--- Log opened Sat Jan 31 00:00:40 2015 | ||
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 264 seconds] | 00:06 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 00:07 | |
-!- ebowden [~ebowden@CPE-144-131-35-125.lns4.dav.bigpond.net.au] has joined ##hplusroadmap | 00:10 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Max SendQ exceeded] | 00:10 | |
-!- phm4242 [~o@boole.london.hackspace.org.uk] has joined ##hplusroadmap | 00:18 | |
-!- voidfire [~voidfire@shellie.darchoods.net] has quit [Ping timeout: 245 seconds] | 00:58 | |
-!- voidfire [~voidfire@shellie.darchoods.net] has joined ##hplusroadmap | 00:59 | |
JayDugger | Any activity using a blockchain for inventory control? | 01:11 |
---|---|---|
JayDugger | I'd ask in a BTC channel, but I don't want to wade through the noise of speculators and wishful groupthinking. | 01:11 |
ebowden | paperbot: http://www.nature.com/ncomms/2013/130702/ncomms3114/full/ncomms3114.html | 01:12 |
justanotheruser | JayDugger: does inventory control require trustless consensus? | 01:39 |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 01:51 | |
-!- lichen [~lichen@c-50-139-11-6.hsd1.or.comcast.net] has joined ##hplusroadmap | 02:05 | |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 02:19 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Quit: Reconnecting] | 02:26 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 02:26 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Quit: Reconnecting] | 02:45 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 02:46 | |
-!- ebowden [~ebowden@CPE-144-131-35-125.lns4.dav.bigpond.net.au] has quit [Remote host closed the connection] | 03:18 | |
-!- delinquentme [~delinquen@162-245-22-166.v250d.PUBLIC.monkeybrains.net] has quit [Ping timeout: 272 seconds] | 03:20 | |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] | 03:36 | |
-!- ebowden [~ebowden@CPE-144-131-35-125.lns4.dav.bigpond.net.au] has joined ##hplusroadmap | 03:41 | |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 03:43 | |
JayDugger | justanotheruser: Yes. | 04:02 |
JayDugger | justanotheruser: Rather, I imagine it can benefit from it. | 04:03 |
justanotheruser | JayDugger: how? | 04:04 |
justanotheruser | inventory? | 04:04 |
justanotheruser | like a companies inventory? Where a group of people count what they have and you believe them? | 04:04 |
JayDugger | Again, yes. | 04:04 |
JayDugger | In particular, I imagine stock and supply for internal uses, rather than inventory for sale. | 04:05 |
justanotheruser | JayDugger: what benefit does a blockchain provide over the people counting signing a message including the counts? | 04:06 |
justanotheruser | that question applies to internal uses as well | 04:10 |
justanotheruser | why can't they workers just commit to a log of all inventory changes the company manages | 04:10 |
JayDugger | Wow...you alternate good questions with bad ones. | 04:11 |
JayDugger | Thank you for posing "what benefit does a blockchain provide over the people counting signing a message including the counts?" | 04:11 |
JayDugger | That gives me something to consider. | 04:12 |
justanotheruser | since my second question is a bad one, do you have an answer? | 04:12 |
JayDugger | Theft, pilferage, laziness... | 04:12 |
JayDugger | None of which a blockchain magically prevents, I admit. | 04:13 |
justanotheruser | yes, I was about to ask that... | 04:13 |
JayDugger | Socratic method aside, have you got an answer to my original question? | 04:14 |
justanotheruser | which one? | 04:14 |
justanotheruser | 04:10 < JayDugger> Any activity using a blockchain for inventory control? | 04:15 |
JayDugger | Yes. | 04:15 |
justanotheruser | maybe by some company using the blockchain buzzword but not understanding what the blockchain is useful for. | 04:15 |
justanotheruser | www.coindesk.com/ibm-executive-block-chain-internet-of-things/ | 04:15 |
justanotheruser | I actually emailed them.. never got a response | 04:15 |
JayDugger | I vaguely remember that article. Thank you for the link. | 04:16 |
justanotheruser | just an fyi, I think they are either idiots for doing this, or smart enough to promote their product with the buzzword blockchain and it will end up looking nothing like a blockchain | 04:17 |
justanotheruser | Socratic method aside, within a company you are already trusting the CEO and his board of directors along with the US government (fraud law enforcement) to not steal your ownership of the company and dividends. It might be interesting for the US government to do some timestamping for businesses | 04:19 |
JayDugger | Yes, rereading that article reveals why it didn't stick the first time. | 04:19 |
justanotheruser | inconsistencies in the database could be found simply by having them publish a log of changes and have individual auditors and shareholders check it too | 04:19 |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-xhnlxqlewjbnfxns] has joined ##hplusroadmap | 04:20 | |
JayDugger | I think you grossly overestimate the technological savvy of your average bureaucrat and average businessman. | 04:20 |
justanotheruser | JayDugger: then you can't expect a blockchain to keep them accountable either | 04:21 |
justanotheruser | if there is a fork (or whatever the equivalent would be called in centralized auditable accounting systems) then that could be proven | 04:21 |
-!- justanot1eruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 04:24 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Quit: Reconnecting] | 04:24 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 04:26 | |
-!- justanot1eruser [~Justan@unaffiliated/justanotheruser] has quit [Client Quit] | 04:28 | |
-!- weles [~mariusz@wsip-174-78-132-9.ri.ri.cox.net] has joined ##hplusroadmap | 05:20 | |
-!- cntrational [~sid@49.205.77.125] has quit [Quit: WeeChat 1.2-dev] | 05:29 | |
kanzure | auuuuuugh motion to ban JayDugger | 05:36 |
phm4242 | reason(s)? | 05:42 |
archels | he says "good morning" a lot | 05:45 |
archels | I mean, mornings? good? | 05:45 |
archels | burn at the stake! | 05:45 |
pasky | now he wished a good evening too | 05:49 |
maaku | kanzure: if an apt-get for electronics could be done more easily (or an apt-get for 3d printing), I'd do that | 05:50 |
maaku | walk before you run | 05:50 |
justanotheruser | kanzure: its your channel. If you ban his it's probably appropriate to ban nmz as well though :P | 05:51 |
kanzure | phm4242: misappropration of lbockchain concepts | 05:53 |
kanzure | archels: greetings are not entirely banworthy | 05:54 |
kanzure | maaku: 3d printing doesn't work- it's not like people are writing millions of pages of documentation about how to make your 3d printed part. and plus, there's usually no dependencies anyway. | 05:54 |
kanzure | justanotheruser: er, what would be the reasons to ban nmz787? | 05:54 |
justanotheruser | wasn't he misunderstanding bitcoin (in a way that is probably worse) with foldingcoin? | 05:55 |
justanotheruser | well I guess I don't know your reason to want to ban JayDugger, but I assume its misunderstanding the blockchain | 05:56 |
kanzure | right.. i think the banworthy offense there was "pretending to understand bitcoin" or something | 05:56 |
phm4242 | kanzure: Which concept did it misappropriate? | 05:56 |
kanzure | phm4242: everything | 05:56 |
justanotheruser | lol | 05:56 |
justanotheruser | phm4242: the PoW part | 05:57 |
kanzure | the inventory part | 05:57 |
kanzure | the blockchain part | 05:57 |
kanzure | all of it | 05:57 |
justanotheruser | wait is phm nmz or JayDugger | 05:57 |
kanzure | er, it was JayDugger | 05:57 |
kanzure | you were just having the conversation with him | 05:57 |
justanotheruser | yes, I thought it was nmz for some reason | 05:57 |
kanzure | i suppose i've heard stranger things? | 05:58 |
justanotheruser | It's not that bad IMO, foldingcoin is more fundamental | 06:00 |
phm4242 | <jrayhawk> Yeah, I'm quite surprised that didn't happen. I personally didn't do it because watching kanzure bash his head against a wall for several hours is actually kinda funny. | 06:01 |
-!- cntrational [~sid@49.205.77.125] has joined ##hplusroadmap | 06:01 | |
justanotheruser | It's not *that* obvious that when you are merely using the blockchain as an auditing tool rather than a tool to guarantee payment without trust, that you can have a centralized system and produce a fraud proof by auditing the companies records in a similar way. | 06:03 |
kanzure | normal databases should be obvious | 06:04 |
kanzure | there are very very few excuses | 06:04 |
justanotheruser | yes, but fraud proofs aren't | 06:05 |
justanotheruser | if you want a similar effect, you can have the company broadcast timestamped records to whatever auditing parties are interested and they can show conflicting messages as proof | 06:05 |
justanotheruser | but even with a blockchain being unnecessary, I wonder how useful such a fraud proof would be. Seems like there would be many tricks around it. "This inventory is defective". "This business meeting cost $1million (it really cost $500k)" | 06:07 |
phm4242 | timestamps. Why didn't satoshi think of that? | 06:08 |
justanotheruser | ... | 06:08 |
justanotheruser | read what I said again. They accomplish the same thing so the blockchain isn't necessary. I'm not saying it can't be accomplished with a blockchain, I'm saying theres no reason to use a blockchain. | 06:09 |
phm4242 | So why did satoshi? | 06:10 |
justanotheruser | why did he what? | 06:10 |
phm4242 | use a blockchain | 06:10 |
justanotheruser | for trustless decentralized consensus | 06:11 |
maaku | phm4242: bitcoin is solving a different problem | 06:11 |
maaku | distributed concensus with an anonymous and dynamic membership set requires a blockchain | 06:11 |
maaku | it's basically the only thing which does | 06:11 |
-!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has joined ##hplusroadmap | 06:12 | |
pasky | phm4242: you want your decisions recorded so they can't be tampered with in the future, but the key here is whether only a single party is making decisions or whether anyone can do them | 06:13 |
kanzure | you shoud be aware that phm4242 is a known troll and wastes your time | 06:14 |
pasky | anyone knows why is it so friggin cold in tokyo (indoors) in winter? | 06:14 |
pasky | japan, pinnacle of civilization, and they are woefully unaware of concepts of heating rooms | 06:15 |
pasky | s/concepts/the concept/ | 06:15 |
phm4242 | justanotheruser: So when you were talking about 'broadcast timestamped records', trust is not an issue? | 06:16 |
maaku | pasky: the japanese 'aesthetic' is no insulation | 06:18 |
pasky | noone isn't sitting on tatami around here anyway | 06:19 |
kanzure | have you two synchronized your terribe ai plans yet? | 06:20 |
phm4242 | kanzure defines trolling as 'getting told why you're wrong' | 06:25 |
kanzure | haha | 06:26 |
-!- ivan` [~ivan@unaffiliated/ivan/x-000001] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] | 06:26 | |
kanzure | keep trying | 06:26 |
kanzure | aren't you the island fucker? | 06:26 |
phm4242 | the what/ | 06:26 |
kanzure | read the logs | 06:27 |
phm4242 | Why? | 06:27 |
phm4242 | should I grep 'island fucker' or what? | 06:28 |
kanzure | 12:18 < p42___> huh? I want you to come, but I don't want to use violence. | 06:29 |
kanzure | your memory sucks and i hate you for this | 06:29 |
-!- ivan` [~ivan@unaffiliated/ivan/x-000001] has joined ##hplusroadmap | 06:29 | |
phm4242 | What does that have to do with fucking islands? | 06:29 |
kanzure | i just told you to read the log, and now where to find it in the logs, so now you can kindly fuck off and go read it | 06:30 |
phm4242 | I've read the logs already. I still have no idea what you're talking about 'island fucker' for | 06:30 |
maaku | kanzure: dereference "you two" | 06:30 |
kanzure | maaku: pasky | 06:30 |
pasky | I admit I forgot maaku has ai plans :) | 06:30 |
maaku | pasky: aparantly we're both working on (terrible) ai plans | 06:31 |
pasky | maaku: do you have ai plans? what ai plans? | 06:31 |
pasky | maaku: my ai plan is https://github.com/brmson | 06:31 |
maaku | i'm happy to sync on that although i'm literally about to go afk. but my bouncer will catch anything | 06:31 |
pasky | me too | 06:31 |
maaku | awesome. i could use an open source watson | 06:31 |
pasky | well of course my ai plans are more than that, but for the rest, i could use an open source watson | 06:32 |
maaku | my plan in one sentence is a fixed CogPrime design with planning engines and model discovery suitable for answering "sudo build me a nanofactory design implementable with current tech" when hooked up to a density-functional quantum simulator for learning/testing | 06:32 |
maaku | but the next stuff i want to do after that requires watson like capabilities ("design me experiments for accomplishing the SENS objectives, organized as a decision tree") | 06:33 |
maaku | anyway, going afk for a number of hours (but online capturing logs) | 06:34 |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 06:34 | |
-!- cntrational [~sid@49.205.77.125] has quit [Quit: WeeChat 1.2-dev] | 06:34 | |
pasky | i think we'll catch each other later, these are very similar to my plans actually but starting with the watson part seemed most meaningful | 06:34 |
