensus process
  873. T 1486094539 20<banjomanjo>30 yeah they do elusive_vxn 
  874. T 1486094546 20<banjomanjo>30 keeping in line with the desires of users
  875. T 1486094553 20<banjomanjo>30 otherwise they don't make profit
  876. T 1486094573 18<ETH-Contract18> @Riiume how do you know the network can handle 2, 4 o 8MB blocks?
  877. T 1486094586 18<Riiume18> Eth, I don't, that was someone else's idea\
  878. T 1486094605 18<ETH-Contract18> Do you think the miners and the full nodes can know that?
  879. T 1486094612 18<wingman218> Riiume: any more?
  880. T 1486094627 18<awfulcrawler18> both sides agree that the network can handle 4MB blocks
  881. T 1486094629 20<banjomanjo>30 Cornell put the upper bounds of safe blocksize at 4 MB a yearish ago
  882. T 1486094636 20<banjomanjo>30 thats not both sides agreeing awfulcrawler 
  883. T 1486094641 20<banjomanjo>30 thats the result of an academic study
  884. T 1486094641 18<Riiume18> ETH-Contract, jwinterm proposed it, I believe
  885. T 1486094647 18<awfulcrawler18> segwit results in max 4MB blocksize
  886. T 1486094650 19<Riiume> banjomanjo, link to study?
  887. T 1486094654 20<banjomanjo>30 max _potential_ blocksize
  888. T 1486094657 18<awfulcrawler18> so both sides agree
  889. T 1486094686 18<awfulcrawler18> if 4MB were not ok then there would be extra restrictions in segwit
  890. T 1486094686 18<jwinterm18> I believe the study estimated that at 4 MB around 10% of nodes would drop off the network
  891. T 1486094714 20<banjomanjo>30 http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf
  892. T 1486094733 18<Riiume18> So then we need more information, i.e. what are the effects of a 10% loss of nodes on long term market cap prospects for Bitcoin?
  893. T 1486094747 18<Riiume18> I suggest somebody (maybe me?) fund that study
  894. T 1486094748 20<banjomanjo>30 4 MB is dependent on certai conditions awfulcrawler 
  895. T 1486094764 20<banjomanjo>30 to achieve a 4 MB block the entire block would have to be stuffed with just multisig transactions
  896. T 1486094782 18<awfulcrawler18> And both sides agree that this is allowable
  897. T 1486094798 20<banjomanjo>30 block weight alots 1 MB of space for TX data(inputs, outputs, legacy TX), and 3 MB for witness data(segwit signatures)
  898. T 1486094827 20<banjomanjo>30 Riiume, I linked the study: http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf
  899. T 1486094841 18<Riiume18> banjomajo, thx
  900. T 1486094845 18<Riiume18> reading
  901. T 1486094899 23* BobBorker (8f9ff401@gateway/web/freenode/ip.143.159.244.123) has joined
  902. T 1486094953 24* elusive_vxn has quit (Quit: Page closed)
  903. T 1486094953 18<ETH-Contract18> which chain will have a greater market cap: a 1MB chain with high confirmation times or a 2/4/8MB chain with lower confirmation times?
  904. T 1486094965 24* BobBorker has quit (Client Quit)
  905. T 1486094981 18<wingman218> Riiume: so why is the BU style limit not one of the options
  906. T 1486095022 20<banjomanjo>30 it is contentious, completely untested, and unstable
  907. T 1486095040 20<banjomanjo>30 it has absolutely zero defense against sybil attacks as well
  908. T 1486095055 18<Riiume18> wingman2, nobody has suggested it, but I think because they are of the belieft that there is no way to convince BitFury+BTCC+FullNodeOperators to accept it
