Logs: freenode/#haskell
| 2021-05-12 01:56:58 | <Cale> | m_shiraeeshi: yep |
| 2021-05-12 01:57:20 | <m_shiraeeshi> | it reminds me of cap-theorem |
| 2021-05-12 01:57:26 | <shanemikel> | permanent divergence is called a "fork" |
| 2021-05-12 01:57:27 | <Axman6> | in DAML you have to be explicit about who can see any contract |
| 2021-05-12 01:57:54 | <Cale> | m_shiraeeshi: Well-behaved miners tend to discard work that they had done to mine a block once they see that a new one got published. |
| 2021-05-12 01:58:02 | → | hyiltiz joins (~quassel@31.220.5.250) |
| 2021-05-12 01:58:02 | × | hyiltiz quits (~quassel@31.220.5.250) (Changing host) |
| 2021-05-12 01:58:02 | → | hyiltiz joins (~quassel@unaffiliated/hyiltiz) |
| 2021-05-12 01:58:21 | <m_shiraeeshi> | why though? you could make it append-only |
| 2021-05-12 01:58:37 | <Cale> | But it's not strictly necessary -- if the block they just mined was harder to mine, perhaps they could publish just after and have theirs accepted instead. |
| 2021-05-12 01:58:39 | <m_shiraeeshi> | so that newer data overwrite older |
| 2021-05-12 01:58:50 | <m_shiraeeshi> | but that could lead to problems I guess |
| 2021-05-12 01:58:58 | <shanemikel> | m_shiraeeshi who is the authority on time or what is the "newest" block? |
| 2021-05-12 01:58:58 | <Cale> | Because it's a wacky distributed database where everyone is considered an adversary |
| 2021-05-12 01:59:16 | <shanemikel> | that's the main issue with distributed systems is no oracle (time is pretty important to agree on) |
| 2021-05-12 01:59:45 | <Cale> | and so you (a) do far more work than any reasonable person would expect, and (b) have to deal with disagreements |
| 2021-05-12 02:00:11 | × | tromp quits (~tromp@dhcp-077-249-230-040.chello.nl) (Ping timeout: 260 seconds) |
| 2021-05-12 02:00:31 | <shanemikel> | m_shiraeeshi what do you wait until every miner in the world has accepted a block before continuing the next TX? How do you know they were all online? |
| 2021-05-12 02:01:09 | → | tdammers joins (~tdammers@77.109.72.213.res.static.edpnet.net) |
| 2021-05-12 02:01:09 | × | tdammers quits (~tdammers@77.109.72.213.res.static.edpnet.net) (Changing host) |
| 2021-05-12 02:01:09 | → | tdammers joins (~tdammers@unaffiliated/tdammers) |
| 2021-05-12 02:01:17 | <Cale> | and there's also all kinds of fun problems at the peer-to-peer layer of how new connections to the network get handled and messages are routed around the network. Bitcoin is pretty stable since most of its routing is ad-hoc, but I have a sneaking suspicion that most networks are more exploitable at the P2P layer than we've yet had to contend with. |
| 2021-05-12 02:01:23 | × | olligobber quits (olligobber@gateway/vpn/privateinternetaccess/olligobber) (Ping timeout: 265 seconds) |
| 2021-05-12 02:01:27 | → | ddellacosta joins (~ddellacos@86.106.143.228) |
| 2021-05-12 02:01:47 | <shanemikel> | Axman6 I bet there are all kinds of hidden goodies on the BTC chain |
| 2021-05-12 02:01:50 | <m_shiraeeshi> | what is the criterion for the transaction being considered committed then? like a minimal number of acknowledgements or something |
| 2021-05-12 02:02:05 | <shanemikel> | It is all probabilistic based on number of confirmations |
| 2021-05-12 02:02:10 | <Cale> | m_shiraeeshi: Sometimes people will wait for a number of blocks to be piled atop it, yeah |
| 2021-05-12 02:02:22 | <shanemikel> | you never know that a block was accepted. only pretty sure after some number of them |
| 2021-05-12 02:03:25 | × | xff0x quits (~xff0x@2001:1a81:5237:1c00:bdb7:e463:e276:44ae) (Ping timeout: 276 seconds) |
| 2021-05-12 02:03:35 | <Axman6> | goodies and very much not goodies |
| 2021-05-12 02:03:35 | <Cale> | It becomes harder and harder to change things farther into the past, because you have to do the work to form all the blocks on top of whatever it was you wanted changed. |