pasky | i want that to do resolution for things like "nanofactory design implementable with current tech", eventually | 06:35 |
pasky | but ideally it'd be useful and making some money for its development much earlier than that | 06:35 |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Client Quit] | 06:36 | |
maaku | i don't think you need watson like tech for the nanofactory example | 06:36 |
maaku | but we can discuss that later. afk for reals | 06:36 |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 06:37 | |
pasky | possibly, maybe I didn't really get the example | 06:37 |
-!- ebowden [~ebowden@CPE-144-131-35-125.lns4.dav.bigpond.net.au] has quit [Remote host closed the connection] | 06:38 | |
justanotheruser | phm4242: trust is an issue, but it isn't any less of an issue with a blockchain | 06:40 |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Client Quit] | 06:40 | |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 06:41 | |
phm4242 | justanotheruser: What are you talking about? Not wanted to trust people is why the blockchain exists. | 06:41 |
kanzure | that's the stupidest thing you've ever said to me | 06:42 |
phm4242 | why? | 06:42 |
kanzure | just because someone makes a transaction that includes some extra data does not make that data trustless you moron | 06:42 |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 06:42 | |
kanzure | how would you even suggest that | 06:42 |
kanzure | why are you wasting our time | 06:42 |
justanotheruser | kanzure: is this the stalker? | 06:43 |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer] | 06:43 | |
kanzure | no | 06:43 |
kanzure | definitely not | 06:43 |
phm4242 | I do stalk Kanz. He just don't know about it. | 06:43 |
justanotheruser | okay, I thought you were saying he was with 12:18 < p42___> huh? I want you to come, but I don't want to use violence. | 06:43 |
kanzure | no i was answering his question about where in the logs | 06:44 |
justanotheruser | I mean the stalker would want you to come to his island to thaw his mom | 06:44 |
kanzure | (that was aorund the time he revealed that he was a fucking idiot) | 06:44 |
kanzure | *around | 06:44 |
chris_99 | look at the IP of p42___ ;) | 06:44 |
justanotheruser | yes | 06:44 |
justanotheruser | it is a hackerspace | 06:44 |
justanotheruser | SOMETHING YOU ARE PROMOTING kanzure | 06:44 |
phm4242 | :-0 | 06:44 |
kanzure | yes hackerspaces often attract totally crazy morons | 06:44 |
justanotheruser | 09:40 < phm4242> justanotheruser: What are you talking about? Not wanted to trust people is why the blockchain exists. | 06:45 |
kanzure | justanotheruser: http://gnusha.org/bitcoin-wizards/2015-01-23.log | 06:45 |
-!- ebowden [~ebowden@CPE-144-131-35-125.lns4.dav.bigpond.net.au] has joined ##hplusroadmap | 06:45 | |
kanzure | whoops | 06:45 |
phm4242 | < jrayhawk> Kanzure is not gracious. | 06:45 |
maaku | so aparantly my wife is delayed and i have more time than I thought | 06:45 |
justanotheruser | Wouldn't it be great if anything done with a blockchain was trust free? | 06:45 |
kanzure | http://gnusha.org/logs/2015-01-23.log | 06:45 |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 06:45 | |
maaku | phm4242: there are very few things about the block chain which are trust-free | 06:46 |
chris_99 | nmz787, about? | 06:46 |
maaku | it does achieve a dynamic, anonymous membership set signature. but that signature covers very little | 06:46 |
ebowden | paperbot: http://www.nature.com/ncomms/2013/130702/ncomms3114/full/ncomms3114.html | 06:46 |
kanzure | no way man anything i publish in a transaction is 100% true | 06:46 |
kanzure | i can't even say it with a straight face :/ | 06:47 |
maaku | and its only what that signature covers, and implications from the validation logic, that you can say is trust-free | 06:48 |
justanotheruser | anything i publish n a transaction is 100% true :| | 06:48 |
justanotheruser | there we go | 06:48 |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 272 seconds] | 06:48 | |
phm4242 | maaku: You mean you mean blockchain technology doesn't work? | 06:49 |
justanotheruser | okay, I change my ban vote to yes | 06:49 |
kanzure | give me a reason to not ban you | 06:49 |
-!- mode/##hplusroadmap [+o kanzure] by ChanServ | 06:49 | |
maaku | phm4242: did you read what I said? it works perfectly fine for what it does. which is a lot less than a lot of people think, but sufficient to build an economy on | 06:50 |
phm4242 | maaku: So what do you mean by 'there are very few things about the block chain which are trust-free'? | 06:50 |
justanotheruser | When I was much younger I thought everyone making a dumb statement was a troll, then I realized there were people so incapable of understanding a concept that they come off as trolls, and now I don't know what to believe | 06:50 |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 276 seconds] | 06:51 | |
@kanzure | justanotheruser: generally, i kick out anyone who is terrible at general reasoning | 06:51 |
justanotheruser | oh wait | 06:51 |
@kanzure | justanotheruser: i don't have enough time to train people to not be idiots | 06:51 |
maaku | anyways. pasky: so kanzure will rip me a new one for not having worked this out in detail, but I think you can represent current capability in a formal planning language | 06:51 |
justanotheruser | I thought jaydugger was tastybuds this first part of the reading | 06:52 |
@kanzure | and idiocy only slows us down | 06:52 |
justanotheruser | what a twist | 06:52 |
@kanzure | JayDugger is a long-time regular, practically ancient | 06:52 |
justanotheruser | .. so phm4242 is jaydugger? | 06:52 |
@kanzure | uh, no | 06:52 |
justanotheruser | wat | 06:52 |
maaku | e.g. AFM depassivation like Zyvex and Moriarty are working on, more blunt AFM tooltip manipulation (which might be probabalistic), etc. | 06:52 |
justanotheruser | 08:56 < justanotheruser> wait is phm nmz or JayDugger | 06:53 |
justanotheruser | 08:56 < kanzure> er, it was JayDugger | 06:53 |
@kanzure | the one asking about inventory was JayDugger | 06:53 |
justanotheruser | so this is tastybuds? | 06:53 |
@kanzure | that's a third person | 06:53 |
justanotheruser | damn, this is a complicated story | 06:54 |
@kanzure | maaku: hey at least you accurately represent my complaints | 06:55 |
justanotheruser | so you wanted to ban jay, phm asked why, and now you want to ban him | 06:55 |
@kanzure | i want to man pete4242 because he's an idiot | 06:55 |
@kanzure | *ban | 06:55 |
justanotheruser | you can see how I thought they were the same person :P | 06:55 |
pasky | maaku: i don't really know anything about nanotech | 06:56 |
maaku | pasky: what I need the watson-like tech for is when we apply the same AGI planner to biology | 06:56 |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Quit: www.byrdirc.com] | 06:56 | |
maaku | because although ab-initio quantum simulators are good enough to do nanotech work, they don't scale up to protiens | 06:57 |
maaku | let alone whole cell simulations | 06:57 |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 06:57 | |
maaku | so i imagine that part of the research objective would have to be accomplished by a watson-like understanding of the existing literature | 06:57 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Ex-Chat] | 06:57 | |
maaku | of course if you have a nanofactory, bootstrapping additional capabilities should be a solved problem :P | 06:58 |
justanotheruser | 11:29 < kanzure> 2 is stupid. typical surface dweller attitude. | 06:58 |
justanotheruser | Is rapture on the h+ roadmap? | 06:58 |
maaku | justanotheruser: define "rapture" | 06:58 |
-!- eudoxia [~eudoxia@r167-56-30-216.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 06:59 | |
justanotheruser | maaku: https://www.youtube.com/watch?v=z3RHenh8vJ8#t=206 | 07:00 |
@kanzure | .title | 07:00 |
yoleaux | Bioshock Intro (720p) - YouTube | 07:00 |
@kanzure | oh brother | 07:00 |
justanotheruser | Rapture seems like a place you would like | 07:01 |
justanotheruser | not sure how easy it would be to make an underwater city though | 07:01 |
justanotheruser | LOL, I love this 12:05 < kanzure> i hate you get the hell out | 07:04 |
eudoxia | kanzure: you still haven't told me your plan to implement skdb | 07:05 |
eudoxia | every second we don't write skdb code is 10^50 planets not colonized by FOSS KSRM probes | 07:05 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Quit: Reconnecting] | 07:11 | |
-!- justanot1eruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 07:12 | |
-!- ivan` [~ivan@unaffiliated/ivan/x-000001] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] | 07:12 | |
@kanzure | i wonder how to convert that into anxiety units | 07:12 |
-!- ivan` [~ivan@192.241.198.49] has joined ##hplusroadmap | 07:13 | |
-!- ivan` [~ivan@192.241.198.49] has quit [Changing host] | 07:13 | |
-!- ivan` [~ivan@unaffiliated/ivan/x-000001] has joined ##hplusroadmap | 07:13 | |
-!- drewbot_ [~cinch@ec2-54-226-96-126.compute-1.amazonaws.com] has quit [Ping timeout: 264 seconds] | 07:16 | |
-!- Merovoth [~Merovoth@gateway/tor-sasl/merovoth] has quit [Remote host closed the connection] | 07:17 | |
-!- drewbot [~cinch@ec2-54-159-143-104.compute-1.amazonaws.com] has joined ##hplusroadmap | 07:20 | |
@kanzure | also, it's in the recent logs | 07:26 |
maaku | KSRM = Kim Stanley Robinson's Mars trilogy? | 07:28 |
maaku | pasky: so I came at AGI because of nanotech | 07:29 |
maaku | i was looking at how to make a nanofactory, and at some point realized it would take an ungodly number of engineering man-years to complete | 07:30 |
maaku | but that's it -- it's basically *just* an engineering problem. every decent university lab has the necessary tools and material | 07:30 |
maaku | so, make an artificial engineer and buy lots of commodity computing hardware to run it seemed a cheaper / faster option | 07:31 |
maaku | using the agi to accomplish other stuff afterwards is just gravy | 07:31 |
@kanzure | ksrm is "kinematic self-replicating machine" | 07:31 |
maaku | ah | 07:32 |
maaku | well the mars triloogy is excelent reading too :) | 07:32 |
-!- Merovoth [~Merovoth@gateway/tor-sasl/merovoth] has joined ##hplusroadmap | 07:37 | |
AmbulatoryCortex | I really wonder if drexlerian nanotech is actually possible, as opposed to to wet nanotech | 07:38 |
justanot1eruser | AmbulatoryCortex: Probably is possible, but also inefficient | 07:41 |
-!- ivan` [~ivan@unaffiliated/ivan/x-000001] has quit [Ping timeout: 252 seconds] | 07:46 | |
-!- xrr [~xrr@gprs-inet-183-27.elisa.ee] has quit [Quit: Leaving] | 07:50 | |
maaku | AmbulatoryCortex: absolutely possible. here's some VASP simulations: http://www.molecularassembler.com/Papers/MinToolset.pdf | 07:51 |
maaku | justanot1eruser: inefficient? | 07:52 |
justanot1eruser | maaku: I'm guessing natures design is better on the atomic scale than replicating our macro scale technologies | 07:53 |
maaku | justanot1eruser: you'd need some insane justifications for that. | 07:55 |
maaku | biology is horribly inefficient | 07:55 |
maaku | biology is putting all the parts in a box, shaking it, and getting the final result | 07:55 |
@kanzure | biology works | 07:56 |
maaku | it's a miracle that it works, and is orders of magnitude slower than physical limits | 07:56 |
justanot1eruser | maaku: and then making that final result slightly different over time towards a local optimum | 07:56 |
eudoxia | the difference is humans optimize for clarity (eg drexler's differential gear) while biology has brutally optimized for raw performance | 07:56 |
@kanzure | not optimized for performance | 07:57 |
maaku | biology is absolutely not optimized for performance | 07:57 |
justanot1eruser | kanzure: no, but survival is often aligned with performance | 07:57 |
maaku | no, it is not | 07:57 |
maaku | it is aligned with redundancy | 07:57 |
maaku | redundancy is counter to performance | 07:57 |
justanot1eruser | ? | 07:57 |
justanot1eruser | I'm talking about survival of an individual organism, are you not? | 07:58 |
AmbulatoryCortex | bacteria manage to reproduce plenty fast in ideal conditions | 07:58 |
@kanzure | geology has no reason to be optimized for performance | 07:58 |
justanot1eruser | how does geology enter here? | 07:58 |
AmbulatoryCortex | guessing mistype | 07:59 |
@kanzure | biology is just a widely misunderstood subfield of geology | 07:59 |
AmbulatoryCortex | lol | 07:59 |
@kanzure | some rocks got up and started walking around and shit | 07:59 |
justanot1eruser | yeah | 08:00 |
justanot1eruser | anyways, organisms that can quickly replicate and use little energy have high performance and can survive | 08:00 |
maaku | justanot1eruser: so i'm in peurto rico, 4,000 mi from my references | 08:02 |
justanot1eruser | life gave us DNA which is probably the best tool for building nanoscale structures in the short term | 08:02 |
maaku | but i believe nanosystems has calculations showing that most of the things biology does can be done 1,000x faster with precision machinery | 08:03 |
justanot1eruser | maaku: That actually sounds reasonable. I did say evolution only brought us to a local optimum. anyways wetware will still probably be our first step to getting to mechanosynthesis | 08:04 |
maaku | which is not surprising ... machines beat biology at macro scale stuff. i think the onus is on you to show biology would be more efficient at the small scale | 08:04 |
