| 00:08:39 | jcorgan: | heh. i'm convinced the concept of "conciousness" as it is presently used (and abused) will evaporate in the same way that "essence of life" evaporated once organic chemistry began to be better understood. |
| 00:10:39 | kanzure: | the wikipedia article literally says "nothing of any consequence has ever been written about it" |
| 00:15:21 | moa: | intelligence is not well-defined either ... the Turing test, imho, is something along the lines of "you'll know it when you see it" |
| 00:15:30 | wumpus: | or, like free will versus determinism just a philosophical distinction that is meaningless in practice - and at most there'd be a brain area that, when stimulated, causes feelings of being conscious |
| 00:16:23 | wumpus: | moa: intelligence is something like 'general problem-solving ability', I think it's fairly well defined, as these things go |
| 00:16:37 | jcorgan: | Daniel Dennet has lots of interesting things to say about all above |
| 00:16:45 | jcorgan: | Dennett* |
| 00:19:45 | wumpus: | I'll look him up |
| 00:24:48 | hulkhogan42o: | are there any dependencies for the sagemath file on bitcoin.ninja? seeing a stacktrace sadly on load |
| 00:53:23 | andytoshi: | hulkhogan42o: shouldn't be, can you 0bin.net the stacktrace? |
| 00:56:38 | gmaxwell: | hulkhogan42o: not at all, but perhaps your sage install isn't complete? |
| 01:32:27 | hulkhogan42o: | andytoshi, gmaxwell: very likely - its the first time i've used sage, so i think i might be missing libraries http://0bin.net/paste/mtJ0bvFoGGEV-2ct#zNHm+cYPIQ5RBjHbxUwdOs00lTMhU7GPUYUBmc9GesM |
| 01:33:14 | hulkhogan42o: | looking at the trace, i think it might be cython related |
| 01:51:11 | hulkhogan42o: | ill work through my distro's package deps to be sure i didnt mess something up, alternatively if there's a quickfix or so ya'll know of i'd happily try it |
| 01:53:41 | andytoshi: | 'fraid not, i don't recognize that |
| 05:05:42 | kanzure: | .title https://news.ycombinator.com/item?id=9418137 |
| 05:05:42 | yoleaux: | Polynomial-Time Hierarchy Is Infinite Under a Random Oracle | Hacker News |
| 06:27:56 | psztorc: | psztorc has left #bitcoin-wizards |
| 07:13:40 | nsh: | i was thinking about this [sort of thing] yesterday wrt to what a system 'knows' vs. what a system is able to demonstrate given appropriate input |
| 07:16:13 | nsh: | .wik Baker-Gill-Solovay oracle |
| 07:16:14 | yoleaux: | "In complexity theory and computability theory, an oracle machine is an abstract machine used to study decision problems. It can be visualized as a Turing machine with a black box, called an oracle, which is able to decide certain decision problems in a single operation. The problem can be of any complexity class." — http://en.wikipedia.org/wiki/Oracle_machine |
| 07:19:32 | nsh: | (e.g. the bitcoin network is facilitated in knowing-in-fact the 'current' UTXO set because it knows-in-principle the validity of any mutation of the UTXO relative a corresponding set of signatures |
| 07:19:33 | nsh: | ) |
| 07:21:57 | nsh: | the ability to know-in-principle -- to have a verificational capacity -- can grow faster than (superadditively wrt) system knowledge-of-fact |
| 07:22:10 | nsh: | (i assert) |
| 07:53:42 | moa: | nsh: "the ability to know-in-principle -- to have a verificational capacity -- can grow faster than (superadditively wrt) system knowledge-of-fact" |
| 07:53:55 | moa: | proof of this would be interesting |
| 07:55:16 | nsh: | * nsh nods |
| 08:05:15 | cameron.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 |