  909. T 1486095066 18<wingman218> so you will never agree to it?
  910. T 1486095070 20<banjomanjo>30 no
  911. T 1486095082 18<Riiume18> wingman2, I would, but the groups I mentioned would not
  912. T 1486095085 20<banjomanjo>30 it removes any promise of predictably in the growth curve of costs for operaitng my node
  913. T 1486095092 20<banjomanjo>30 I would not accept that under any circumstances at all
  914. T 1486095101 20<banjomanjo>30 unless free quantum computers start raining from the sky
  915. T 1486095143 18<Riiume18> Interesting sidenote, BTC is more resistant to quantum annealing methods than Cryptonight
  916. T 1486095214 19<wingman2> banjomanjo: so you want to be able to run a node above all else?
  917. T 1486095242 20<banjomanjo>30 yes, that is non-negotiable
  918. T 1486095251 18<jwinterm18> I like the idea of a dynamic blocksize, but I think BU is not a good idea
  919. T 1486095257 20<banjomanjo>30 if I am not verifying the blockchain, bitcoin is not trustless(or trustminimized, whatever you want to say)
  920. T 1486095260 18<ETH-Contract18> LAST 1000 BLOCKS === Bitcoin Unlimited blocks: 220 ( 22% ) Bitcoin Classic blocks: 10 ( 1% ) SegWit blocks: 238 ( 23.8% ) ===
  921. T 1486095273 20<banjomanjo>30 I will not support any form of bitcoin that loses that trustless property
  922. T 1486095292 20<banjomanjo>30 that makes it no different than Paypal or a network of Banks
  923. T 1486095323 18<ETH-Contract18> interesting debate good night everyone
  924. T 1486095326 24* ETH-Contract has quit (Quit: Page closed)
  925. T 1486095432 18<wingman218> interesting i want to be able to send a transaction above all else
  926. T 1486095447 20<banjomanjo>30 well, I'm the one who has to pay to store it
  927. T 1486095457 18<wingman218> so do i
  928. T 1486095460 20<banjomanjo>30 I actually pay for your transaction forever
  929. T 1486095462 20<banjomanjo>30 you pay for it once
  930. T 1486095478 18<wingman218> i run a node
  931. T 1486095519 20<banjomanjo>30 and you have clearly just stated doing that is low on your priority list
  932. T 1486095529 20<banjomanjo>30 I came here for trustless, not a new Paypal
  933. T 1486095539 20<banjomanjo>30 as did from what I see most of the people in Bitcoin
  934. T 1486095570 18<awfulcrawler18> you run a node of your own free will you don't get to guilt btc users about it
  935. T 1486095581 18<wingman218> i wanted machine to machine transactions
  936. T 1486095661 20<banjomanjo>30 my running a node is what guarantees the rules of the network are enforced
  937. T 1486095667 20<banjomanjo>30 its not just an optional thing I do because I want to
  938. T 1486095685 20<banjomanjo>30 it is the only fully secure way to use Bitcoin, and it reinforces the lessened security for those who do not run a node
  939. T 1486095694 20<banjomanjo>30 it is not just a "thing I do", so do not trivialize it as such
  940. T 1486095716 18<awfulcrawler18> you aren't a hero so stop portaying yourself as such. Lots of people run nodes dude
  941. T 1486095722 20<banjomanjo>30 Bitcoin doesn't just magically keep its rules in force just cause
  942. T 1486095727 20<banjomanjo>30 it does because of the large number of nodes running
  943. T 1486095743 20<banjomanjo>30 the less nodes running, the more the guarantee of core rules staying the same is lessened
  944. T 1486095752 23* adam3us (~adam3us@unaffiliated/adam3us23) has joined
  945. T 1486095773 20<banjomanjo>30 I didn't portray myself as anything, I stated why I run a node and why it is important for the security of my money
  946. T 1486095780 20<banjomanjo>30 and how it helps reinforce that security for others
  947. T 1486095876 23* pero (~pero@unaffiliated/pero23) has joined
  948. T 1486095905 18<adam3us18> here's a peterotodd concept: bitcoin security depends on censor-resistant bandwidth, not raw bandwidth.