maaku | justanot1eruser: people have said that, but no one has come up with a reasonable way to bridge from biology to drexlarian like nanotech | 08:04 |
justanot1eruser | maaku: make really inefficient error prone mechanosynthesis with wetware is my guess | 08:05 |
@kanzure | that's not an actual proposal | 08:05 |
maaku | whereas there at least exists a roadmap for doing so from AFM microscopy | 08:05 |
maaku | and, what kanzure said | 08:05 |
maaku | on the other hand, we have demonstrated mechanosynthesis on gold and silicon surfaces | 08:06 |
justanot1eruser | kanzure: no, it's a guess. I'm not sure we know enough about how we can manipulate DNA to know whether we can/how to create a useful structure | 08:06 |
@kanzure | well, there's dna origami, but you should look at proteins instead | 08:07 |
eudoxia | is dna origami actually useful for anything | 08:07 |
maaku | eudoxia: scaffolding | 08:08 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 272 seconds] | 08:08 | |
-!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 08:08 | |
maaku | justanot1eruser: the problem is that DNA and proteins are all wobbly at the nano scale | 08:09 |
justanot1eruser | maaku: I'm not sure AFMs are the correct path since they're addative and the useful structures will probably be unstable when created in that manner | 08:09 |
pasky | maaku: ah! that's a pretty interesting take - but then again, do you need an agi to build a nanofactory in the first place? i think it's way easier to go about this by using more specialized means, i.e. decomposing the problem to organizing current knowledge, automatically creating structured databases from unstructured data, inference on top o fthe database... | 08:09 |
-!- Viper168_ is now known as Viper168 | 08:09 | |
justanot1eruser | I should specify additive layer by layer | 08:09 |
maaku | justanot1eruser: it's not unlike 3d printing, so long as you can come up with a way of removing material in a bath | 08:10 |
justanot1eruser | maaku: what is the bath? Electrons? | 08:10 |
maaku | it seems zyvex is doing that (additive silicon and germanium then remove the germanium in a path) | 08:10 |
maaku | *bath | 08:10 |
@kanzure | http://diyhpl.us/~bryan/papers2/nanotech/freitas_process/notes.txt | 08:10 |
maaku | but yeah, the freitas process is not strictly additive | 08:10 |
maaku | pasky: "do you need an agi" absolutly not. give me either (a) 500 million dollars and 10 years, or (b) an AGI and a basement lab | 08:11 |
maaku | since I think the agi would cost less than $500m+, that seemed the less resource intensive path | 08:12 |
justanot1eruser | maaku: how much freedom does that process allow? Could I construct arbitrary molecules and wash the stuff I don't need? | 08:12 |
AmbulatoryCortex | course, to make the AGI, you might need the former | 08:12 |
maaku | justanot1eruser: i don't know, zyvex seems pretty tight lipped about what they're doing. you can read the outline of their proposal in the foresight nanotech roadmap though. | 08:12 |
maaku | AmbulatoryCortex: 10 years, yes, the hundreds of millions, probably not | 08:13 |
pasky | maaku: i think the agi will be more expensive than $500m still at this point, since you need a few more damn good ideas, and to get these, you need to spend resources to generate all sorts of bad and good ideas | 08:13 |
maaku | pasky: that's where i differ from mainstream opinion. to build the idiot savant capable of designing a nanofacotry, i think we have most of the good ideas already sorted out | 08:14 |
justanot1eruser | why does the number $500 million keep getting tossed around in this channel? Is it just coincidence? | 08:16 |
@kanzure | budgets, man | 08:16 |
justanot1eruser | is that the 2015 ##hplusroadmap budget? | 08:16 |
@kanzure | well the 2014 budget was just under $3 million | 08:16 |
justanot1eruser | do you have donors? | 08:17 |
maaku | in this case $500m was a random number i came up with on the spot. | 08:17 |
justanot1eruser | ah | 08:17 |
@kanzure | fuck donors | 08:18 |
maaku | freitas and merkle quote $50m - $100m I think, but they're optimistic imho | 08:18 |
@kanzure | 1) do hard work 2) demand payment 3) i'll spend my money however i please | 08:18 |
justanot1eruser | because the guy that almost got banned earlier was talking about a $500m island | 08:18 |
maaku | whereas an AGI could really be 1-2 people over ten years, so ~$1-2m | 08:18 |
-!- ebowden [~ebowden@CPE-144-131-35-125.lns4.dav.bigpond.net.au] has quit [Remote host closed the connection] | 08:19 | |
maaku | http://www.wisdomination.com/screw-motivation-what-you-need-is-discipline/ | 08:20 |
maaku | .title | 08:20 |
yoleaux | Screw motivation, what you need is discipline. | Wisdomination | 08:20 |
@kanzure | screw discipline, what they need is meth or adderall | 08:22 |
@kanzure | possibly cocaine | 08:22 |
maaku | pasky: the "organize current data, structure databases, inference [and planning] on top" works for stuff we know a lot about, e.g. biology | 08:22 |
maaku | hence why I think it's necessary for the SENS research task | 08:23 |
@kanzure | you can get pretty far manually doing that | 08:23 |
@kanzure | http://diyhpl.us/~bryan/papers2/longevity/aging_roadmap.txt | 08:23 |
maaku | but for nanofactories, the amount of info out there is basically nil | 08:23 |
@kanzure | (this guy's thesis was that "everyone sucks and nobody reads the fucking literature, they just want expert systems to do the work for them eventually") | 08:23 |
maaku | most of it is right here: http://www.molecularassembler.com/Nanofactory/Publications.htm | 08:23 |
maaku | heh | 08:23 |
maaku | yeah you could do all of this manually | 08:24 |
maaku | but then you have Aubrey de Gray (sp?) asking for $100m to do just that | 08:24 |
maaku | (also insanly optimistic, i think) | 08:24 |
@kanzure | afaik aubrey doesn't read that much any more | 08:24 |
pasky | maaku: i think i'd need to understand nanotech a lot more to be able to meaningfully discuss engineering it automatically :) | 08:26 |
maaku | fair enough. i'd have to say the same about biology to be honest | 08:26 |
maaku | but i'd love a description of why you're working on agi and what tasks you want it to accomplish similarly | 08:27 |
pasky | I wrote it once (I think even pasted in this channel) but it didn't really make that much sense I guess | 08:30 |
pasky | maaku: http://pasky.or.cz/dev/brmson/grand-plan.pdf | 08:32 |
pasky | re-reading it, i do think it makes sense but is way less concrete than i remembered writing it ;) | 08:32 |
maaku | it made sense, and it is a workable plan | 08:38 |
maaku | if i had to bootstrap agi, that's what I'd do | 08:38 |
maaku | what i'm doing right now is bootstrapping via a different industry :P | 08:38 |
-!- eudoxia [~eudoxia@r167-56-30-216.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 08:39 | |
maaku | i'm involved in a bitcoin startup, and if that has any kind of positive exit, i should be able to skip the compete-with-Google-and-IBM phase and jump straight to "build me a nanofactory" | 08:39 |
-!- DumpsterD1ver [~loki@vpn166.sdf.org] has quit [Quit: Leaving] | 08:39 | |
maaku | but that requires a 5+ year self-funded roadmap | 08:40 |
maaku | if you want to make money now, knowledge organization is probably where it's at | 08:40 |
-!- DumpsterD1ver [~loki@vpn166.sdf.org] has joined ##hplusroadmap | 08:43 | |
maaku | pasky: just be careful to structure it in such a way that you can sell the business to one of those heavyweights without losing the technology | 08:44 |
maaku | e.g. make the tech open source but the database proprietary or something | 08:44 |
maaku | this may require some careful thought | 08:44 |
-!- weles [~mariusz@wsip-174-78-132-9.ri.ri.cox.net] has quit [Ping timeout: 256 seconds] | 08:45 | |
maaku | pasky: what's your background? | 08:45 |
maaku | n/m found your web page | 08:47 |
-!- dustinm` [~dustinm@192.241.142.243] has joined ##hplusroadmap | 08:47 | |
pasky | maaku: so far it's all open source, and I don't have the details of restricting something figured out yet; my working plan is to keep it open source until I'm sure I shouldn't | 08:49 |
maaku | imho it should always be open source | 08:49 |
maaku | i assume this is your research project? any problems from your university if you spin it out? | 08:50 |
@kanzure | pasky's background is the git underground | 08:50 |
@kanzure | and stuff | 08:50 |
@kanzure | actually i have no good way to summarize pasky | 08:51 |
maaku | i'll save a git vs monotone argument for another day | 08:51 |
-!- dustinm` [~dustinm@192.241.142.243] has quit [Ping timeout: 252 seconds] | 08:53 | |
maaku | "answering physics questions in university exams" -- do you have more info about this | 08:53 |
-!- eudoxia [~eudoxia@r167-56-30-216.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 08:53 | |
-!- erasei [erasei@x.kaw.cc] has quit [Ping timeout: 244 seconds] | 08:53 | |
maaku | ? how smart is it? is it able to plug into models, or is it just searching for already worked out examples? | 08:53 |
-!- erasei [erasei@x.kaw.cc] has joined ##hplusroadmap | 08:54 | |
-!- eudoxia [~eudoxia@r167-56-30-216.dialup.adsl.anteldata.net.uy] has quit [Client Quit] | 08:55 | |
maaku | on the roadmap to a "build me a nanofactory", i'm going to build an agent that can figure out and solve problems in introductory physics and engineering books | 08:55 |
maaku | i'm trying to see if what you're doing is related | 08:55 |
-!- eleitl [~eugen@v8.ativel.com] has joined ##hplusroadmap | 08:57 | |
@kanzure | eleitl: hello. maaku and pasky have agi plans. | 08:57 |
@kanzure | and eudoxia wants your autograph | 08:58 |
eleitl | :) | 08:58 |
eleitl | who are maaku and pasky? | 08:58 |
@kanzure | some dudes | 08:59 |
eleitl | dudes with money and clue? | 08:59 |
phm4242 | Nope. Inane AGI speculation. | 08:59 |
-!- phm4242 was kicked from ##hplusroadmap by kanzure [phm4242] | 08:59 | |
@kanzure | ignore that one | 08:59 |
@kanzure | they were here a moment ago... | 08:59 |
@kanzure | have you heard anything from zyvex in forever? | 09:00 |
eleitl | No. But I'm not keeping current. | 09:00 |
pasky | maaku: should be no issues from my university :) i hope to be able to get some investment and spin it off in late spring / summer, i'll see how it goes | 09:00 |
pasky | no idea how to really go about it yet :) | 09:00 |
eleitl | pasky: what is your proposed bootstrap route? | 09:01 |
maaku | eleitl: i'm Mark Friedenbach, previously worked at Transcriptic and NASA's Astrobiology and Lunar Science institutes before that. Currenty co-founder at blockstream.com. Who are you? | 09:01 |
eleitl | Hi Mark. I'm Eugene Leitl. Early transhumanist of diverse interests, worked in cryonics. | 09:01 |
pasky | maaku: regarding the physics questions - I just started to look around a week ago and so far it seems they are basically converting the physics problems to some logical forms with predicates and then spew out pattern-based Modelica statements based on these predicates, and running simulations; I don't really like the approach that much, but it's a difficult problem | 09:01 |
pasky | eleitl: Hi Eugene. I'm Petr Baudis, random programmer and researcher, previously working e.g. on Computer Go and various machine learning stuff, currently working on https://github.com/brmson | 09:02 |
pasky | ...which is a question answering system. | 09:03 |
eleitl | Hi Petr. Pleased to meet you. | 09:03 |
pasky | and I guess this gadget is my proposed bootstrap route to agi if you meant that :) but I'm not actually thinking about agi that hard since building a sensible QA system is hard enough right now. | 09:04 |
@kanzure | eleitl: https://www.youtube.com/watch?v=SRu02F6AOmg&t=40s | 09:04 |
-!- dustinm` [~dustinm@192.241.142.243] has joined ##hplusroadmap | 09:04 | |
eleitl | Pick and place a la gatling. Nice. | 09:05 |
eleitl | pasky: I think AGI is hard enough you have to go biologically inspired if not outright biologically derived. Which is computationally very expensive, obviously. | 09:06 |
eleitl | Another approach is ALife, but that requires even more computation. | 09:06 |
maaku | eleitl: nice to meet you. andytoshi has inspired me to stop cryoprocrastinating. if i have any questions about that stuff i might ask you | 09:06 |
@kanzure | eleitl: i stumbled into this alife paper a while back.... http://diyhpl.us/~bryan/papers2/ai/The%20evolution%20of%20self-replicating%20computer%20organisms%20-%20A.%20N.%20Pargellis.pdf | 09:07 |
maaku | eleitl: you working in the AI space? | 09:07 |
maaku | eleitl: i'll differ with you on that one. biological intelligence is not optimized for being comprehensible | 09:08 |
eleitl | maaky: sure, I can tell you everything you like about cryonics. Unfortunately, probably none of it will be exactly great news. | 09:08 |
pasky | eleitl: so I think that's an argument I had with kanzure back when I posted http://pasky.or.cz/dev/brmson/grand-plan.pdf first | 09:08 |
@kanzure | i was not aware that eleitl agreed with me on biologically-accurate emulation | 09:09 |
@kanzure | however, i always suspected that i was right | 09:09 |
maaku | lol | 09:09 |
eleitl | maaku: I've done my reading on AI since 1980s and had my debates in 1990s/2000s. | 09:09 |
