| 00:40:48 | Taek: | if we ended up colonizing Mars, would Bitcoin be forced to stay on Earth? |
| 00:41:03 | Taek: | 10 minute block time would stop being sufficient I imagine |
| 00:41:17 | Taek: | esp. if we got to the further planets, or some asteroids, etc. |
| 00:47:37 | kanzure: | colonization is not necessary to ask that sort of question (imagine distant non-colonization-reated equipment) |
| 00:50:18 | amiller: | sometimes mars is within 10 minutes away |
| 00:50:20 | amiller: | other times not |
| 00:52:16 | kanzure: | instead of having different currencies for ever-expanding light-spheres it would be nice to have some sort of matrioshka shell thing that was built for the concept of a larger sphere of payments that take (extremely) longer to confirm |
| 00:54:11 | kanzure: | oh right, that doesn't work for the same reasons it didn't work last week and the week before that. |
| 00:55:40 | Taek: | It's interesting to think of how that could be applied on Earth today |
| 00:56:06 | Taek: | Have different spheres for each continent, country, town |
| 00:56:22 | Taek: | each with a substantially quicker block time than the last |
| 00:56:43 | kanzure: | wont work because you can double spend in another "town" level at the same time |
| 00:56:55 | Taek: | "Chicago experiencing a 51% attack all week" |
| 00:57:18 | Taek: | kanzure you'd probably need coins that are only useful in a particular sphere |
| 00:57:46 | Taek: | eg if you are able to spend coins in chicago, then NYC won't recognize them as in a NYC wallet |
| 00:58:10 | Taek: | other problems aside, ofc |
| 01:02:54 | amiller: | i think there's a kind of proof of work you can do if you are within a tiny lightsphere |
| 01:02:57 | amiller: | rather than far apart |
| 01:03:47 | amiller: | if there's a group of parties that has a really large hashpower among them, but the distribution of that hashpower is sufficiently distributed that like there's some 30% of it that's a minimum light distance away from some other 30% |
| 01:04:00 | amiller: | then they can't build sequential proofs of work bigger than that |
| 01:04:27 | amiller: | so if you had optimal fiber optic peering or something you could build a region local proof of work that couldn't be attacked even by a more powerful miner as long as the more powerful miner is far away |
| 01:05:36 | gmaxwell: | amiller: the lightsphere limited one doesn't work for offline parties. So peopel can't easily come and leave from the system. |
| 01:07:32 | amiller: | i think it does |
| 01:07:58 | amiller: | it's like the difference between a sequentially-hard proof of work and one like bitcoin. |
| 01:08:09 | kanzure: | is there an equivalent localized version of the problem of speed of light constraints? like, nodes aready don't get information instantly anyway, as they would due to light delay.. |
| 01:08:18 | amiller: | you can do an offline-checkable proof of "concentrated work" |
| 01:08:50 | gmaxwell: | amiller: it doesn't. say there are a bunch of parties mining in your lightsphere but you're not paying attention. Then later people from mars do ton of mining and send a time delayed alternative history. Then an evil group of local martians continue that one. Now you join the scene, how can you distinguish these two histories? one which was forged far away? |
| 01:11:57 | kanzure: | *related (not reated) |
| 01:13:28 | amiller: | hrm. maybe you can interact *now* with the local chain and quickly figure out that's one's more current than the other |
| 01:13:57 | gmaxwell: | amiller: sure but they're both being run locally now "local martians". But the far away one has lots more work (martian computers) before you joined. |
| 01:14:11 | amiller: | hmm |
| 01:14:12 | gmaxwell: | So it looks to you like the real chain is the imposter: less work, and both are local. |
| 01:14:27 | amiller: | so the martians did a lot of work when they were on mars, but now they've landed and they're your neighbors and you can'ttell which one was there the whole time |
| 01:14:38 | amiller: | that's a neat problem good point |
| 01:16:30 | kanzure: | 18:14 < fenn> it would take a long time to agree on a price for anything |
| 01:16:34 | gmaxwell: | I think it's related to some of the POS simulation problems. |
| 01:16:36 | kanzure: | 18:16 < fenn> no shipping, this is an information-only product |
| 01:17:49 | kanzure: | so trade might be off limits, but maybe there would be some proof system for "locking" funds in one region, but making them available in another? |
| 01:18:49 | kanzure: | some of this has to be constrained by the nature of interstellar existence anyway, trade isn't one-to-one with current concepts i imagine |