  949. T 1486095922 20<banjomanjo>30 hmm...
  950. T 1486095934 18<wingman218> That's interesting
  951. T 1486096013 20<banjomanjo>30 Yeah, I think thats a very succinct way of pointing out throughput alone is not the main value prospective 
  952. T 1486096129 20<banjomanjo>30 wingman2, you said you wanted machine to machine transactions as if they are not possible?
  953. T 1486096168 19<wingman2> 12:18 < banjomanjo> I came here for trustless, not a new Paypal
  954. T 1486096185 20<banjomanjo>30 yes, but you said that as if machine to machine payments are not possible
  955. T 1486096187 20<banjomanjo>30 they are
  956. T 1486096310 18<wingman218> its what i was interest in bitcoin for
  957. T 1486096330 18<wingman218> *interested
  958. T 1486096335 20<banjomanjo>30 well then it seems the two chief reasons we are here are not mutually exclusive by anymeans
  959. T 1486096351 20<banjomanjo>30 and in fact I would argue machine to machine payments it dependent on bitcoin staying trustless as well
  960. T 1486096372 18<wingman218> not really
  961. T 1486096387 20<banjomanjo>30 well thats good to have established, a tiny step of progress :)
  962. T 1486096447 18<wingman218> also i don't do them
  963. T 1486096509 18<wingman218> The idea of driving down the highway and negotiating with all of the access points for internet is not really possible
  964. T 1486096549 20<banjomanjo>30 O.o
  965. T 1486096552 20<banjomanjo>30 why not?
  966. T 1486096572 20<banjomanjo>30 your phone is pinging back and forth with every WAP in range of it all day
  967. T 1486096662 18<wingman218> I mean it wasn't possible until the idea of lightning networks came out
  968. T 1486096693 20<banjomanjo>30 yes it was, just not as efficiently
  969. T 1486096702 20<banjomanjo>30 Satoshi has capacity for one way payment channels built in since day one
  970. T 1486096722 20<banjomanjo>30 and even discussed how the economics of the system would make it unviable for microtransactions to stay on chain forever
  971. T 1486096748 20<banjomanjo>30 had*
  972. T 1486096839 24* pero (~pero@unaffiliated/pero24) has left ("Leaving")
  973. T 1486096890 19<wingman2> banjomanjo: It would be super inefficient for each car to create a new transaction every hundred meters with a node it has never seen before and may never see again
  974. T 1486096994 18<wingman218> Anyway it's why I was originally excited about bitcoin, but that was dashed pretty early
  975. T 1486097173 18<wingman218> i will never get what i orginally wanted from bitcoin
  976. T 1486097183 20<banjomanjo>30 yes you will :)
  977. T 1486097228 18<wingman218> i might get it from lightning
  978. T 1486097254 18<wingman218> which i support
  979. T 1486097291 18<Riiume18> wingman2, we have our preferences but... remember our task is to achieve consensus, regardless of what we personally would want
  980. T 1486097304 18<wingman218> and i support segwit
  981. T 1486097328 18<Riiume18> Then you must present a strategy by which SegWit can obtain consensus
  982. T 1486097358 18<Riiume18> I want our group to stay focused on that
  983. T 1486097365 18<wingman218> Riiume: we don't acheive consensus
  984. T 1486097384 18<wingman218> bitcoin acheives consensus
  985. T 1486097399 18<Riiume18> wingman2, correct... well let me be more precise...
  986. T 1486097418 18<Riiume18> We want over 95% of the hashpower on the existing protocol to accept the consensus
  987. T 1486097431 18<wingman218> we can't
  988. T 1486097448 18<wingman218> won't happen
  989. T 1486097453 18<Riiume18> i disagree
  990. T 1486097479 18<Riiume18> if I agreed with that, I wouldn't have formed this channel to begin with
  991. T 1486097510 18<wingman218> "12:50 < Riiume> i disagree "and thats why it won't happen
  992. T 1486097524 18<wingman218> "i disagree"
  993. T 1486097623 18<wingman218> you can't have consensus when "i disagree"
  994. T 1486097625 20<banjomanjo>30 guys, I think you are severely underestimating the support Segwit has
  995. T 1486097640 18<jwinterm18> consensus != unanimous consensus
  996. T 1486097643 20<banjomanjo>30 60% of the visible broadcasting nodes are upgraded to enforce it
  997. T 1486097664 20<banjomanjo>30 almost every major business in the space is onboard, either having completed code updates, in process with them, or intending to
  998. T 1486097688 24* fvcxza has quit (Ping timeout: 260 seconds)
  999. T 1486097689 20<banjomanjo>30 do you really think it is rational in the long term for miners to ignore pretty much the vast majority of the entire rest of the ecosystem?