pasky | eleitl: I personally don't think AGI needs to be biologically inspired (or at least not to the gory details of simulating firing neurons etc.) | 09:09 |
eleitl | What exactly do you mean when you say biologically accurate, kanzure? | 09:09 |
@kanzure | i mean "read a bunch of papers, write good emulators" | 09:09 |
@kanzure | *neuroscience papers | 09:09 |
@kanzure | *neurophysiology papers | 09:09 |
eleitl | I agree about that one. | 09:10 |
eleitl | pasky: biological information processing is opaque, so they only chance you have is by going to low level physics, and see what you can learn from digitized neuranatomy | 09:10 |
pasky | eleitl: same with alife, that is not even on my radar at all; that's in the field of the people who want to build a robot who will learn touch objects and understand basic relations like a human child, but I'm looking at AGI from a completely different perspective, I don't even think it needs to be embodied or particularly represented in the physical or any artifical world | 09:10 |
maaku | if your goal is to create an artificial human with all its perks and flaws, yes biology-inspired is probably a good route | 09:10 |
eleitl | So, unfortunately, you will need the pesky spiking neurons, at least initially. | 09:11 |
pasky | eleitl: but I don't want to build an artificial human brain | 09:11 |
pasky | that's not a very interesting task for me | 09:11 |
eleitl | pasky: information processing needs hardware, so of course it is embodied | 09:11 |
maaku | if you just want to automate certain tasks, e.g. engineering, then i find bio-emulation pathways less convincing | 09:11 |
eleitl | Automation is easy, but real intelligence is not. | 09:11 |
eleitl | So I will remain politely skeptical of your effort, unless it involves the total connectome. | 09:12 |
@kanzure | wellll | 09:12 |
pasky | if we start throwing around phrases like "real intelligence", I don't think that's a particularly inspiring discussion for me | 09:12 |
@kanzure | arguably the total connectome is not necessary | 09:12 |
maaku | i'll have to ask you to taboo "real intelligence" here and define what you mean | 09:12 |
@kanzure | like, auditory cortex is probably unnecessary | 09:12 |
@kanzure | and some of the motor cortex can probably be dropped | 09:13 |
maaku | for me the task i'm interested in is creative problem solving and model building | 09:13 |
eleitl | I define real intelligence as ability of artificial system to occupy 50% of better of all human work niches | 09:13 |
@kanzure | why not just human-like cognitive ability | 09:13 |
maaku | eleitl: right, ok. so we're talking about different things | 09:13 |
eleitl | So not just spikes in ability landscapes, but flooding the entire space | 09:13 |
@kanzure | (or is that something you would ask to taboo as well?) (and if so, why) | 09:13 |
maaku | for me AGI is an ability to solve problems in arbitrary, unexpected domains | 09:14 |
eleitl | if it can't compete with people across the board is not really intelligence | 09:14 |
eleitl | maaku: that's what people do, yes | 09:14 |
maaku | eleitl: people do much more than that though | 09:14 |
maaku | if my AGI is unable to express what it feels like to fall in love, i'm okay with that | 09:15 |
eleitl | I have not been able to find a better benchmark than the 50% definition above. | 09:15 |
eleitl | why should your AI be unable to learn what falling in love means? | 09:15 |
eleitl | There are people who have never fallen in love, so all they can do is read about that. | 09:15 |
@kanzure | this is going to escalate into the wrong conversation ("is it okay to not experience love but still understand it") | 09:16 |
maaku | eh... yeah i think kanzure is right | 09:16 |
maaku | this is not the place for that conversation | 09:16 |
eleitl | I'm trying to use a definition of abilities people are being paid for | 09:16 |
eleitl | As a benchmark it's not all that bad, and measurable. | 09:17 |
maaku | what I was trying to get at before was "for what reason do you want to make an AGI?" | 09:17 |
@kanzure | eleitl: why does GRG suck so much? | 09:17 |
eleitl | Old gerontologists can't evolve. | 09:17 |
@kanzure | but why not. is that going to happen to me? | 09:17 |
maaku | if your task is "replace humans 50% of the time", then having a human-inspired intelligence is a good idea | 09:17 |
eleitl | An AGI is a tool to restart progress, which has been stalling. | 09:18 |
pasky | eleitl: so how would you measure it? because it doesn't seem all that unlikely to me that machines today occupy 50% of better of all human work niches compared to 300 years ago | 09:18 |
eleitl | An AGI which is not useful is of no interest to me. | 09:18 |
maaku | if your goal is "automate things humans find hard" -- what i'm doing -- then there's good reason to believe that human-inspired AGI is an inefficient path | 09:18 |
eleitl | pasky: the number of employees hired. | 09:18 |
eleitl | Human intelligence is young, and if we have figured out the basics we can improve it reasonably quickly, if it does not require embodiment. | 09:19 |
pasky | progress has been stalling? I must be living in a different world! | 09:19 |
eleitl | If a path is sterile, it might be the most efficient one, but it's still irrelevant in practice. | 09:19 |
pasky | I kind of like the definition "ability to solve problems in arbitrary, unexpected domains" right now | 09:19 |
eleitl | This is pretty much the textbook definition, but as a metric it's useless. | 09:20 |
eleitl | By looking at the job market for people vs. AI you're letting the whole world evaluate your systems. All you do is collecting statistics. | 09:20 |
eleitl | By 50% or better you're avoiding the asymptotic issues of 100%. | 09:21 |
@kanzure | eh, maybe you cna extract some evolutionary pressure from job markets, but i dunno | 09:21 |
maaku | eleitl: it's very much a useful metric. | 09:22 |
maaku | i write a program that is able to answer physic and engineering questions when hooked up to a newtonian physics simulator | 09:22 |
eleitl | maaku: what is your measurement procedure for intelligence? | 09:22 |
eleitl | Mine is job statistics. Easy as pi. | 09:22 |
maaku | when it's able to do that, try it on chemistry questions when hooked up to a density-functional simulator | 09:22 |
maaku | then ask it to build nanomachines | 09:22 |
eleitl | maaku: DFT is probably not the right tool, but you're exactly right | 09:23 |
maaku | eleitl: does it get the job done. | 09:23 |
maaku | i don't think that kind of measurement is useful or interesting | 09:23 |
maaku | i'm working on AGI to solve a problem. solving that problem is interesting | 09:23 |
eleitl | I happen to disagree. The history of AI tends to be full of empty grandiosity. I prefer results. | 09:24 |
maaku | eleitl: DFT is not the right tool for which part? | 09:24 |
@kanzure | maaku: it's a little unfair for you to bring up newtonian physics simulators after our previous conversations. i mean, at least address the original objections.... | 09:25 |
eleitl | Are you trying to do um or pm? I happen to think that the chemistry part is boring. So you need to simulate ~ms of ~um^3. | 09:25 |
maaku | kanzure: i'm blanking on the specific critique you are talking about | 09:25 |
@kanzure | "making a perfect simulator is also an impossible task" | 09:26 |
eleitl | I don't really understand what maaku's point is. | 09:26 |
@kanzure | "machine learning tends to be optimized for the simulation you write" | 09:26 |
eleitl | Who cares about anything perfect? | 09:26 |
maaku | in this case i'm talking about a trainign set -- given a newtonian physics simulator, learn the model that makes that simulator work and use it to answer questions from a textbook on newtonian physics | 09:26 |
@kanzure | machine learning will exploit the fuzzyness in the simulator | 09:26 |
@kanzure | that is why you would care | 09:26 |
pasky | eleitl: your metric maybe makes sense (if you can measure it; I have the trouble with the "Better 50%", I don't know how to define that), but it says "have we got agi yet, for a wihle now?" | 09:26 |
eleitl | 50% or more of all job slots occupied by nonhumans. | 09:27 |
pasky | that seems like kinda silly question to me, I'd like to ask "how I just built an agi?" or even "okay, what remains to be done to build an agi?" your metric doesn't help me with that | 09:27 |
eleitl | *All* job slots. CEO, gardener, physician, nanotechnologist. | 09:27 |
eleitl | Actually, your AGI would be capable of starting to fill up the slots -- all of them -- if it's good enough. | 09:28 |
eleitl | Before, you're limited to simulations. | 09:28 |
eleitl | That tells you something, see ALife progress. | 09:28 |
pasky | eleitl: oh like that! I didn't even realize you mean it this way; the notion that AGI would hold "jobs" seems most extraordinary to me | 09:28 |
maaku | kanzure: right well this is why we run the tests on newtonian physics -- I'm sure that the AGI will early on find edge cases in the simulator and abuse them | 09:28 |
maaku | but it's easier to tell when that happens | 09:28 |
eleitl | pasky: yes, definitely, artificial systems are already competing on the job market, but only in isolated areas | 09:29 |
eleitl | A real general intelligence would be able to do any job, orelse it wouldn't be actually general | 09:29 |
maaku | vs when you ask it to design a nanofactory and it comes up with something weird. maybe the thing it found is real? maybe not? we don't know | 09:29 |
maaku | eleitl: no, artificial systems pretty much run our world today | 09:30 |
eleitl | If it does the job it doesn't matter how weird it is | 09:30 |
pasky | eleitl: but the robots in car factories definitely are replacing human jobs; but they don't hold jobs themselves - and they create other human jobs for robot maintenance; I would by default assume it'll be the same way with AGI | 09:30 |
maaku | from air traffic control to distribution networks to architectural design to layout of supermarkets, etc. | 09:30 |
pasky | just as regular computers and their databases replaced huge swathes of low-paid office workers since the 50s | 09:30 |
pasky | or there used to be an actual job of being a "computer", that is carrying out arithmetic calculations | 09:30 |
eleitl | Precisely my point: you can already compute a human equivalence index by looking at the job market. | 09:31 |
eleitl | But you'll notice glaring absence of machines in certain wide areas of it. | 09:31 |
eleitl | That's because our current systems are idiot savants. | 09:31 |
maaku | i don't know how you would -- exactly how many people would it take to manually do what a narrow AI system does today? unclear because capabilities differ | 09:31 |
pasky | but what do I learn from the job market? okay it has this structure now, it had that structure in 1915; many jobs popped up, many disappeared, some because of machines, some for different reason | 09:31 |
pasky | how do I derive your metric from that? | 09:31 |
maaku | but i think it's arguing a point i find uninteresting | 09:32 |
eleitl | I find it interesting what you find unininteresting. | 09:32 |
@kanzure | eleitl: i think GRG has no good plans | 09:33 |
pasky | I wonder if we have less prostitues because of all the porn on the internet | 09:33 |
@kanzure | eleitl: someone needs to sit down and write useful plans for $10M, $100M, $10B in funding for our various mutual problem areas | 09:34 |
maaku | kanzure: that would be useful, although probably a rabbit hole of disagreement | 09:34 |
@kanzure | that's fine actually | 09:34 |
@kanzure | having separate disagreeing plans is fine | 09:34 |
maaku | since most of us compute that based on gut feeling, not actual costed plans | 09:34 |
@kanzure | having them at all is useful | 09:34 |
eleitl | I agree GRG is not making progress | 09:34 |
maaku | GRG? | 09:34 |
@kanzure | they proposed a weekly conference call -_- | 09:34 |
@kanzure | .wik http://en.wikipedia.org/wiki/Gerontology_Research_Group | 09:35 |
yoleaux | "The Gerontology Research Group (GRG) is a global group of researchers in various fields that is known for verifying and tracking supercentenarians, or people who are at least 110 years old. The group also aims to further gerontology research with a goal of reversing or slowing aging. It was founded in 1990 by L. Stephen Coles and Stephen M." — http://en.wikipedia.org/wiki/Gerontology_Research_Group | 09:35 |
eleitl | Gerontologists are conservative, and they don't learn from the dead gerontologists. | 09:35 |
eleitl | http://chronopause.com/chronopause.com/index.php/2011/05/30/going-going-gone/ | 09:36 |
eleitl | Few people remember who these even were. | 09:36 |
@kanzure | bunch of dead people | 09:36 |
eleitl | Right. That so many with high hopes died should send the other ones a message. | 09:36 |
@kanzure | you mean high hopes is insufficient? whaaaat | 09:37 |
eleitl | Yet they still continue to believe they're the ones who're going to make the breakthrough. | 09:37 |
eleitl | But there isn't any breakthrough to make as far as I can tell. It's an issue of control. | 09:37 |
@kanzure | bodies are dumb anyway | 09:37 |
maaku | i'm sorry i have a mental filter regarding most ageing research | 09:37 |
eleitl | maaku: it's wise of you. I still track it on a low priority, should something interesting emerge. | 09:38 |