| 2021-05-12 02:04:18 | <Cale> | (well, blocks representing an equivalent amount of work) |
| 2021-05-12 02:04:25 | <shanemikel> | It's all based on game theory AFAIK... miners are incentivized to produce the next block and make $$ they don't want to do extra work for the same reward |
| 2021-05-12 02:04:26 | <m_shiraeeshi> | in a way it resembles how git works, right? |
| 2021-05-12 02:04:34 | → | xff0x joins (~xff0x@2001:1a81:523e:f300:70f:5849:6a09:4e61) |
| 2021-05-12 02:04:37 | <m_shiraeeshi> | git hashes things |
| 2021-05-12 02:04:37 | <shanemikel> | yeah |
| 2021-05-12 02:04:41 | × | xcmw quits (~textual@cpe-69-133-55-43.cinci.res.rr.com) (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 2021-05-12 02:04:59 | <shanemikel> | git is based on a sort-of merkle dag right? |
| 2021-05-12 02:05:07 | × | oxide quits (~lambda@unaffiliated/mclaren) (Ping timeout: 252 seconds) |
| 2021-05-12 02:05:30 | × | isovector quits (~isovector@172.103.216.166) (Ping timeout: 258 seconds) |
| 2021-05-12 02:05:41 | → | davve_ joins (davve@bsd.douchedata.com) |
| 2021-05-12 02:06:05 | × | davve quits (davve@bsd.douchedata.com) (Read error: Connection reset by peer) |
| 2021-05-12 02:06:26 | × | lambdabot quits (~lambdabot@silicon.int-e.eu) (Ping timeout: 240 seconds) |
| 2021-05-12 02:06:50 | <Cale> | But yeah, all of this kind of thing is unbelievably computationally expensive and power hungry. There are ways to improve the situation a bit vs. proof-of-work, but they're never really going to come close to postgres performance-wise. |
| 2021-05-12 02:06:52 | × | `slikts quits (~nelabs@wikipedia/reinis) (Quit: No Ping reply in 180 seconds.) |
| 2021-05-12 02:06:55 | → | chris__ joins (~chris@81.96.113.213) |
| 2021-05-12 02:07:58 | → | `slikts joins (~nelabs@2a00:d880:5:395::37ca) |
| 2021-05-12 02:07:58 | × | `slikts quits (~nelabs@2a00:d880:5:395::37ca) (Changing host) |
| 2021-05-12 02:07:58 | → | `slikts joins (~nelabs@wikipedia/reinis) |
| 2021-05-12 02:08:07 | → | sdrodge joins (~sdrodge@unaffiliated/sdrodge) |
| 2021-05-12 02:08:16 | → | lambdabot joins (~lambdabot@silicon.int-e.eu) |
| 2021-05-12 02:08:17 | × | lambdabot quits (~lambdabot@silicon.int-e.eu) (Changing host) |
| 2021-05-12 02:08:17 | → | lambdabot joins (~lambdabot@haskell/bot/lambdabot) |
| 2021-05-12 02:08:31 | <Cale> | There's stuff like Tezos' proof-of-stake, where people are granted power over the network in accordance with how much monopoly money they have (which was initially determined by how much real world money they gave the Tezos devs) |
| 2021-05-12 02:09:45 | <m_shiraeeshi> | what about ethereum? it runs arbitrary computations in blockchain somehow? |
| 2021-05-12 02:09:49 | <Cale> | (The idea being that if you want your monopoly money to continue being regarded as valuable, you had better act in good faith) |
| 2021-05-12 02:10:03 | <shanemikel> | Theory on the street is that PoW is less vulnerable to money-printing power grab schemes.. |
| 2021-05-12 02:10:06 | × | apache8080 quits (~rishi@wsip-70-168-153-252.oc.oc.cox.net) (Ping timeout: 240 seconds) |
| 2021-05-12 02:11:07 | <Cale> | Ethereum started out proof-of-work, and I think there's a plan in the works to change it over to proof-of-stake. |
| 2021-05-12 02:11:27 | → | justan0theruser joins (~justanoth@unaffiliated/justanotheruser) |
| 2021-05-12 02:12:01 | <monochrom> | I want to proof my cake and eat it too. |
| 2021-05-12 02:12:12 | <m_shiraeeshi> | some crypto jargon I see here |
| 2021-05-12 02:12:17 | <shanemikel> | It seems like a conflict of interest could emerge between network operators and stakeholders. How do they incentivize network operation? An indirect incentive for stakeholders to maintain network integrety? |
| 2021-05-12 02:12:20 | <Axman6> | bake it with ASICs |
| 2021-05-12 02:12:27 | × | justanotheruser quits (~justanoth@unaffiliated/justanotheruser) (Ping timeout: 250 seconds) |