  1000. T 1486097701 20<banjomanjo>30 that eventually will directly conflict with their motive to turn a profit
  1001. T 1486097724 20<banjomanjo>30 the network is not just the miners, and the miners don't make money without everyone else
  1002. T 1486097728 18<wingman218> we have 10 months
  1003. T 1486097734 20* banjomanjo 30shrugs
  1004. T 1486097743 20<banjomanjo>30 softforks have taken longer than this to roll out before
  1005. T 1486097801 18<wingman218> We have 10 months and then everyone has to upgrade to start signaling segwit again
  1006. T 1486097901 18<wingman218> So that we don't run out of flags because of people not upgrading their software
  1007. T 1486097910 18<Riiume18> <Hey all, I'm going afk, please keep the conversation directed towards the stated goal with an eye towards actionable plans.>
  1008. T 1486097923 18* Riiume goes afk
  1009. T 1486097952 18<awfulcrawler18> Would a 2MB hardfork be an acceptable compromise?
  1010. T 1486097998 20<banjomanjo>30 instead of segwit, no
  1011. T 1486098012 20<banjomanjo>30 potentially down the line if there is consensus for it, I wouldn't be opposed in practice
  1012. T 1486098015 18<awfulcrawler18> 2MB + segwit?
  1013. T 1486098018 20<banjomanjo>30 but that is entirely dependent on segwit
  1014. T 1486098022 20<banjomanjo>30 not immediately after, no
  1015. T 1486098025 20<banjomanjo>30 that wouldn't be safe
  1016. T 1486098039 20<banjomanjo>30 I'd want Schnorr signatures + OWAS first
  1017. T 1486098046 20<banjomanjo>30 and again, it would be entirely dependent on consensus
  1018. T 1486098056 20<banjomanjo>30 if everyone is onboard, fine, but if there is contention over it no
  1019. T 1486098080 20<banjomanjo>30 a hardfork without unanimity over a puny 1 MB increase that does nothing in the longrun is an insane risk to me
  1020. T 1486098104 18<wingman218> Actually, about that, core should change the end date for segwit activation to five years in the future.