| 01:19:58 | kanzure: | if you pay someone a year away to give you data (if you could), it would take a year (at minimum) for an alert of the payment to reach them, and then some amount of time for data to get back to you (unless it was local and encrypted already?) |
| 01:21:04 | kanzure: | 18:19 < fenn> you could send it to a local escrow agent who tell his partner on alpha centauri to forward alphacoins to your seller |
| 01:21:08 | kanzure: | 18:20 < fenn> problem is alpha escrow services gets screwed when the exchange rate changes |
| 01:24:47 | kanzure: | what does orion's arm have to say about this |
| 01:27:19 | kanzure: | :( they took the lazy path http://www.orionsarm.com/eg-article/537285c1964e8 |
| 03:12:45 | Eliel: | synereo seems like an interesting idea. That is, combining a social network with a monetary system. They appear to intend to use a miner based system, however, I started wondering if it might be usable to piggyback timestamping to people's status updates. The system naturally has social (trust) network links included. |
| 08:05:14 | wilhelm.freenode.net: | topic is: This channel is not about short-term Bitcoin development | http://bitcoin.ninja/ | This channel is logged. | For logs and more information, visit http://bitcoin.ninja |
| 08:05:14 | wilhelm.freenode.net: | Users on #bitcoin-wizards: andy-logbot jtimon NewLiberty Logicwax Dr-G2 rdponticelli damethos vmatekole midnightmagic hollandais copumpkin fanquake p15 [7] atgreen artilectinc roconnor emsid Anduck nuke1989 koshii gnusha_ espes__ Sangheili forrestv epscy CryptOprah tromp_ pi07r stonecoldpat asoltys NikolaiToryzin Graet Nightwolf tacotime justanotheruser iddo mortale Guest21768 waxwing ebfull Adlai d4de^^ c0rw1n DoctorBTC Graftec Luke-Jr zenojis devrandom spinza |
| 08:05:14 | wilhelm.freenode.net: | Users on #bitcoin-wizards: Max_H3adr00m EasyAt wizkid057 gavinandresen eristisk gribble go1111111 MRL-Relay shesek rfreeman_w wiretapped altoz HaltingState Aquent Emcy dansmith_btc2 super3 nanotube Starduster burcin SDCDev Meeh LarsLarsen samson_ optimator_ wumpus Grishnakh mmozeiko jgarzik grubles berndj [d__d] livegnik reick nsh Dyaheon Fistful_of_coins SomeoneWeird Krellan Iriez [Derek] comboy arowser warren coryfields_ [\\\] pigeons sipa gmaxwell Taek kanzure _2539 |
| 08:05:14 | wilhelm.freenode.net: | Users on #bitcoin-wizards: LaptopZZ_ lnovy dgenr8 hguux Alanius K1773R phantomcircuit kumavis heath a5m0 andytoshi bobke petertodd BigBitz BrainOverfl0w digitalmagus CodeShark smooth BlueMatt @gwillen sl01 artifexd michagogo tromp bbrittain weex Kretchfoop Muis Hunger-- jrayhawk_ firepacket yoleaux mappum jbenet zibbo_ Eliel amiller crescendo btc_ kgk coryfields poggy UukGoblin danneu catcow TD-Linux [Tristan] helo otoburb ryan-c roasbeef pajarillo Keefe Gnosis ahmed_ |
| 08:05:14 | wilhelm.freenode.net: | Users on #bitcoin-wizards: so starsoccer kinlo Apocalyptic mr_burdell fluffypony @ChanServ phedny lechuga_ abc56889 throughnothing harrow |
| 12:35:36 | Taek: | amiller, I could be mis-thinking, but the difficulty raising attack will still take 10^400 tries to succeed on Bitcoin. |
| 12:35:54 | Taek: | It's just that if you care about reverting to any point instead of to a specific point, each try takes the same amount of time |
| 12:36:22 | Taek: | where if you are trying to revert to a specific point, each try takes exponentially more time than the previous |
| 12:38:37 | Taek: | Though I'm realizing I did the probability wrong. It's probably not as severe as 10^400 tries |
| 12:41:33 | Taek: | The DRA is more interesting in the specific point case |
| 12:42:09 | Taek: | We already know that an attacker with 'p' mining power can do a double spend of length 6 with a probability of 1 if they try enough times |
| 12:42:29 | Taek: | Each time the attacker tries, they have p^6 chance of success |
| 12:42:40 | Taek: | even without using a DRA |
| 12:43:11 | Taek: | But if the attacker keeps trying to revert to the same point, the probability falls off exponentially as they fall behind |
| 12:43:38 | Taek: | with the DRA though, they can keep a constant probability of reverting to the same point, regardless of how far they fall behind |
| 12:45:49 | Taek: | The tradeoff being exponenetial time investment |
| 15:23:17 | SomeoneWeird: | SomeoneWeird is now known as Guest12848 |
| 15:52:30 | fanquake_: | fanquake_ is now known as fanquake |
| 17:49:36 | fanquake_: | fanquake_ is now known as fanquake |
| 18:34:36 | davidlatapie: | davidlatapie has left #bitcoin-wizards |
| 20:04:43 | NewLiberty_: | NewLiberty_ has left #bitcoin-wizards |
| 22:02:47 | Guest12848: | Guest12848 is now known as SomeoneWeird |
| 22:12:56 | davidlatapie: | davidlatapie has left #bitcoin-wizards |
| 22:46:54 | samson2: | samson2 is now known as samson_ |