eleitl | in 30 years, nothing interesting has. | 09:38 |
@kanzure | emulation seems like a more viable option | 09:38 |
maaku | some people were god damn lucky in the gene lottery, and others probably extended it a few years or decades through good habits | 09:38 |
@kanzure | and also has the benefit of fixing most of the aging problems | 09:38 |
eleitl | even if you're lucky, it buys you perhaps 20-30 years. | 09:39 |
eleitl | Not bad, but not game-changers. | 09:39 |
maaku | but it seems unlikely to yield a better payoff than "exercise, eat right, and don't smoke" | 09:39 |
maaku | anyway i only need to live 30-ish more years... | 09:39 |
@kanzure | "store your wisdom teeth" is some fall-out and good advice | 09:39 |
eleitl | I'm popping dozens of pills, but I'm making no illusions about their effect on longevity. | 09:40 |
eleitl | It keeps my vitals in decent shape, though. | 09:40 |
@kanzure | nah vitamins are just a scam | 09:40 |
maaku | eleitl: however i do realize that being an 80's baby i'm fortunate to have the luxury to ignore it though | 09:41 |
eleitl | "I only need to live 30-ish more years" -- that's so many were saying in 1980s and 1990s. | 09:41 |
eleitl | Quite a few of them are dead now. | 09:41 |
eleitl | maaku: you're not different from people who were born last century | 09:41 |
@kanzure | eleitl: does it make you feel better that you outlived them | 09:41 |
@kanzure | eleitl: because it should | 09:41 |
maaku | eleitl: the world i live in is | 09:41 |
eleitl | don't fool yourself, you're the easiest person to fool | 09:41 |
@kanzure | "this time it's different" | 09:42 |
eleitl | there is no point in feeling good outliving the old and diseased | 09:42 |
maaku | eleitl: the last generation (sorry) made a mistake that is the variant of the planning fallacy | 09:42 |
@kanzure | .to fenn http://parts.io/ | 09:42 |
yoleaux | kanzure: I'll pass your message to fenn. | 09:42 |
maaku | we/they could see that radical life extension was possible | 09:42 |
eleitl | they're poor bastards, and we're poor bastards, too | 09:42 |
@kanzure | speak for yourself | 09:42 |
maaku | as humans we estimate things that seem crazy hard as 30 years away. it's just our nature | 09:43 |
eleitl | maaku: my model of the future is that it is worse than the past | 09:43 |
eleitl | Unlike what many think technical progress has stalled across the board. | 09:43 |
pasky | ah, and that again :) | 09:43 |
pasky | gee | 09:43 |
eleitl | We're still making progress in some select areas, but that is dependant on a number of external factors that are not going in a good direction. | 09:44 |
eleitl | Exponentials are not the norm, and they never last long. | 09:44 |
@kanzure | yeah he's sort of the eternally optimistic pessimist | 09:44 |
eleitl | I'm an optimist because I believe it's all would have been completely avoidable | 09:45 |
maaku | on the other hand, now we actually have technical pathways to achieving it, or at least the Kurzweilian graphs which show it happening in 30-ish years | 09:45 |
maaku | eleitl: I have absolutely no way to relate to that | 09:45 |
@kanzure | kurzweil is my bitch | 09:45 |
eleitl | maaku: Kurzweil is a kook | 09:45 |
maaku | eleitl: wat.. i just.. wwhat? explonential progress is the only truly universal fact of human history | 09:46 |
maaku | it has always held true in every culture in every era | 09:46 |
maaku | at some point it will stop. it must as we hit physical limits. but there's still plenty of room at the bottom... | 09:47 |
maaku | (well it won't stop, it's just we'll start the exponential expansion outward into space) | 09:47 |
eleitl | Exponentials don't last long, and be it for relativistic limits. | 09:48 |
eleitl | Until you can pop off recursive spacetimes with traversable portals. | 09:48 |
maaku | ok true. at some point we simply become quadratic as we expand outwards in all direcctions. | 09:48 |
* pasky became convinced, hmm, 12 minutes ago that this discussion is a waste of time | 09:48 | |
maaku | but it doesn't *stop* | 09:48 |
pasky | maaku: I know, someone is wrong on the internet! but... | 09:48 |
eleitl | maaku: I know many people who think like you. These people should take the time analyzing the physical layer issues. They might learn something in the process. | 09:49 |
eleitl | pasky is exactly right | 09:49 |
eleitl | I'm not debating, just going through the notions | 09:49 |
pasky | that was probably my biggest mistaken assumption :) | 09:49 |
eleitl | I've had such debates in the last 30 years, and I'm tired of them. | 09:50 |
* maaku is wasting time while he waits for somebody | 09:50 | |
pasky | oh, still? wow | 09:50 |
maaku | eleitl: then what are you doing here? | 09:50 |
pasky | eleitl: you should've said that before :) | 09:50 |
eleitl | They never convince anyone. | 09:50 |
maaku | i can't imagine a point of view more counter to ##hplusroadmap | 09:50 |
@kanzure | he is here because this is unfortunately the only place that has any ambition or skill or money | 09:50 |
maaku | we have money? | 09:51 |
@kanzure | and also because i kick out all the fucktards | 09:51 |
eleitl | kanzure is unfortunately correct | 09:51 |
maaku | :P | 09:51 |
eleitl | it is easy to get money if your ideas sound right | 09:51 |
@kanzure | (or you, you know, have an actual income) | 09:51 |
@kanzure | (or uh, ahem, inheritance) | 09:52 |
eleitl | AGI still makes a good pitch to the right kind of people | 09:52 |
eleitl | I would milk the time while it lasts | 09:52 |
@kanzure | eleitl: i think the emulation pitch is pretty strong right now | 09:53 |
@kanzure | but it also sounds synonymous with hard work, so i can see why some would prefer to avoid it | 09:53 |
eleitl | it unfortunately takes a lot more money that that kind of funding can raise | 09:53 |
@kanzure | markham got a billion eurobucks | 09:54 |
maaku | it's finally, finally back in vogue | 09:54 |
@kanzure | uh, markram | 09:54 |
@kanzure | wow i got his name wrong. that doesn't happen often. | 09:54 |
maaku | but i'm not sure i'm so optimistic | 09:54 |
eleitl | Yeah, and produced a major eclat in the process. | 09:54 |
maaku | bitcoin seems a better path to money at the moment | 09:54 |
eleitl | And one billion in connectomics is chump change. | 09:55 |
eleitl | Remember exascale isn't even here yet, and it will be delayed | 09:55 |
@kanzure | eclat? | 09:55 |
eleitl | He might well burn all the money, and then have very little to show for it. | 09:56 |
eleitl | Brouhaha. The neuroscientists think theirs is a zero sum budget. | 09:56 |
maaku | pasky: this is why no matter what you make sure your project stays open source | 09:56 |
maaku | find some other way to ensure competative advantage despite your tech being open source | 09:56 |
eleitl | They think they should have gotten part of Markram's money. | 09:56 |
eleitl | maaku: what kind of hardware are you currently using? | 09:57 |
maaku | that way if/when you exit or go under, the tech stays usable and not trapped | 09:57 |
@kanzure | eleitl: markram is a neuroscientist, though | 09:58 |
eleitl | Open source is a given, but if you take money from venture capitalists, there are plenty of strings attached | 09:58 |
@kanzure | eleitl: yeah i did see lots of people complaining that they weren't given a chunk | 09:58 |
@kanzure | open source strategy is too subtle for most people to explain | 09:58 |
eleitl | Yes, but the rest of them don't believe he is truly one of them. And they want a share of the pie. | 09:58 |
@kanzure | trezor recently rewrote their git commit history because they didn't understand what their open source licensing was useful for | 09:58 |
pasky | what kind of strings? | 09:58 |
eleitl | control of IP | 09:58 |
@kanzure | wait why don't they think markram is a neuroscientist? | 09:59 |
pasky | kanzure: it was a PR stunt, more or less, they reverted it | 09:59 |
maaku | eleitl: for AGI? i'm still at the fooling-around-on-my-laptop stage | 09:59 |
@kanzure | pasky: is it possible that they are idiots? | 09:59 |
maaku | actually more of fooling-around-on-the-whiteboard | 09:59 |
eleitl | maaku: would you scale to millions and billions of nodes? | 09:59 |
maaku | it's a hobby project, i do bitcoin at my dayjob (and being co-founder, it consumes more than 40hrs a week) | 09:59 |
@kanzure | 40 hours + 1 soul | 09:59 |
@kanzure | + 1 marriage | 10:00 |
maaku | heh, yeah | 10:00 |
maaku | that describes it well | 10:00 |
maaku | we actually have a 40hr work culture | 10:00 |
pasky | kanzure: I know them personally and at least stickac has plenty of open source experience; it seems it's complicated and they are in a kinda tough spot and some of them indeed panicked | 10:00 |
maaku | but being a co-owner really throws that out the window | 10:00 |
@kanzure | pasky: right.... i have enough problems trying to explain an open source strategy to nmz787 for example. i can't imagine working with more than one person that doesn't understand open source. | 10:00 |
eleitl | maaku: are you at a bank, or an exchange? | 10:00 |
maaku | eleitl: bitcoin core tech company | 10:01 |
maaku | http://blockstream.com/about/ | 10:01 |
maaku | eleitl: i think millions let alone billions is overkill, although it could scale | 10:01 |
@kanzure | and me: https://ledgerx.com/about-the-team/ | 10:01 |
maaku | oh yeah kanzure is in the bitcoin-riches-plz camp too :) | 10:02 |
maaku | i'm imagining 10k to 100k nodes for the applications i'm considering | 10:02 |
maaku | so, a small datacenter | 10:02 |
@kanzure | technically i am in the compliance game | 10:02 |
eleitl | Oh, you've got Adam Back. | 10:02 |
@kanzure | yes they are serious | 10:02 |
pasky | (fwiw, i tried to jump the bitcoin bandwagon once but now i think it's crazy to bet much money on it and it's likely to stay a niche for quite a long while now) | 10:02 |
maaku | and Greg Maxwell and Pieter Wuille ;) | 10:03 |
eleitl | maaku: do you do asynchronous shared-nothing message-passing? | 10:03 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 10:03 | |
maaku | we're considering it but not doing anything with it at this time | 10:03 |
maaku | there are some useful protocols that would benefit from that capability | 10:03 |
@kanzure | .g site:blockstream.com filetype:pdf sidechains | 10:04 |
yoleaux | http://www.blockstream.com/sidechains.pdf | 10:04 |
eleitl | if you want to do AGI on a sea of nodes, you need to do that. Nothing else will scale. | 10:04 |
eleitl | I haven't been reading bitcoin-dev@ in a while. | 10:04 |
@kanzure | eleitl: opinion on mixmaster or mixminion or anonymous remailers? | 10:06 |
maaku | oh you mean in the AGI design? i haven't 'been considering any secure message passing architectures | 10:07 |
maaku | i'm assuming it's all in one data center with physical security | 10:07 |
maaku | well.. sort-of. i also have a nuclear option downloadable client that uses a bitcoin-derived p2p layer for connecting nodes | 10:07 |
eleitl | maaku: I'm talking about supercomputer architectures, where message passing has zero security | 10:08 |
eleitl | you might believe you don't need that much crunch, but you'll need that much crunch | 10:09 |
maaku | it would be crazy irresponsible to deploy, but could work if hooked up to a dead-mans switch as a protection against harassment by crazy LW people | 10:09 |
maaku | eleitl: oh given we were talking about bitcoin i thought you meant bitmessage or mixmaster like systems :) | 10:09 |
eleitl | residential WAN latency and throughput are really not relevant to large scale minds, unless the nodes are very fat indeed. And they aren't. | 10:09 |
maaku | yeah obviously some sort of message passing system would be used, but until operational I don't have reason to believe that OpenMPI based stuff would be insuffient.. | 10:10 |
eleitl | you can't have shared memory, and you have to take care of the data/code logistics | 10:11 |
eleitl | think 3d torus, and map your stuff there | 10:11 |
eleitl | if you're doing SMP work it doesn't really scale beyond 10^3 nodes tops | 10:12 |
eleitl | a good testbed would be developing on a Xeon Phi | 10:14 |
eleitl | older Xeon Phi are a dize a dozen | 10:14 |
eleitl | s/dize/dime | 10:14 |
maaku | the architecture i'm considering would be message passing | 10:15 |
maaku | and resiliant to noisy/lost messages | 10:15 |
eleitl | that would be remarkably fitting | 10:15 |
eleitl | add to it small-world network assumption, and you will go places | 10:16 |
eleitl | (torus is small-world) | 10:16 |
eleitl | allright, I need to run off to spend time with the family | 10:16 |
eleitl | y'all take care | 10:16 |
-!- eleitl [~eugen@v8.ativel.com] has quit [Quit: leaving] | 10:17 | |
maaku | thanks for the advise eleitl | 10:17 |
maaku | oh he left | 10:17 |
-!- FourFire [FourFire@2a02:270:2015:cafe:c148:1a38:1e7e:49f3] has joined ##hplusroadmap | 10:20 | |
-!- eudoxia [~eudoxia@r167-56-30-216.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 10:24 | |
eudoxia | oh no, i missed the leitl | 10:26 |
maaku | eudoxia: eleitl is seemingly well respected here? | 10:32 |
eudoxia | well you'd expect that to be the case given this is actually his fan club | 10:32 |
maaku | ? | 10:32 |
@kanzure | this is the official eugen leitl fan club | 10:33 |
@kanzure | the internet's first and only | 10:33 |