  1021. T 1486098116 18<awfulcrawler18> 1MB increase either does nothing or is an insane risk...
  1022. T 1486098129 18<awfulcrawler18> if it 'does nothing' how can it be risky?
  1023. T 1486098143 24* NLNico has quit (Ping timeout: 248 seconds)
  1024. T 1486098160 20<banjomanjo>30 because of the risk of consensus failure if its not unanimous
  1025. T 1486098175 20<banjomanjo>30 1 MB increase in the long term is just a bandaid that will not last long at all
  1026. T 1486098197 20<banjomanjo>30 not worth risking a failed hardfork over a bandaid that is in no way even close to a permenant solution
  1027. T 1486098249 18<wingman218> I don't like the idea of 2mb, But I don't have a problem with a failed hard fork
  1028. T 1486098268 24* Riiume (446a9f30@gateway/web/freenode/ip.68.106.159.4824) has left
  1029. T 1486098273 18<awfulcrawler18> consensus isn't 100% of people agreeing to do the same thing
  1030. T 1486098274 23* Riiume (446a9f30@gateway/web/freenode/ip.68.106.159.4823) has joined
  1031. T 1486098276 20<banjomanjo>30 well then I think you are miscalculating the security model of the network enormously wingman2 
  1032. T 1486098284 24* Riiume (446a9f30@gateway/web/freenode/ip.68.106.159.4824) has left
  1033. T 1486098286 18<awfulcrawler18> the ETH / ETC fork was a form of consensus as well
  1034. T 1486098290 20<banjomanjo>30 when it comes to a hardfork, yes it is awfulcrawler 
  1035. T 1486098299 20<banjomanjo>30 ETH/ETC was not a form of consensus
  1036. T 1486098302 20<banjomanjo>30 it was a consensus failure
  1037. T 1486098312 18<awfulcrawler18> no it is an emergent property of the decentralized system
  1038. T 1486098316 20<banjomanjo>30 and it resulted in two separate chains to the economic detriment of both
  1039. T 1486098324 20<banjomanjo>30 no it wasn't, it was a consensus failure
  1040. T 1486098329 18<awfulcrawler18> there is no 'failure', there is just the state of the system
  1041. T 1486098330 20<banjomanjo>30 "emergent property" is a vague buzzword
  1042. T 1486098336 20<banjomanjo>30 that means "anything that happens" essentially
  1043. T 1486098340 18<awfulcrawler18> just because you don't like the outcome doesn't mean it's a failure
  1044. T 1486098340 20<banjomanjo>30 thats nonsense
  1045. T 1486098350 18<awfulcrawler18> 'consensus' as you use it is a vague buzzword
  1046. T 1486098367 20<banjomanjo>30 no it is not, it is the network maintaining enforcement of the same consensus rules
  1047. T 1486098376 20<banjomanjo>30 the network diverging is a failure of consensus
  1048. T 1486098393 18<awfulcrawler18> Network A ----> Network A + network B
  1049. T 1486098397 18<awfulcrawler18> Network A still exists
  1050. T 1486098402 18<awfulcrawler18> diverging isfine
  1051. T 1486098408 20<banjomanjo>30 no it is not
  1052. T 1486098414 20<banjomanjo>30 it resulted in a loss of value for both chains
  1053. T 1486098427 18<wingman218> i don't have a problem with that
  1054. T 1486098428 20<banjomanjo>30 the loss of money on both sides due to the technical shortcomings not modeling diverging forks
  1055. T 1486098430 20<banjomanjo>30 double spends
  1056. T 1486098439 20<banjomanjo>30 well guess what wingman2, almost everyone else does
  1057. T 1486098441 20<banjomanjo>30 this is money
  1058. T 1486098442 18<awfulcrawler18> that's something you personally have a problem with. Code doesn't care
  1059. T 1486098450 20<banjomanjo>30 if you don't care about your money maintaining its value, guess what, you are in the wrong place
  1060. T 1486098473 18<wingman218> no i'm not
  1061. T 1486098480 18<awfulcrawler18> If you want to compromise you have to accept that you'll have less than 100% agreement
  1062. T 1486098482 20<banjomanjo>30 I came here in good faith to discuss
  1063. T 1486098496 20<banjomanjo>30 and so far have been met with nothing but ultimatums for hardforks
  1064. T 1486098505 20<banjomanjo>30 and a complete dismissal of factual aspects of how this system functions
  1065. T 1486098525 20<banjomanjo>30 no I don't awfulcrawler 
  1066. T 1486098530 20<banjomanjo>30 if we don't get agree, status quo it is
  1067. T 1486098545 18<wingman218> nothing but ultimatums for hardforks no hardforks
  1068. T 1486098564 18<awfulcrawler18> yeah you don't want to compromise. Topic is 'compromise to avoid contentious hardfork'
