#bitcoin-wizards

/

      • dnaleor has quit
      • yoleaux has quit
      • Belkaar has quit
      • Belkaar joined the channel
      • Belkaar has quit
      • Belkaar joined the channel
      • darpvader has quit
      • rmwb has quit
      • rmwb joined the channel
      • rmwb has quit
      • rmwb joined the channel
      • Chris_Stewart_5 has quit
      • Chris_Stewart_5 joined the channel
      • luke-jr joined the channel
      • dppsan joined the channel
      • Dizzle joined the channel
      • rmwb has quit
      • dppsan has quit
      • legogris has quit
      • legogris joined the channel
      • rmwb joined the channel
      • rmwb has quit
      • rmwb joined the channel
      • yoleaux joined the channel
      • TheSeven joined the channel
      • marcoagner joined the channel
      • Dizzle has quit
      • _whitelogger joined the channel
      • MaxSan joined the channel
      • rmwb has quit
      • rmwb joined the channel
      • dabura667 joined the channel
      • danydan has quit
      • Giszmo joined the channel
      • dabura667 joined the channel
      • rmwb has quit
      • rmwb joined the channel
      • CheckDavid joined the channel
      • rmwb has quit
      • priidu joined the channel
      • rmwb joined the channel
      • priidu has quit
      • rmwb has quit
      • rmwb joined the channel
      • jonasschnelli joined the channel
      • BashCo joined the channel
      • dnaleor joined the channel
      • BashCo joined the channel
      • Ylbam joined the channel
      • yoleaux has quit
      • yoleaux joined the channel
      • rmwb has quit
      • rmwb joined the channel
      • rmwb has quit
      • cdecker joined the channel
      • daszorz joined the channel
      • JackH joined the channel
      • rmwb joined the channel
      • rmwb has quit
      • rmwb joined the channel
      • cluckj has quit
      • cluckj joined the channel
      • AaronvanW joined the channel
      • Aaronvan_ joined the channel
      • rmwb has quit
      • Alina-malina joined the channel
      • dnaleor has quit
      • Alina-malina joined the channel
      • rmwb joined the channel
      • pro joined the channel
      • rmwb has quit
      • jannes joined the channel
      • rmwb joined the channel
      • MaxSan joined the channel
      • JackH joined the channel
      • kanzure
      • nona_ joined the channel
      • rmwb has quit
      • MaxSan joined the channel
      • nona_ has quit
      • fluffypony
        lol peacoq
      • kanzure
        "Proof of work without all the work" https://arxiv.org/pdf/1708.01285.pdf
      • hmm doesn't cite andytoshi's document.
      • JackH
        where is his paper?
      • got a link?
      • tromp
        also discussed on Hacker News at https://news.ycombinator.com/item?id=14945860
      • kanzure
        JackH: hm?
      • tromp: thank you.
      • JackH
        thx kanzure
      • do we btw have any material on low-level/op_code, that is not necessarily Script/Bitcoin, but general low level op_code's and different mechanisms on low-level interactions?
      • Its a portion I am looking into quite a lot, but it is difficult to find precisely something that would mimic the architecture of Bitcoin
      • even if its non decentralized
      • kanzure
        how about any interpreted small language ever?
      • JackH
        yeah I looked a lot over assembly and microcode as well
      • kanzure
        JackH: the topic you are seeking is just "language design". it's a big subtopic in computer science.
      • or interpreter design i guess.
      • i don't know what you want.
      • tromp
        i have an interpreter for the smallest language on my shirt:-)
      • Chris_Stewart_5 joined the channel
      • JackH
        I guess I am looking for examples of designs of methods via op_codes in languages. The interest really came after seeing MAST conceived
      • afk11 joined the channel
      • belcher has quit
      • laurentmt joined the channel
      • if a softfork locks in, and it is chain specific lock in as sipa said earlier, then for what reason dont we enable a lock-in mechanism on node level after X blocks from activation? Wont this guarantee soft forks locked in, stay locked in? What is the downside?
      • belcher joined the channel
      • gmaxwell
        JackH: you are a new node, syncing, far back early in the chain, someone has forked it, activated the softfork early in it.. zot. you are now fried.
      • you can't reorg onto the real chain because it's invalid.
      • JackH
        right now the "real chain" is via DNS discovery to other nodes, no?
      • kanzure
        it's based on rule validation
      • Guyver2 joined the channel
      • gmaxwell
        JackH: the real chain is the valid chain with the most work.
      • but with your suggestion a node would no longer converge to that if it saw a fork with an early activation first.
      • JackH
        if a bip9 softfork locks in, and we wait 100 blocks (for example) we can be quite certain that not only will it stay activated, but that it will also be the longest chain. if miners reorg 100 blocks back, it will be bad for them as the nodes would kick them off
      • 100 is just an example, could be 250
      • gmaxwell
        no.
      • when you start up, I might give you a 100 block fork off at block 1 (for example). You don't know any better yet, so you follow it. Later you hear about a better chain from other peers, and you switch to that.
      • with your suggestion, if my fork activated the softfork earlier than the best chain did, you wouldn't reorg anymore.
      • kristofferR joined the channel
      • JackH
        ok I get it now, but this also implies that we never know the "longest" chain, by amount. At some point we must assume miners are not going to be able to reorg, for example 5000 blocks after activation
      • kanzure
        length is insufficient, the rule is validity too.
      • rmwb joined the channel
      • gmaxwell
        JackH: no, because I can fork at 0 make 5000 blocks in a few minutes with a miner. Bitcoin never knows it's on the longest chain but it's always moving towards it.
      • someone could give you some old fork, but you won't follow it for long...
      • betawaffle
      • kanzure
        betawaffle: read your scrollback....
      • betawaffle
        lol, sorry
      • laurentmt joined the channel
      • darpvader joined the channel
      • gmaxwell
        I think the title of the paper is misleading.
      • it's not an alternative proof of work scheme, it's a scheme for registering ephemerial identities for sybil resistance.
      • tuxcanfly
        gmaxwell: I've always wondered, wouldn't sybil resistance make proof of work superflous? just distributed mining randomly across identified parties
      • *distribute mining rewards
      • gmaxwell
        if you can strongly assume 2/3rds are honest, and people don't join and leave.
      • in that paper their notion of sybil resistance is very limited. :P
      • contrapumpkin joined the channel
      • The thing to compare this work to is "proof of idle".
      • betawaffle
      • tuxcanfly
        gotta read the paper. proof of idle seems interesting.
      • gmaxwell
        the idea of the paper appears to be everyone generates a public key. Then node join the network by doing a large amount of POW based on the network state. They show people the work and get added to a list of nodes.
      • then at random times, the system (which consists of members registered via the prior process) demands everyone do POW and return responses ASAP. Anyone who doesn't return 'in time' is banned from the system and must make a new identity to rejoin.
      • so this lets nodes spend most of their time not working, but its inherently online... it doesn't build up a hard to forge history.
      • kanzure
        i would like to see more thought put into online/interactive protocols, in general. or some work towards showing limits to what more interaction can enable. (for example: getting all UTXO owners to sign off on state updates)
      • tuxcanfly
        gmaxwell: wouldn't the system members game it because they know when the POW demand goes out, but others don't
      • I guess unless you trust all system members
      • gmaxwell
        tuxcanfly: unclear to me, I don't understand how they make their distributed beacon. I just read it very quickly (under 2 minutes) just to determine what the heck general area they were working on. :)
      • tuxcanfly
        gmaxwell: did you have a different concept for proof of idle in your mind?
      • gmaxwell
        I don't recall how adiabat's scheme beaconed either. :)