| 2021-05-12 02:12:27 | <Cale> | Proof of stake has the rather enormous advantage of not wasting everyone's graphics cards (and potentially custom hardware) on solving otherwise-meaningless cryptographic puzzles. |
| 2021-05-12 02:13:27 | <Cale> | Still, you have every node on the network duplicating and checking all the computation that happens on it |
| 2021-05-12 02:13:33 | × | urodna quits (~urodna@unaffiliated/urodna) (Quit: urodna) |
| 2021-05-12 02:14:08 | <Cale> | So it's still pretty expensive and wasteful power-wise to do everything, relative to traditional computer programs. |
| 2021-05-12 02:14:18 | <shanemikel> | Also, if BTC was proof of stake US gov would already be one of the top 1% network controllers (AFAIK because of silk road seizure) |
| 2021-05-12 02:14:29 | <Cale> | Nice |
| 2021-05-12 02:14:43 | → | ddellac__ joins (~ddellacos@ool-44c73afa.dyn.optonline.net) |
| 2021-05-12 02:15:47 | <Cale> | So we just need to convince the BTC folks to go proof of stake, and then convince the US government to shut it down, and then we can rest for a bit before figuring out how to EOL the rest of cryptocurrency. |
| 2021-05-12 02:15:58 | <shanemikel> | LOL |
| 2021-05-12 02:16:03 | × | davve_ quits (davve@bsd.douchedata.com) (Read error: Connection reset by peer) |
| 2021-05-12 02:16:10 | → | davve joins (davve@bsd.douchedata.com) |
| 2021-05-12 02:16:22 | <shanemikel> | *sorry EOL |
| 2021-05-12 02:16:34 | <Cale> | (end of life) |
| 2021-05-12 02:16:58 | <shanemikel> | ahaha |
| 2021-05-12 02:17:31 | <Cale> | I just want it to be over |
| 2021-05-12 02:17:34 | <Cale> | haha |
| 2021-05-12 02:17:56 | <nshepperd> | proof of work blockchain solves a very specific problem, which is byzantine consensus in the absence of God. if you're an organisation providing a service your don't need it since you have a god |
| 2021-05-12 02:19:42 | <shanemikel> | yeah most interesting to me is geopolitics of it. India trying to shut it down, china w/ huge mining power. Texas trying to become mining hotspot. at least it will be interesting. |
| 2021-05-12 02:20:40 | <m_shiraeeshi> | I heard that at some point all the coins are going to be mined |
| 2021-05-12 02:20:43 | × | ddellac__ quits (~ddellacos@ool-44c73afa.dyn.optonline.net) (Ping timeout: 265 seconds) |
| 2021-05-12 02:20:52 | <shanemikel> | especially the timing of it during global power competition and the breakdown of postwar consensus |
| 2021-05-12 02:21:39 | <m_shiraeeshi> | what limits coins from being mined indefinitely? |
| 2021-05-12 02:22:14 | <shanemikel> | I think the rewards for new blocks diminish over time on BTC ad infinitum? |
| 2021-05-12 02:23:32 | <m_shiraeeshi> | also I got the impression that it's becoming increasingly harder to mine coins |
| 2021-05-12 02:23:49 | <shanemikel> | Which will be the case as long as the major BTC fork is controlled by devs, clients & miners who want to adhere to Satoshi's original vision? |
| 2021-05-12 02:24:03 | <m_shiraeeshi> | ok, so what happens after that? people start again from the clean slate? |
| 2021-05-12 02:24:13 | <m_shiraeeshi> | erase all the history? |
| 2021-05-12 02:24:15 | <shanemikel> | The dev communities are another interesting locus of control. |
| 2021-05-12 02:24:36 | <shanemikel> | No, you can fork the chain and change the protocol |
| 2021-05-12 02:26:46 | <m_shiraeeshi> | so is there some kind of a garbage collection in blockchain? |
| 2021-05-12 02:26:49 | <shanemikel> | Get super active on GitHub, join crypto community, talk some Capos (power devs) into adopting your new community guidelines, sneak some malevolent stuff in the backdoor, promote people who are secretly allied with you to hostile takeover of the crypto, profit. |
| 2021-05-12 02:26:59 | → | FinnElija joins (~finn_elij@gateway/tor-sasl/finnelija/x-67402716) |
| 2021-05-12 02:26:59 | finn_elija | is now known as Guest29940 |
All times are in UTC.