  1069. T 1486098572 18<jwinterm18> no ultimatum, no ultimatum, you're the ultimatum
  1070. T 1486098573 18<awfulcrawler18> good luck
  1071. T 1486098588 20<banjomanjo>30 no where did I say no hardforks
  1072. T 1486098597 20<banjomanjo>30 and I literally *just* laid out the conditions under which I would accept one
  1073. T 1486098623 20<banjomanjo>30 that was unbelievably disingenuous of you to say
  1074. T 1486098656 18<jwinterm18> if every node and 100% of hash power agrees with hardfork?
  1075. T 1486098663 18<jwinterm18> that seems...unlikely
  1076. T 1486098668 20* banjomanjo 30shrugs
  1077. T 1486098673 18<awfulcrawler18> did I say you said 'no hardforks'...(because I didn't)
  1078. T 1486098690 20<banjomanjo>30 kicking people off the network they bought into is a complete betrayal of the social contract of bitcoin in my mind
  1079. T 1486098720 18<awfulcrawler18> forking doesn't kick people off though
  1080. T 1486098736 20<banjomanjo>30 yes it does, again, you act like the value of a token is irrelevant
  1081. T 1486098737 20<banjomanjo>30 its not
  1082. T 1486098754 18<awfulcrawler18> uh...it doesn't. Fork = network A -> network A + network B
  1083. T 1486098758 18<awfulcrawler18> network A still exists
  1084. T 1486098762 18<awfulcrawler18> stay on network A
  1085. T 1486098773 20<banjomanjo>30 you are kicking people off the network in terms of economic network effect
  1086. T 1486098783 20<banjomanjo>30 effectively the same thing
  1087. T 1486098785 18<wingman218> yes
  1088. T 1486098785 18<awfulcrawler18> no people optionally move from A to B
  1089. T 1486098816 20<banjomanjo>30 and again, I've been over this, its not that simple
  1090. T 1486098824 20<banjomanjo>30 you ridiculously oversimplify things
  1091. T 1486098835 20<banjomanjo>30 a diverging fork is counterfeiting a token and its distribution
  1092. T 1486098842 20<banjomanjo>30 those tokens cost electricity and energy to produce
  1093. T 1486098851 20<banjomanjo>30 and they are being replicated for essentially free to speculate on
  1094. T 1486098854 20<banjomanjo>30 i.e. counterfeiting
  1095. T 1486098901 18<awfulcrawler18> I will launch a new blockchain called dollaridoodle and give all btc holders one dollaridoodle for every bitcoin they have. Have I counterfeited anything?
  1096. T 1486098969 20<banjomanjo>30 that is not a rational argument, that is a description of a totally different thing
  1097. T 1486098982 20<banjomanjo>30 its a new token, it is not a diverging fork claiming to be the original network and token
  1098. T 1486098991 20<banjomanjo>30 that is a wild conflation of two different things
  1099. T 1486098994 18<awfulcrawler18> it's a thought experiment which describes the same thing as a PoW fork
  1100. T 1486099000 20<banjomanjo>30 no it doesn't
  1101. T 1486099001 20<banjomanjo>30 at all
  1102. T 1486099009 18<awfulcrawler18> it does...exactly
  1103. T 1486099013 20<banjomanjo>30 no it doesn't 
  1104. T 1486099019 18<awfulcrawler18> this is brick-wall stuff
  1105. T 1486099024 18<awfulcrawler18> I'll shut up now don't worry
  1106. T 1486099026 20<banjomanjo>30 its describes a network that results in the same distribution 
  1107. T 1486099028 20<banjomanjo>30 it is not the same thing
  1108. T 1486099038 20<banjomanjo>30 dude, you are the brick wall here
  1109. T 1486099046 20<banjomanjo>30 I am actually explaining reasoning behind what I say
  1110. T 1486099051 20<banjomanjo>30 you just keep asserting things
  1111. T 1486099058 20<banjomanjo>30 and do not provide any rationalization for it
  1112. T 1486099085 18<wingman218> anyway
  1113. T 1486099106 18<wingman218> 12:58 < Riiume> <Hey all, I'm going afk, please keep the conversation directed towards the stated goal with an eye towards
  1114. T 1486099109 18<wingman218> actionable plans.>
  1115. T 1486099114 18<wingman218> so
  1116. T 1486099217 19<wingman2> you like hard forks or not awfulcrawler and banjomanjo doesn't matter. what can we do to make this go smoothly