| 08:05:15 | cameron.freenode.net: | Users on #bitcoin-wizards: andy-logbot bedeho2 pgokeeffe Mably justanotheruser p15x adam3us1 HostFat moa hktud0 priidu nuke_ arubi Luke-Jr b_lumenkraft rusty1 ryanxcharles bliljerk101 unlord_ hulkhogan42o kanzure hashtagg jgarzik [7] NewLiberty_ face Meeh p15 Dr-G2 airbreather Taek MoALTz cbeams_ PaulCapestany waxwing antgreen wallet42 OneFixt RoboTeddy sneak sparetire catcow DocHex dc17523be3 Starduster morcos lmatteis mappum CoinMuncher helo isis Adlai LeMiner SDCDev |
| 08:05:15 | cameron.freenode.net: | Users on #bitcoin-wizards: pollux-bts Hunger- Relos spinza heath crescendo c0rw1n Guest50139 Tiraspol nickler Guest55444 Emcy CodeShark prodatalab trstovall dansmith_btc dgenr8 sadoshi GAit melvster mkarrer Kwelstr [d__d] binaryatrocity gnusha luny BananaLotus guruvan lmacken rustyn harrigan Eliel Cory K1773R warren grandmaster xapp Alanius brand0 poggy ajweiss Zouppen Krellan stonecoldpat phedny aakselrod HM s1w lechuga__ EasyAt_ yorick afdudley mm_0 lnovy cdecker wiz |
| 08:05:15 | cameron.freenode.net: | Users on #bitcoin-wizards: leakypat platinuum koshii smooth SubCreative Madars tromp_ throughnothing_ amiller sparetire_ gmaxwell Xzibit17 adams_ forrestv richardus Fistful_of_coins sdaftuar andytoshi Iriez copumpkin nsh cfields_ wumpus jbenet NeatBasisW dasource dardasaba ebfull luigi1111w jonasschnelli merlincorey [ace] eric a5m0 nephyrin null_radix Sqt mikolalysenko sturles GreenIsMyPepper harrow vonzipper berndj manan19 comboy jaromil catlasshrugged_ Apocalyptic |
| 08:05:15 | cameron.freenode.net: | Users on #bitcoin-wizards: cryptowest_ runeks__ Graet indolering Keefe petertodd jcorgan larraboj ryan-c jessepollak gribble tromp mr_burdell d9b4bef9 starsoccer weex SwedFTP yrashk artifexd kumavis otoburb huseby midnightmagic BlueMatt TD-Linux mariorz hguux___ wizkid057 Anduck kyuupichan Logicwax phantomcircuit luigi1111 nanotube yoleaux gavinandresen AdrianG livegnik optimator fluffypony cursive roasbeef_ espes__ pigeons warptangent STRML michagogo sl01 Muis |
| 08:05:15 | cameron.freenode.net: | Users on #bitcoin-wizards: coryfields_ kinlo gwillen epscy Oizopower CryptOprah BrainOverfl0w MRL-Relay azariah @ChanServ davout |
| 08:11:29 | nsh: | i guess it's colloquially roughly the statement that reasoning/processing compresses better than data wedded to interpretation |
| 08:11:49 | nsh: | which is somewhat supported by neuro/cognitive-science |
| 08:12:23 | nsh: | (we retain the ability procedurally recreate memories with relative consistency, rather than 'storing' them) |
| 08:56:35 | wumpus: | nsh: sounds believable to me; I'm not entirely sure about the 'can grow faster' part, but it follows that the ability to perform a task requires less complexity than the ability to understand it and explain it. For organisms, nature just 'runs' all the processes on the bare physical substrate, no need for an intermediate layer of mathematical interpretation. On the other hand, once there is a mathematical representation it is |
| 08:56:35 | wumpus: | possible to manipulate it at an abstract level and consider improvements much faster than 'blind evolution' could. |
| 08:56:53 | nsh: | hmm |
| 08:57:24 | nsh: | * nsh nods |
| 09:14:22 | wumpus: | I do love the decentralized approaches and cycles that are the result of evolution, though slower. Self-sustaining, self-healing, self-propagating systems, eventually relying just on input from the sun. Whereas 'explicit' intelligent approaches require burning up stored energy constantly, and heavily centralized production under special conditions. |
| 09:25:39 | moa: | wumpus: for now, it may just be our current era and stage of technological evolution |
| 09:26:17 | moa: | like dinosaurs came before mammals |
| 09:26:26 | wumpus: | moa: yes |
| 09:27:56 | moa: | just one exogenous shock away from the next leap forward |
| 11:12:00 | wumpus: | gmaxwell: re: Friendship is Optimal, I hadn't expected to enjoy a story about my little pony of all things, but I'm at the third chapter now and I find it quite intruiging, as you said it's really a story about the potential consequences of general AI, maximizing ponies and friendship is just its utility function :-) |
| 11:17:01 | lmatteis: | guys i'm a newb with networking stuff. but who helps establish a connection between two peers that have not opened ports? the only way is to have peers that *do* have opened ports actually help with hole punching between the former two peers, right? |
| 11:17:30 | lmatteis: | how's this done with bitcoin? |
| 11:20:47 | gwillen: | wumpus: it only gets weirder from there. But I enjoyed it a lot. |
| 11:22:06 | gwillen: | lmatteis: as far as I know, bitcoin does not use any kind of NAT holepunching; two nodes behind NATs simply don't connect directly to one another |
| 11:22:21 | gwillen: | (it's possible my information is outdated, since I'm getting it from a forum post from 2012.) |
| 11:22:41 | gwillen: | Instead they see information passed through other nodes. |
| 11:22:48 | lmatteis: | gwillen: so you must incentivize peers to keep their ports opened? |
| 11:23:32 | gwillen: | well, I think there's a healthy supply of peers that are not behind NATs, so their ports are always open |
| 11:23:41 | gwillen: | and I think NATted peers mostly count on connecting to those |
| 11:23:52 | gwillen: | I don't think there's a substantial number of people manually forwarding ports through their routers for bitcoin |
| 11:25:29 | moa: | there's UPNP enabled clients also |
| 11:25:50 | gwillen: | true, if your router supports UPnP your ports may get forwarded for you automatically |
| 11:25:58 | lmatteis: | moa: what's that? is that simply opened ports? |
| 11:26:20 | moa: | if that is the default setting in the router, yes |
| 11:26:42 | lmatteis: | but you still have to configure your bitcoin app to listen on it right? |
| 11:26:54 | moa: | kind of |
| 11:27:15 | moa: | some binaries support it by default others need it enabled, depends |
| 11:27:25 | moa: | on how it was rolled |
| 11:27:37 | lmatteis: | so it can sort of figure out by itself what port to listen on? |
| 11:30:01 | moa: | i think it defaults to 8333 unless you tell it otherwise, it's all configurable, there's a max connections setting also |
| 11:37:17 | moa: | has anyone played around with other non-UPnP hole punching approaches yet btw? |
| 12:22:11 | wumpus: | gwillen: heh they even have a currency called 'bits' |
| 12:56:50 | midnightmagic: | wumpus: Also read: Blindsight by Peter Watts, and the followup Echopraxia. And, if you ever played D&D and remember it fondly, there's Harry Potter and the Natural 20. :) And depending on your ability to wade through more words than I've ever seen in one place "for free", Worm by Wildbow. |
| 12:57:06 | midnightmagic: | heh heh heh.. and now I've infected the rest of you too. |
| 13:11:32 | wumpus: | midnightmagic: I've read blindsight and its successor, Watt's biologist way of writing takes some getting used to, but liked the atmosphere a lot, and also that they have tons of references to state-of-the-art research. Will take a look at those others |
| 13:22:24 | fluffypony: | * fluffypony refuses to read midnightmagic's comment for fear he spend the next week readin |
| 13:49:11 | zooko: | Hello. |
| 13:49:20 | kanzure: | no |
| 13:49:34 | zooko: | My location: Boulder, United States |
| 13:49:41 | zooko: | Oops. |
| 13:49:52 | fluffypony: | lol |
| 13:50:00 | zooko: | Experimenting with phone IRC app. |
| 13:50:08 | zooko: | Can't figure out how to who. |
| 13:51:25 | zooko: | Wow, I just put two and two together that IRC is dating/sex app on phones! It's in that category off apps in the app store. |
| 13:52:45 | fluffypony: | hah hah |
| 13:52:51 | fluffypony: | * fluffypony puts on his robe and his wizard hat |
| 13:53:48 | zooko: | 8-/ |
| 13:56:59 | tromp: | hi zooko |
| 13:57:20 | zooko: | Hi tromp! |
| 13:57:37 | tromp: | any word from the zerocash ppl:-? |
| 13:59:59 | wumpus: | zooko: haha, seriously? I hope freenode is not one of the default networks then :) |
| 14:01:34 | zooko: | Eli said that the Cuckoo paper did not have a compelling argument that there isn't another tmto . |
| 14:02:02 | zooko: | He's reading the Catena paper next. |
| 14:05:03 | zooko: | Hello JoiIto. |
| 14:05:19 | JoiIto: | Hi zooko |
| 14:05:19 | tromp: | sounds like they're committed to hashcash then |
| 14:05:51 | zooko: | What do you mean? |
| 14:06:43 | tromp: | alternatives to hashcash require mre faith in tmto resistence |
| 14:06:47 | tromp: | more |
| 14:08:20 | zooko: | No, I'm going to choose a memory-oriented PoW. |
| 14:08:54 | tromp: | so? that doesn't contradict the use of hashcash |
| 14:09:00 | zooko: | The Catena paper starts off on the right foot by describing The Pebble Game. |
| 14:09:14 | zooko: | Oh, I don't know what you mean by hashcash. |
| 14:09:43 | tromp: | hashcash is only the most widely used PoW:-) |
| 14:10:28 | tromp: | find a nonce where H(header) |
| 14:10:35 | zooko: | Ok, then I don't understand what you mean by memory-oriented PoW not contradicting ... |
| 14:11:00 | tromp: | your H will be memory oriented |
| 14:12:52 | zooko: | Ok, I understand, thanks. :-) |
| 14:13:12 | tromp: | you'll pick symmetric over assymetric PoW |
| 14:13:28 | tromp: | asymmetric |
| 14:18:14 | zooko: | tromp: yes, I think I will. |
| 14:18:55 | zooko: | I don't think we require the feature of fast verification, and the symmetric approach is better-studied. |
| 14:18:56 | pigeons: | is there a good argument presented somewhere for why "memory-hard" PoW is a good thing? I think mainly the claim i see is that botnet majority is more "fair" than asic owner majority |
| 14:20:08 | zooko: | I don't recall for certain, but I think vitalik may have written a blog post about it that I liked. |
| 14:21:01 | instagibbs: | Not sure I find that botnet argument super compelling. We're banking on the fact that ASICs are hard to get/setup/run. |
| 14:21:34 | pigeons: | right, for sure if that's the case its only temporary |
| 14:21:48 | instagibbs: | I mean it's still an interesting question, to be sure. |
| 14:22:52 | tromp: | dram-majority is more fair than asic-majority |
| 14:23:13 | zooko: | Fair is an interesting word. |
| 14:23:30 | zooko: | My children use it a lot. |
| 14:24:41 | zooko: | I sometimes think "fair" just means "good". |
| 14:25:25 | zooko: | The other day I heard a comment on econtalk that when Adam Smith used the word "fair" he meant it in the 18th |
| 14:25:51 | zooko: | Century sense, which was a sports metaphor. "Fair" meant "not foul". |
| 14:26:13 | Eliel: | zooko: I think the word "good" is much more ambiguous than fair. |
| 14:27:49 | zooko: | True. |
| 14:28:23 | zooko: | But, I for one certainly don't know whether asic mining is more or less fair than dram mining. :-) |
| 14:33:08 | instagibbs: | Egalitarian might be a better word. |
| 14:33:10 | Eliel: | but, at least in games, fair tends to mean a system that's not preset to give significant advantage to any specific subgroup of players. |
| 14:34:35 | pigeons: | i think in this context, "fair" is often used to mean more diverse and "decentralized" and "distributed" |
| 14:36:48 | zooko: | Ah, yes that I do have an opinion about. |
| 14:37:20 | zooko: | Eliel: good definition. Thanks. |
| 14:40:45 | Eliel: | usually, though, people experienced in the game aren't considered a subgroup in this respect though. |
| 14:41:22 | zooko: | :-) |
| 14:41:49 | zooko: | Yeah, it's not fair for the pros to always win! |
| 14:42:05 | zooko: | My children say things like that, too... |
| 14:42:40 | tromp: | tic-tac-toe is fair |
| 14:42:41 | zooko: | Brb |
| 14:43:10 | ajweiss: | GOOD EVENING DR. FALKEN |
| 14:43:12 | tromp: | connect-4 is unfair |
| 14:45:56 | zooko`: | zooko` is now known as zooko |
| 14:46:44 | Eliel: | zooko: The game of Go has a standard method of compensating for the difference in the player's experience levels. It helps stronger player learn even when playing against weaker players. |
| 14:47:23 | zooko: | Yeah, that's a good feature. |
| 14:47:34 | Eliel: | That method has also become the unit of measurement for comparing player strength |
| 14:47:51 | zooko: | I still remember that time tromp gave me a 9 stone handicap on a 9x9 board and I still lost. Argh--I hated that. |
| 14:48:00 | Eliel: | :D |
| 14:48:24 | Eliel: | yeah, 9 stones is hardly enough to bridge the gap between a beginner and an expert. |
| 14:48:33 | zooko: | Yeah. That wasn't fair. |
| 14:48:34 | Eliel: | ... even on 9x9 board it appears |
| 14:49:38 | zooko: | Way to go JoiIto on http://gavintech.blogspot.com/2015/04/joining-mit-media-lab-digital-currency.html |
| 15:02:28 | jeremyrubin: | come hang out on #mit-dci if you want to chat about it! |
| 15:03:14 | fluffypony: | ah this is where the IRC dating thing starts... |
| 15:05:20 | zooko: | Hiya adam3us. |
| 15:10:39 | amiller: | wow, neat |
| 15:18:44 | DocHex: | DocHex has left #bitcoin-wizards |
| 15:24:45 | justanot1eruser: | justanot1eruser is now known as justanotheruser |
| 16:24:13 | adam3us: | zooko: hi |
| 19:16:28 | blazes816: | blazes816 is now known as tcrypt |
| 21:11:45 | lmatteis: | what's a good irc channel to talk about issues regarding p2p and network; how to find peers and topics of that kind? |
| 21:12:43 | gmaxwell: | there is a ##p2p-hackers |
| 21:13:53 | lmatteis: | with only 7 ppl |
| 21:14:31 | lmatteis: | gmaxwell: do you know if any experiments have been done on the propagation times of data across the network? |
| 21:14:49 | lmatteis: | and also, if the gossip protocol is truly random and not biased towards specific peers |
| 21:16:04 | gmaxwell: | across what network? |
| 21:16:14 | lmatteis: | bitcoin sorry |
| 21:16:26 | gmaxwell: | there is some data on that that is old and rubbish. |
| 21:16:58 | gmaxwell: | Bitcoin Core has intentional sleeps in the message handling; and all the data that was there was before a number of changes that reduced delays a zillion fold. |
| 21:18:04 | gmaxwell: | It's more or less 'random'; take care with biasing, it's very easy to create vulerabilities to partitioning |
| 21:22:35 | lmatteis: | i wonder if implementing NAT traversal techniques in clients, and therefore increasing the number of peers that can connect to one another, would speed up propagation of data in the bitcoin network |
| 21:22:54 | lmatteis: | it would certainly help with decentralization |
| 21:23:01 | lmatteis: | although not much |
| 21:23:17 | gmaxwell: | Ahem: http://arxiv.org/abs/1502.07373 "The Spy in the Sandbox -- Practical Cache Attacks in Javascript" |
| 21:24:05 | gmaxwell: | lmatteis: perhaps not, if it increases the connectivity around slow ratelimited hosts. |
| 21:25:23 | fluffypony: | maybe it's better to tend towards well-configured hosts that have fat bandwidth |
| 21:25:33 | fluffypony: | than try and make everyone on a 56k dial-up connection play their part |
| 21:25:33 | Taek: | "allows a remote adversary recover information belonging to other processes, other users and even other virtual machines running on the same physical host as the victim web browser" am I misunderstanding or is that a very serious problem? |
| 21:25:58 | fluffypony: | Taek: JavaScript in general is a very serious problem ;) |
| 21:26:19 | gmaxwell: | Taek: this is why all cryptographic software should be hardened against those kinds of sidechannels. |
| 21:26:31 | lmatteis: | this seems from February. Wonder what's Chrome's response was |
| 21:26:53 | moa: | it's some wild stuff so question is how prevalent is it already |
| 21:27:40 | moa: | NoScript ftw |
| 21:27:59 | lmatteis: | fluffypony: well in bitcoin you have fat nodes because they're incentivized to be around (mine). other p2p networks don't have that incentive and i fear that the percentage of unNATted peers is very low |
| 21:28:24 | lmatteis: | also i mean, it's p2p, so the more decentralization, the better |
| 21:28:54 | gmaxwell: | lmatteis: huh? mining doesn't have anything to do with running nodes visible on the network. |
| 21:29:04 | Taek: | you could always make incentives. For example have each peer offer some reasonable reward to the first person who shows them a piece of critical information (such as a block) |
| 21:29:13 | gmaxwell: | (and no competent miner exposes his mining node directly to the network) |
| 21:29:27 | lmatteis: | gmaxwell: don't you have to broadcast it as soon as you find it? |
| 21:29:35 | moa: | lmatteis: there is also the relay node network you may not have heard about |
| 21:29:43 | phantomcircuit: | gmaxwell, plenty of incompetent miners... |
| 21:29:51 | gmaxwell: | Taek: you mean an incentive to sybil attack the network by connecting to everyone in order to race stuff in a microsecond earlier? |
| 21:30:03 | fluffypony: | hmmmmm...do solo miners add "identifiable" data to the coinbase data? |
| 21:30:11 | fluffypony: | (by default I mean) |
| 21:30:20 | gmaxwell: | lmatteis: sure; but _reliable_ fast connectivity is more important than being connected to many people. |
| 21:30:27 | gmaxwell: | fluffypony: course not. |
| 21:30:34 | lmatteis: | what do you mean? |
| 21:30:42 | lmatteis: | why would need fast connectivity |
| 21:30:47 | fluffypony: | thought so |
| 21:30:53 | Taek: | it would incentives nodes with high connectivity and quick propagation of stuff, I don't see where the sybil attack comes in |
| 21:32:38 | moa: | lmatteis: http://sourceforge.net/p/bitcoin/mailman/message/31604935/ |
| 21:33:40 | gmaxwell: | Taek: because sybil attacking the network is what you just described pays people to do. |
| 21:38:18 | Taek: | gmaxwell: hmm. It does create incentives for nodes to make sure that they are the only ones who are ever able to send you a block. But it also incentivizes competition. It's only a problem when the attacker takes control of your peerlist, which other nodes have incentives to prevent |
| 21:39:02 | moa: | lmatteis: some more current info https://bitcointalk.org/index.php?topic=766190.0 |
| 21:40:57 | gmaxwell: | Taek: it incentivzes ~agressively~ wasting everyone elses resources. Other than large miners (who can use efficient tools like relay network) no one else cares about tens of milliseconds in propagation delays. |
| 21:43:27 | zookosphone: | zookosphone is now known as zooko |
| 22:00:05 | lmatteis: | here's a good article on NAT traversing techniques https://hal.inria.fr/hal-00945700/file/paper.pdf |