@kanzure | (our meetings are every tuesday at midnight) | 10:33 |
maaku | i had seriously never heard of him before | 10:35 |
@kanzure | there are not that many transhumanists out there, you know | 10:36 |
@kanzure | eudoxia: best to use .to or .tell i suppose | 10:37 |
-!- mode/##hplusroadmap [-o kanzure] by kanzure | 10:39 | |
maaku | what has he done that is deserving of fandom? (just curious) | 10:41 |
kanzure | lately, cryonics, but also he's been known to put up great defenses against retard transhumanists over email | 10:42 |
kanzure | http://lists.extropy.org/pipermail/extropy-chat/2008-August/044716.html | 10:43 |
AmbulatoryCortex | I understand some of those words, kanzure! | 10:46 |
AmbulatoryCortex | progress! | 10:46 |
maaku | his anti-progress views are strange | 10:47 |
maaku | i couldn't tell if he was trolling me or not | 10:48 |
eudoxia | he is a peak-oilist | 10:48 |
kanzure | like i said, he's optimist-pessimist | 10:48 |
kanzure | he doesn't believe there's been *no* progress | 10:48 |
eudoxia | in one of his reddit posts he summarized his views as, basically: there's ongoing progress but also ongoing environmental destruction etc., we'll have to see which trend beats the other | 10:49 |
kanzure | meh | 10:49 |
kanzure | i have never cared for that particular subset of his views :p | 10:50 |
kanzure | anyway, he's not afraid of doing work so that is a useful property of ihm as well | 10:50 |
-!- delinquentme [~delinquen@162-245-22-166.v250d.PUBLIC.monkeybrains.net] has joined ##hplusroadmap | 10:50 | |
kanzure | whereas i spent a lot of time trying to shame the world transhumanist association into doing transhumanist tech projects, which did not go well at all | 10:50 |
eudoxia | i don't think i've heard that particular story | 10:51 |
maaku | environmental destruction arguments are nonsense -- and i say this as an enviralmentalist | 10:51 |
FourFire | kanzure, best to be friendly with the few transhumanists you happen upon then, right? | 10:52 |
kanzure | no | 10:52 |
kanzure | best to kick them out and murder them | 10:52 |
FourFire | why not? | 10:52 |
FourFire | why so? | 10:52 |
kanzure | because they are a waste of my time and resources | 10:52 |
maaku | "environmental destruction" means changing the environment in ways that we don't value, but it's a logical fallacy to jump to that destroying human intelligence & progress | 10:52 |
kanzure | and they are completely ineffective | 10:52 |
FourFire | and you are? | 10:52 |
kanzure | in the 200 years of their ideation never has one of them actually decided to learn some biology | 10:52 |
kanzure | yes i have a demonstrated track record of writing good code, at least | 10:52 |
FourFire | srs? | 10:52 |
maaku | FourFire: the extropian community has been less than worthless with respect to achieving their goals | 10:53 |
kanzure | sitting around watching the stock market for the singularity is not a good plan of action | 10:53 |
FourFire | I had no idea, I thought you were just acting arseholish for no particular reason besides personality deficits | 10:53 |
maaku | with some notable exceptions -- freitas and merkle being my favorites | 10:53 |
kanzure | go fuck yourself | 10:53 |
kanzure | argh what is wrong with you people | 10:53 |
kanzure | what do you think this channel is for | 10:53 |
FourFire | maaku, well you can't have negative progress, unless you mean PR wise | 10:53 |
FourFire | but sure | 10:53 |
kanzure | i am an asshole towards you because you don't do anything | 10:54 |
maaku | FourFire: sure you can. you can convince other people that "sitting aroun watching the stock market for the singularity" is all they need to do | 10:54 |
kanzure | and also because your ideas are often bad | 10:54 |
maaku | thanks, Kurzweil | 10:54 |
FourFire | kanzure, sharing progress, maybe encouraging eachother, not being arseholes at anyone who is genuinely interested | 10:54 |
kanzure | i am not going to encourage you. you have only produced crap, in the past. | 10:54 |
maaku | we need people saying "YOU ARE GOING TO DIE UNLESS YOU HUNKER DOWN AND GET TO WORK!" | 10:55 |
FourFire | I was unaware that I've actually produced anything, have I claimed to in the past? | 10:55 |
maaku | not, "sign up for cryonics and you'll be ok!" | 10:55 |
FourFire | maaku, yeah, I should do that to myself | 10:55 |
kanzure | you have certainly brought up lots of stuff in here that has been less than stellar | 10:55 |
kanzure | and also, your lack of production and work is also a good explanation of my dislike of you | 10:55 |
kanzure | so that's good news | 10:55 |
kanzure | at least i'm consistent | 10:55 |
maaku | i don't care about past performance as long as you are willing to make an effort now | 10:56 |
maaku | (also I don't know what FourFire has worked on, so can't judge) | 10:57 |
kanzure | he has worked on bringing lesswrong crap in here | 10:57 |
kanzure | actually i'm pretty upset that he just interprets all of my comments to him as a personality defecit. instead of, you know, reading what i'm saying. really? there's no other conceivable reason for my intense anger at your poor reasoning skills? | 10:59 |
kanzure | maybe how you ignore my explanations about what this channel is for? nah that's impossible, right? | 10:59 |
FourFire | as far as I'm concerned and recall I haven't stated what I've worked on in here before, though sure, me presenting ideas which aren't 100% constructive could seem like a dick move to someone who has spent time building an optimal environment fro collaboration, but I don't consider that to be the case | 10:59 |
maaku | then FourFire, step one is apply your now mastered rationality skills to seeing if any of those unconventional lesswrong ideas actually make sense (hint: many of them don't) | 11:00 |
FourFire | the majority of my interest in LW and the surrounding community has been in improving my personal effectiveness towards the goals I see fit, and indeed more accurately determining what those goals are | 11:00 |
maaku | what has disgusted me about transhumanism is that people get the message: "steps to immortality: (1) sign up for cryonics, (2) sit back and chat while watching those smooth exponentials, (3) there is no step three, but enjoy the singularity!" | 11:00 |
kanzure | that's certainly digusting | 11:01 |
kanzure | er.... disgusting.. | 11:01 |
FourFire | ok, well cryonics for one has always been a last resort thing ever since I found out about it, I consider it to be 99.5% as bad as dying in any other way | 11:01 |
FourFire | but then I have similar reservations about uploading | 11:02 |
kanzure | although, to be fair, i would argue that the majority of transhumanists do not sign up for cryonics by default | 11:02 |
kanzure | which is perhaps even worse | 11:02 |
maaku | (FourFire: I post on LW too by the way, though I get low karma for being counter-culutural there) | 11:02 |
FourFire | kanzure, wat, how can you have it both ways? | 11:02 |
kanzure | haha maau is counter-cultural? | 11:02 |
kanzure | FourFire: if doing nothing is bad, then doing cryonics is only slightly better | 11:02 |
AmbulatoryCortex | cyronics still shreds your cells, unless there's been some development I'm not aware of recently | 11:02 |
kanzure | *haha maaku is counter-cultural? | 11:02 |
FourFire | maaku, I posted a little but not anything of consequence | 11:03 |
FourFire | AmbulatoryCortex, much better than your atoms being all over the place | 11:03 |
kanzure | it's stupid that transhumanism is actually counter-cultural there. they don't know their own skin from the walls. | 11:03 |
FourFire | at least with cryonics they're sorta in the same positions that they were when you were alive | 11:03 |
FourFire | ok, fine, here is an idea by a quantum physicist I know for making cheap, dense flash memory: http://memory.oyhus.no/ | 11:04 |
FourFire | if someone here can get it made then I consider that progress! | 11:05 |
maaku | kanzure: i'm not sure transhumanism is counter-cultural there. they're mostly pro-cryonics, pro-life extension, etc. | 11:05 |
FourFire | I'm shortly going to purchase the OpenBCI kit and donate it to the hackerspace I participate in | 11:06 |
kanzure | just not pro-doing-anything | 11:06 |
AmbulatoryCortex | I think cryonics is a false hope | 11:06 |
kanzure | it's not a hope. it's just a better alternative to cremation. | 11:06 |
eudoxia | maaku: i think kanzure meant they're kurzweilian transhumanists rather than roll up the sleeves and pump glycol into a cadaver transhumanists | 11:06 |
maaku | AmbulatoryCortex: cryonics is better than the alternative, that's all | 11:06 |
maaku | eudoxia: oh they're worse than that. if you actually roll up your sleeves and work on this, you're murdering humanity! | 11:07 |
FourFire | during my free time I will do some biofeedback experiments and afterwards attempt to make a collab team to program a Brain Computer Interface software | 11:07 |
maaku | seriously i've gotten death threats on #lesswrong (which is why i'm no longer there) | 11:07 |
kanzure | neat | 11:07 |
FourFire | what?? | 11:07 |
AmbulatoryCortex | nice | 11:07 |
FourFire | by who? | 11:07 |
eudoxia | that's got to be some kind of achievement | 11:07 |
kanzure | maaku: well according to the etc group, i'm a bioterrorist and according to eliezer i'm going to blow up the planet (because transhumanists/ai are bad, dawg) | 11:07 |
eudoxia | 'told to stop working on AI by a brony on #lesswrong' | 11:07 |
FourFire | yeah, I mean I've said all manner of off things in there | 11:07 |
FourFire | the worst I've gotten is temp bans for "time travelling" | 11:08 |
maaku | i didn't recognize or remember the nick, although i'm sure I cound find it in the logs. probably some kid | 11:08 |
kanzure | might have been me | 11:08 |
maaku | wasn't you | 11:09 |
kanzure | :( | 11:09 |
maaku | but yeah, i wasn't joking earlier about the dead-mans switch to release a UFAI | 11:09 |
kanzure | hm? | 11:09 |
maaku | and it's fucked up that this should even be an issue | 11:09 |
eudoxia | wait who has a UFAI on a dead man's switch | 11:10 |
eudoxia | an* UFAI | 11:10 |
maaku | kanzure: genetic search + reinforcement learner + programmable goals + "mind transmission" communication, wrapped up with the bitcoin p2p protocol for peer discovery and relay | 11:10 |
kanzure | no i mean | 11:10 |
kanzure | what do you mean by you weren't joking? | 11:10 |
FourFire | kanzure, I'm afraid you'll have to remind me on "my explanations about what this channel is for?" then I could tell you whether I've just completely misinterpreted or ignored them | 11:10 |
kanzure | FourFire: you just claimed you thought this channel was for my outbursts, you dork | 11:10 |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 11:11 | |
maaku | eudoxia: I don't yet, but I will. | 11:11 |
kanzure | maaku: i think that the claim that singinst can prevent ufai is silly and should never be trusted | 11:12 |
maaku | kanzure: +1 on that | 11:12 |
kanzure | and even if they were capable of that, then we s[huld prevent them from achieving control of the planet | 11:12 |
kanzure | btw fenn has been living with them for a few weeks now | 11:12 |
FourFire | hehe, it's been a long time since someone called me a dork, do we still use that word these days? | 11:12 |
FourFire | I'm fully aware that there's some massive communication problem between the two of us, but I remain unconvinced that it is solely due to me being stupid | 11:13 |
maaku | yay we got an inside mole! | 11:14 |
AmbulatoryCortex | eh? | 11:14 |
maaku | i'm actually not too worried about the siginst people, but rather their crazy, literally insane followers | 11:15 |
maaku | goertzel has apparantly been receiving death threats too | 11:15 |
kanzure | maaku: are you in california? when you're not in puerto rico? | 11:15 |
maaku | yeah | 11:15 |
maaku | San Jose | 11:15 |
kanzure | you should meet feen | 11:15 |
kanzure | fenn | 11:15 |
maaku | i'd like to | 11:15 |
kanzure | and tell him to bring steve | 11:16 |
FourFire | "you just claimed you thought this channel was for my outbursts, you dork" kanzure I admit that I can't see where I did that, care, to quote it back to me? | 11:16 |
kanzure | wtf? | 11:16 |
maaku | who is steve? | 11:16 |
kanzure | 10:53 < FourFire> I had no idea, I thought you were just acting arseholish for no particular reason besides personality deficits | 11:16 |
kanzure | FourFire: you don't seem to be capable of carrying a conversation.... | 11:16 |
FourFire | yes, ok in retrospect that was a stupid comment | 11:16 |
kanzure | maaku: steve is the only reason singinst has an ounce of legitimacy | 11:17 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds] | 11:17 | |
maaku | i would have put nate/so8res as that reason | 11:17 |
maaku | (i've been trying to pull him away from MIRI craziness, unsuccessfully) | 11:17 |
FourFire | I know that you consider me your intellectual inferior, or perhaps just unfortunately undereducated, but I don't understand why that uatomatically translates into you being a verbal dick to me, however I'm glad we have this dialogue going now, so don't get offended now. | 11:17 |
kanzure | s08res not found | 11:17 |
maaku | FourFire: kanzure is a dick to everyone | 11:18 |
maaku | http://lesswrong.com/user/So8res/ | 11:18 |
FourFire | I genuinely want to find out what ticks you off so much about me, and fix it if it's not too hard | 11:18 |
kanzure | maaku: http://lesswrong.com/user/Steve_Rayhawk/overview/ | 11:19 |
maaku | FourFire: good. as far as I can tell that's why he's a dick. if you take at as indication to learn more about yourself, you're the right type of person. if not, you disappear and that is also good | 11:19 |
FourFire | thanks for that maaku | 11:21 |
eudoxia | not true, i haven't disappeared and all i've done is write a commit for skdb once | 11:22 |
FourFire | the potential benefit from attaining the cooperation of people who are insufficient in personality to get around kanzure's personality are not worth the time for the benefit they will give, which answers my first question today | 11:22 |
kanzure | because i have never seen you do anything but misunderstand | 11:23 |
kanzure | which makes investment in you totally worthless | 11:23 |
kanzure | also reddit. that's not a good sign either. | 11:23 |
maaku | FourFire: right, I think | 11:24 |
kanzure | *and* lesswrong, geeze | 11:24 |
kanzure | eudoxia: you are bad at hiding the other piles of code you write | 11:24 |
eudoxia | kanzure: haha, yes, but a lisp ORM isn't going to invent molecular machines ;C | 11:24 |
FourFire | oh yeah I wanted to write up some detailed things in the diyhpl.us/wiki filling in some bits which are marked "will fill in later" "obvious stuff" | 11:25 |
kanzure | stranger claims about lisp have been made, i suppose | 11:25 |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 11:25 | |
kanzure | maaku: steve is basically the singinst version of gmaxwell | 11:26 |
FourFire | kanzure, TBH most of the time I spend on reddit goes towards raising the, what LW would call "sanity waterline" but what is more correctly termed "effective altruist-transhumanist culture" in the futurology subreddit | 11:26 |
eudoxia | that's probably a waste of time | 11:27 |
FourFire | oh and evening out the collapse and darkfuturology commenters which are too pessimistic, yeah it probably is | 11:27 |
eudoxia | kanzure: here's a funny coincidence, though, there's a chemist who attended a foresight conference and built a common lisp compiler on LLVM to run massively distributed computational chemistry stuff | 11:28 |
FourFire | I guess I'll continue my previous strategy, and report when I've actually done something i consider valuable: if I do manage to make the BCI work, I'll open source it | 11:28 |
eudoxia | maybe i can convince him to, idk, run an ab-initio simulation of all those impossible gears | 11:28 |
kanzure | er, why? | 11:29 |
kanzure | convince him to clean up nanoengineer :p | 11:29 |
eudoxia | kanzure: he spent two years writing a compiler so he wouldn't have to touch C++ anymore | 11:29 |
kanzure | FourFire: i don't think you are going to find eeg useful | 11:31 |
FourFire | kanzure, thanks for the tip, what makes you say that? | 11:33 |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 264 seconds] | 11:33 | |
kanzure | decades of disappointment | 11:33 |
-!- Vutral_ [~ss@p5B2A49CB.dip0.t-ipconnect.de] has joined ##hplusroadmap | 11:33 | |
FourFire | I'm buying it for the hackerspace, and though it's not necessarily useful for transhumanist agenda, I'm sure someone here will do something cool with it even if it's insufficient for my needs | 11:34 |
FourFire | would you suggest something else i could spend 800 USD on then? | 11:34 |
eudoxia | inb4 molecular biology textbooks | 11:35 |
eudoxia | then again we have libgen for that | 11:35 |
kanzure | hopefully you guys would burn me at the stake if i suggested paying for textbooks | 11:35 |
-!- narwh4l [~michael@unaffiliated/thesnark] has joined ##hplusroadmap | 11:37 | |
kanzure | yo narwh4l | 11:38 |
narwh4l | hey kanzure | 11:38 |
FourFire | eudoxia, I can just borrow them from the university library anyway | 11:40 |
FourFire | I got a recentish copy of Molecular Biology of The Cell but have been too busy to read through metabolism :/ | 11:41 |
-!- narwh4l [~michael@unaffiliated/thesnark] has quit [Remote host closed the connection] | 11:42 | |
delinquentme | im . really . fucking . happy . | 11:48 |
delinquentme | kanzure, nmz787 thanks for the halp ... instrument response time is perceivably instant | 11:49 |
kanzure | got a paycheck? | 11:49 |
delinquentme | that too ! | 11:49 |
delinquentme | im not dirty fucking poor hahah | 11:49 |
kanzure | good times | 11:49 |
FourFire | So, kanzure what would you advise me in regards to studying Biology at a University, is it: A waste of time and you don't care || B not worth it but you'd tell me to anyway since I'm useless to you || C Worth it and you'd recommend it because it would help me help you || D I could make better use of my time reading textbooks and programming (or whatever you do) ? | 11:50 |
kanzure | do you have money? | 11:50 |
FourFire | I live in a country where it's free to attend | 11:50 |
FourFire | but no. | 11:51 |
kanzure | do you live in the united states? | 11:51 |
delinquentme | FourFire, my .02 : if you think you can make it through the typically boring a fucking curriculum and still love it ... do it | 11:52 |
delinquentme | I dont think its a stretch to say college lecture classes / testing structure are where creativity goes to die. | 11:52 |
FourFire | kanzure, Norway | 11:53 |
FourFire | delinquentme, I'm not certain, but between the hackerspace and work, which is actually mentally engaging and not soul crushing in the slightest, I could probably make it through five years required for a masters degree in cellular biology and molecular chemistry, which is what I'd go for | 11:57 |
TMA | FourFire: It is mainly the leisure to pursue the learning -- the college might or might not be helpful in your knowledge quest. | 11:57 |
FourFire | TMA, I'm afraid I couldn't parse your first sentence, care to rephrase? | 11:59 |
TMA | FourFire: Sure. It'll take some rethinking. | 12:00 |
TMA | FourFire: I would compare college to instant ramen or fastfood -- easily accessible knowledge in standardized packages. Passable but not quite gourmet nor healthy. | 12:02 |
FourFire | fair enough, I'd compare it to politically correct food | 12:04 |
TMA | FourFire: If your goal is to learn, then college might provide you with the basics. However in order to learn I think that it is more important to have the leisure to learn than any method -- college, self-study, MOOC, ... | 12:04 |
FourFire | you're 100% not weird according if you take university, and you automatically get some degree of respect from people who care about pieces of paper | 12:04 |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-xhnlxqlewjbnfxns] has quit [Quit: Connection closed for inactivity] | 12:04 | |
TMA | the paper -- it is the main bonus of college/university -- it boosts your earning power so you can have more leisure to learn later in life | 12:06 |
FourFire | yeah, fair enough: the goal is to acquire the resources to actually have an impact on indefinite biological lifespan research, or it's lesser subgoal, life extension | 12:06 |
TMA | I am not well versed in biology myself, but I take it that a good college is the fastest way to get the basics necessary for starting to make an impact. | 12:08 |
FourFire | if that means I need a degree in order to work for SENS then so be it, if I can be an autodidact and do the same more quickly (more doubtful, personally) then that. | 12:08 |
kanzure | biology is not a high-paying career | 12:08 |
FourFire | I have taken college biology, I'm talking solely about University | 12:08 |
delinquentme | FourFire, its worth mentioning that those who funnel in money from other industries are doing as much good as SENS | 12:08 |
delinquentme | and what kanzure said | 12:09 |
kanzure | if you enjoy being broke and poor then congratulations welcome to the glamorous life of biology | 12:09 |
FourFire | delinquentme, "other industries" don't capture my interest like biology does, and thus I predict I will be less productive | 12:09 |
delinquentme | FourFire, show me your passion by working a hard ass well paying job then finish up work and head to the lab | 12:10 |
FourFire | I don't care about glamour or wealth either, I care about the end results, even if my short term motivation system is counter-productive | 12:10 |
delinquentme | How passionate are you ? | 12:10 |
TMA | delinquentme: I have tried two universities. Only one of them was killing the creativity. | 12:12 |
FourFire | having an odd 60 years of "upper middle class" lifestyle is worthless against multiple millennia of meh (and you probably get a much higher variation too) | 12:12 |
delinquentme | FourFire, i dont follow what you mean | 12:14 |
delinquentme | TMA do research if its what you truly want. | 12:14 |
delinquentme | where im at I'm funneling my day job funds into research | 12:15 |
delinquentme | With money: the lever gets much larger. | 12:15 |
FourFire | delinquentme, enough that my judgement is clouded regarding Eliezer's rather unorthodox outbursts because he proclaims death to be the enemy | 12:16 |
FourFire | delinquentme, what research? your own? | 12:16 |
delinquentme | hehe He unfriended me because hes a social justice warrior | 12:18 |
delinquentme | FourFire, myself and a few friends | 12:18 |
-!- eudoxia [~eudoxia@r167-56-30-216.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 12:27 | |
FourFire | delinquentme, I meant that a "good life" is worth very little compared to a "poor unglamorous" indefinite lifespan, which could well contain 60 years of wellbeing anyway | 12:28 |
delinquentme | FourFire, I think you're missing what my point is | 12:34 |
delinquentme | you can spend your money however you like | 12:34 |
delinquentme | but just having it... will give you more options. | 12:35 |
FourFire | yes you're implying that I could multiply my research effectiveness by earning money and donating it | 12:35 |
FourFire | or that, yeah | 12:35 |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-ijbgrnynrfthyvxx] has joined ##hplusroadmap | 12:36 | |
FourFire | I have a job, it's not well paying for me atm, but it's in a high margin business, and will be in the last half of jobs to be automated | 12:36 |
TMA | not well paying job in high margin business sounds like a very bad method of participation on that business -- better to own that business | 12:40 |
FourFire | I'll get higher pay as soon as I stop being a trainee | 12:41 |
FourFire | ATM I get the crappy pay of 15.50 USD/Hour, but I get some perks in the job | 12:42 |
TMA | FourFire: then you'll get more of the leisure I was talking about earlier -- leisure to pursue your learning or your research | 12:43 |
FourFire | I am of the opinion that I lack the required personal qualities to own successful business | 12:45 |
FourFire | of course I haven't taken the opportunity to test this yet. | 12:45 |
FourFire | I don't see myself doing so in the forseeable future as it would be a risky investment of the majority of my resources for an uncertain probability of attaining an uncertain gain | 12:46 |
TMA | FourFire: based on my calculations and big mac index your salary would be slightly above average in the Czech Republic where I live; this translates to an elevated level of stress when trying to work and study with university imposed deadlines | 12:54 |
FourFire | so that's bad? | 12:54 |
TMA | FourFire: on the other hand you have said that your job is not of the soul-sucking variety; the overall level of stress might be easily comparable to one full time soul-sucking job with no other stressors | 12:56 |
TMA | FourFire: so it might be doable :) | 12:56 |
TMA | FourFire: the deadlines will force you to learn faster [and shallower] -- the price is stress, the gain is breadth of knowledge [I wouldn't have learned some things were they not compulsory part of the curriculum -- yet I find the insights obtained from them useful] | 13:01 |
TMA | maybe the best plan is to start, learn a little, network a lot and drop out -- several of the big enterpreneurs did just that | 13:02 |
-!- souljack [souljack@shell.xshellz.com] has quit [Remote host closed the connection] | 13:16 | |
-!- FourFire [FourFire@2a02:270:2015:cafe:c148:1a38:1e7e:49f3] has quit [Ping timeout: 276 seconds] | 13:22 | |
justanot1eruser | maaku: I was thinking about incentive problems with freicoin then it hit me that freicoin has a double meaning | 13:37 |
-!- FourFire [FourFire@2a02:270:2015:cafe:c148:1a38:1e7e:49f3] has joined ##hplusroadmap | 13:58 | |
FourFire | TMA, again, I feel that I lack certain personality ... requirements to "network a lot" (and I don't primarily mean that I'm a spotty nerd with no social skills) I feel that the majority of the gain of going to university would be the knowledge shortly followed by the certification of holding a degree | 14:00 |
kanzure | that's insane | 14:02 |
FourFire | I just don't think I'm entrepreneur material (and I can't even spell that word :( | 14:02 |
FourFire | kanzure, in which way, I see several | 14:02 |
-!- night [~Adifex@unaffiliated/adifex] has quit [Ping timeout: 265 seconds] | 14:05 | |
kanzure | you can just buy a certificate from lots of schools, try vietnam | 14:08 |
FourFire | but those certifications aren't worth much as a result | 14:13 |
FourFire | and having loads of random certifications just looks like you're collecting them, which in the eyes of the people I want to seem credible to, would be actually bad :/ | 14:14 |
-!- xrr [~xrr@gprs-inet-183-27.elisa.ee] has joined ##hplusroadmap | 14:16 | |
-!- FourFire [FourFire@2a02:270:2015:cafe:c148:1a38:1e7e:49f3] has quit [Quit: Leaving] | 14:28 | |
TMA | FourFire: nowadays I would put the degree and certificate first and the knowledge second [because the knowledge can be pursued by other means]. | 14:30 |
-!- drewbot_ [~cinch@ec2-107-22-1-5.compute-1.amazonaws.com] has joined ##hplusroadmap | 14:34 | |
-!- delinquentme [~delinquen@162-245-22-166.v250d.PUBLIC.monkeybrains.net] has quit [Remote host closed the connection] | 14:36 | |
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap | 14:39 | |
-!- drewbot [~cinch@ec2-54-159-143-104.compute-1.amazonaws.com] has quit [Ping timeout: 272 seconds] | 14:39 | |
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Namaste] | 14:44 | |
-!- Qfwfq [~WashIrvin@unaffiliated/washirving] has quit [Ping timeout: 245 seconds] | 14:54 | |
-!- Qfwfq [~WashIrvin@unaffiliated/washirving] has joined ##hplusroadmap | 14:59 | |
archels | so I think I'm going to go for CircuiTikz after all to typeset these circuit diagrams | 15:14 |
archels | InkScape sounded nice in theory because of the freedom, but there are way too many mouseclicks involved | 15:14 |
kanzure | bloop | 15:27 |
kanzure | well you could always write raw svg and load into inkscape? | 15:27 |
kanzure | although why use inkscape for anything circuit related? | 15:28 |
-!- ebowden [~ebowden@CPE-144-131-35-125.vic.bigpond.net.au] has joined ##hplusroadmap | 15:32 | |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Quit: Ping timeout: 245 seconds : USERID : UNIX : ByrdIRC] | 15:35 | |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 15:35 | |
archels | because of freedom in case I need a new part, or need to customise something | 15:37 |
-!- drewbot_ [~cinch@ec2-107-22-1-5.compute-1.amazonaws.com] has quit [Read error: Connection reset by peer] | 15:42 | |
-!- drewbot_ [~cinch@ec2-107-22-1-5.compute-1.amazonaws.com] has joined ##hplusroadmap | 15:43 | |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Quit: Ping timeout: 264 seconds : USERID : UNIX : ByrdIRC] | 15:48 | |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 15:49 | |
-!- delinquentme [~delinquen@162-245-22-166.v250d.PUBLIC.monkeybrains.net] has joined ##hplusroadmap | 16:09 | |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Ex-Chat] | 16:12 | |
-!- justanot1eruser is now known as justanotheruser | 16:20 | |
nmz787 | archels: why not use inkscape? | 16:31 |
nmz787 | archels: if you want to write some custom software, and are also interested in autorouting, maybe we can collaborate | 16:31 |
nmz787 | archels: s/inkscape/kicad/ | 16:31 |
nmz787 | my fingers typed from the wrong memory buffer | 16:32 |
-!- night [~Adifex@unaffiliated/adifex] has joined ##hplusroadmap | 16:34 | |
kanzure | logging log collector http://www.fluentd.org/ | 17:01 |
maaku | justanotheruser: double-meaning? | 17:05 |
-!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Ping timeout: 264 seconds] | 17:06 | |
maaku | .help .tell | 17:06 |
yoleaux | Relay a telegram to someone | 17:06 |
maaku | .tell maaku testing | 17:06 |
yoleaux | maaku: Talking to yourself is the first sign of madness. | 17:06 |
maaku | hah | 17:06 |
maaku | .tell FourFire fuck utility; take what you're interested in college. art history? philosophy? whatever. drop your job as the opportunity cost is too high and self-learn bio-engineering, comp sci, or whatever transhumanist contribution you want to make. establish a reputation by coding. | 17:10 |
yoleaux | maaku: I'll pass your message to FourFire. | 17:10 |
kanzure | i'm not really sure where to make money in norway | 17:11 |
justanotheruser | maaku: is there not supposed to be a double meaning? | 17:12 |
justanotheruser | "frei"? | 17:12 |
maaku | kanzure: he may have to move elsewhere after university | 17:14 |
maaku | it comes from 'freigeld', Gesell's proposed currency, which itself did have a double-meaning i think | 17:14 |
maaku | it's not a play on my name as many people think (different root, spelled differently) | 17:15 |
maaku | 'freigeld' does literally mean 'free money' | 17:15 |
kanzure | instead of university he could spend that time making money | 17:16 |
kanzure | "notmysurnamemarket" | 17:16 |
maaku | haha now that was a double meaning | 17:17 |
justanotheruser | ok neverming | 17:17 |
maaku | we were absolutely going for 'free markets' in 'freimarkets' | 17:17 |
maaku | justanotheruser: but seriously what was the two meanings you thought? i want to know in case there's a branding issue | 17:17 |
justanotheruser | maaku: well now theres threeish | 17:18 |
justanotheruser | I thought it was just german-free coin | 17:18 |
maaku | which it is, go on | 17:18 |
justanotheruser | then I realized it was made by mark frei- | 17:19 |
justanotheruser | but I assume its actually made by mark frie | 17:19 |
maaku | heh yeah. 'frie-' means peaceful, 'frei-' mean free | 17:20 |
maaku | that one's absolutely a coincidence. besides freigeld predating freicoin by about 100 years, jorge coined the 'freicoin' before i found the project | 17:21 |
maaku | btw what was the incentive issue? | 17:21 |
justanotheruser | while demmurage may be good for society as a whole, it is bad for an individual | 17:22 |
justanotheruser | I'm still skeptical that society can be efficient with a deflationary currency | 17:23 |
justanotheruser | or at least as efficient as it is now | 17:24 |
maaku | ah well that's a complicated one to answer in one sentence, but it comes down that in an economy based on demurrage currency real prices are lower | 17:25 |
justanotheruser | yep. | 17:25 |
maaku | due to effects on interest rates | 17:25 |
maaku | a longer discussion probably deserves to be on #freicoin | 17:25 |
justanotheruser | the incentive problem isn't so much something that breaks freicoin as something that hurts adoption | 17:26 |
kanzure | working capital accumulation and saving sounds like a useful thing to me | 17:26 |
kanzure | instead of financing crazy projects with crazy amounts of debt, you can just finance stuff by doing good work | 17:26 |
justanotheruser | kanzure: in practice it has been problematic | 17:26 |
kanzure | in what practice tho | 17:27 |
justanotheruser | I'm not sure how to answer that | 17:27 |
justanotheruser | but maybe the answer is an example of a country using a deflationary currency? | 17:27 |
kanzure | bitcoin is a first | 17:27 |
kanzure | there has been been an actually scarce money | 17:27 |
justanotheruser | not gold? | 17:27 |
justanotheruser | inb4 alchemy | 17:28 |
kanzure | nobody uses gold like that because it's inconvenient | 17:28 |
maaku | justanotheruser: right, once adopted the benefit should be obvious by comparison, but you don't get those benefits until a large part of the supply chain starts accepting free money (free by Gesell's definition) | 17:30 |
kanzure | maaku: i have some complaints about the costs of capital equipment | 17:31 |
justanotheruser | kanzure: there have been convenient gold based currencies | 17:31 |
maaku | kanzure: don't we all :) | 17:31 |
kanzure | that's just paper gold | 17:31 |
justanotheruser | sure | 17:31 |
kanzure | also the actual available supply of gold is not known, so why would anyone know if the total amount of paper gold adds up? | 17:31 |
justanotheruser | they wouldn't | 17:32 |
kanzure | fascinating | 17:32 |
justanotheruser | lol | 17:32 |
justanotheruser | the currency wasn't trustless | 17:32 |
kanzure | https://www.youtube.com/watch?v=cFods1KSWsQ | 17:32 |
maaku | justanotheruser: there is Woergl to point to | 17:34 |
-!- archels_ [charl@toad.stack.nl] has joined ##hplusroadmap | 17:35 | |
maaku | but i'm more interested in what you get when you combine freicoin scarce money with (old-)ripple credit (both tend towards 0% interest) and commodity basket units of account | 17:36 |
maaku | that's the direction we are slowly headed towards in freicoin | 17:36 |
maaku | (with freimarket extensions) | 17:36 |
-!- JayDugger1 [~jwdugger@pool-173-57-55-138.dllstx.fios.verizon.net] has joined ##hplusroadmap | 17:37 | |
-!- delinquentme_ [~delinquen@162-245-22-166.v250d.PUBLIC.monkeybrains.net] has joined ##hplusroadmap | 17:38 | |
-!- ParahSailin_ [~parahsail@unaffiliated/parahsailin] has joined ##hplusroadmap | 17:41 | |
-!- archels [charl@unaffiliated/archels] has quit [Write error: Connection reset by peer] | 17:43 | |
-!- delinquentme [~delinquen@162-245-22-166.v250d.PUBLIC.monkeybrains.net] has quit [Ping timeout: 256 seconds] | 17:43 | |
-!- JayDugger [~jwdugger@pool-173-57-55-138.dllstx.fios.verizon.net] has quit [Ping timeout: 256 seconds] | 17:43 | |
-!- cursive [~sqrt@unaffiliated/cursive] has quit [Ping timeout: 256 seconds] | 17:43 | |
-!- poohbear [~tigger@unaffiliated/tigger] has quit [Ping timeout: 256 seconds] | 17:43 | |
-!- faceface [~dbolser@nat8021.ebi.ac.uk] has quit [Ping timeout: 256 seconds] | 17:43 | |
-!- drazak [~bleh@198.52.199.197] has quit [Ping timeout: 256 seconds] | 17:43 | |
-!- andytoshi [~andytoshi@unaffiliated/andytoshi] has quit [Ping timeout: 256 seconds] | 17:43 | |
-!- ParahSailin [~parahsail@unaffiliated/parahsailin] has quit [Ping timeout: 256 seconds] | 17:43 | |
justanotheruser | but you can still study the economic effects of a deflationary currency | 17:43 |
-!- cursive [~sqrt@54.84.83.73] has joined ##hplusroadmap | 17:43 | |
-!- cursive [~sqrt@54.84.83.73] has quit [Client Quit] | 17:43 | |
-!- Netsplit *.net <-> *.split quits: dpk, Urchin, lichen | 17:43 | |
-!- faceface_ [~dbolser@nat8021.ebi.ac.uk] has joined ##hplusroadmap | 17:43 | |
-!- poohbear [~poohbear@unaffiliated/tigger] has joined ##hplusroadmap | 17:44 | |
justanotheruser | I don't think the economics change that much even when you have a trusted party controlling the currency | 17:44 |
-!- drazak_ [~bleh@198.52.199.197] has joined ##hplusroadmap | 17:44 | |
-!- andytoshi [~andytoshi@wpsoftware.net] has joined ##hplusroadmap | 17:44 | |
-!- Netsplit over, joins: lichen, dpk, Urchin | 17:48 | |
justanotheruser | maaku: so freimarkets is being developed for bitcoin and everything from bitcoin including sidechains etc will be moved to freicoin as soon as its ready? | 17:52 |
maaku | justanotheruser: once it is open sourced by blockstream, it can be ported. jorge and I have written into our employment contracts permission to do exactly this even if it were to pose a conflict of interest to blockstream | 17:59 |
maaku | (which it doesn't at this time, and I don't foresee it being a conflict, but just in case we insisted on that clause) | 17:59 |
maaku | jorge's full time job right now is basically implementing a confined-scope version of freimarkets | 17:59 |
justanotheruser | yes, I wouldn't expect it to be a short term conflict given the market caps no offense :x | 18:08 |
maaku | well even if freicoin completely overtook bitcoin (one can dream), it should affect BSC. we make our money on currency-neutral services | 18:14 |
maaku | *it should not | 18:44 |
-!- ebowden [~ebowden@CPE-144-131-35-125.vic.bigpond.net.au] has quit [Remote host closed the connection] | 19:08 | |
-!- cntrational [~sid@49.205.77.125] has joined ##hplusroadmap | 19:08 | |
-!- cntrational [~sid@49.205.77.125] has quit [Client Quit] | 19:09 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Read error: Connection reset by peer] | 19:28 | |
-!- ebowden [~ebowden@CPE-144-131-35-125.vic.bigpond.net.au] has joined ##hplusroadmap | 19:36 | |
-!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has quit [Read error: Connection reset by peer] | 19:56 | |
-!- delinquentme_ [~delinquen@162-245-22-166.v250d.PUBLIC.monkeybrains.net] has quit [Ping timeout: 255 seconds] | 20:12 | |
nmz787 | sheena: does the speed of dispensing for the cheese matter? i.e. slowly over 5 seconds versus a 'real quick squirt'? | 20:29 |
-!- cntrational [~sid@49.205.77.125] has joined ##hplusroadmap | 20:38 | |
-!- cntrational [~sid@49.205.77.125] has quit [Quit: WeeChat 1.2-dev] | 21:05 | |
nmz787 | sheena: this is what I have so far, it isn't much but I am still learning this CAD system http://www.3dvieweronline.com/share/Y9Ma7JmyO0wKZyY/Y9Ma7JmyO0wKZyY | 21:33 |
-!- phm4242 [~o@boole.london.hackspace.org.uk] has joined ##hplusroadmap | 21:39 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 21:51 | |
-!- phm4242 [~o@boole.london.hackspace.org.uk] has quit [Ping timeout: 276 seconds] | 21:58 | |
-!- phm4242 [~o@boole.london.hackspace.org.uk] has joined ##hplusroadmap | 22:02 | |
-!- soylentbomb [~k@unaffiliated/soylentbomb] has quit [Ping timeout: 272 seconds] | 22:29 | |
-!- xrr [~xrr@gprs-inet-183-27.elisa.ee] has quit [Ping timeout: 265 seconds] | 22:37 | |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-ijbgrnynrfthyvxx] has quit [Quit: Connection closed for inactivity] | 22:54 | |
JayDugger1 | Good morning, everyone. | 23:06 |
-!- ebowden [~ebowden@CPE-144-131-35-125.vic.bigpond.net.au] has quit [Remote host closed the connection] | 23:16 | |
--- Log closed Sun Feb 01 00:00:41 2015 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!