  1117. T 1486099238 18<wingman218> core should change the end date for segwit activation to five years in the future.
  1118. T 1486099247 20<banjomanjo>30 thats not how that works
  1119. T 1486099253 18<wingman218> what do you think
  1120. T 1486099255 20<banjomanjo>30 that in itself would require a softfork I believe
  1121. T 1486099340 20<banjomanjo>30 core can't just "change things"
  1122. T 1486099347 20<banjomanjo>30 people have to choose to run their updated software
  1123. T 1486099353 20<banjomanjo>30 they have no control whatsoever beyond writing code
  1124. T 1486099358 20<banjomanjo>30 the rest is entirely up to the users
  1125. T 1486099372 18<wingman218> core should change the end date for the possibility of a segwit activation to five years in the future. not the ~10 months it is now.
  1126. T 1486099380 20<banjomanjo>30 they can't do that!
  1127. T 1486099389 20<banjomanjo>30 1) people would have to update their clients
  1128. T 1486099398 20<banjomanjo>30 2) miners would have to softfork to modify BIP9
  1129. T 1486099405 20<banjomanjo>30 they can't just change shit
  1130. T 1486099447 18<wingman218> core should change the end date for the possibility of a segwit activation to five years in the future *with a new flag*. not the ~10 months it is now.
  1131. T 1486099528 19<wingman2> banjomanjo: https://github.com/bitcoin/bitcoin/blob/812714fd80e96e28cd288c553c83838cecbfc2d9/src/chainparams.cpp#L97
  1132. T 1486099599 18<wingman218> We have to start everything again after November 15th, 2017.
  1133. T 1486099615 18<wingman218> Everyone has to start signaling again.
  1134. T 1486099726 20<banjomanjo>30 this is ridiculous
  1135. T 1486099734 20<banjomanjo>30 you want a "compromise" to avoid a contentious hardfork
  1136. T 1486099750 20<banjomanjo>30 but you want to take the SF solution to things right now, and set it up to hopefully take longer?
  1137. T 1486099805 20<banjomanjo>30 thats effing ridiculous
  1138. T 1486099808 18<wingman218> no That's not what that
  1139. T 1486099831 18<wingman218> It's literally the date segwit dies
  1140. T 1486099834 20<banjomanjo>30 no its not
  1141. T 1486099842 20<banjomanjo>30 its the date the bit has to be reset
  1142. T 1486099852 18<wingman218> well yes
  1143. T 1486099868 18<wingman218> And then it has to be done all over again
  1144. T 1486099897 18<wingman218> everyone has to upgrade again
  1145. T 1486100023 20<banjomanjo>30 so?
  1146. T 1486100032 18<wingman218> Without that end date, the slow march of time will get us segwit
  1147. T 1486100033 20<banjomanjo>30 you're suggestion requires updating again anyway
  1148. T 1486100036 20<banjomanjo>30 wait til then
  1149. T 1486100093 18<wingman218> Just by everyone and eventually having to upgrade
  1150. T 1486100104 18<wingman218> *Just by everyone eventually having to upgrade
  1151. T 1486100220 18<jwinterm18> thread got scrubbed: https://www.reddit.com/r/Bitcoin/comments/5rqvwc/btcnegotiate_on_freenode_a_working_group_to/
  1152. T 1486100231 18<jwinterm18> imagine that
  1153. T 1486100241 18<wingman218> bu will die over time if they start having problems keeping up with security fixes and such
  1154. T 1486100348 18<adam3us18> bitcoin is going through updates anyway, 0.14 is coming in a few months.
  1155. T 1486100448 18<wingman218> i really don't like that end date
  1156. T 1486100811 23* vogelito (~Adium@fixed-190-149-187-190-149-151.iusacell.net23) has joined
  1157. T 1486100899 18<vogelito18> Good night. It would be nice if the logs included the nick, otherwise pretty hard to read
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement