--- Log opened Mon Apr 15 10:58:38 2013
10:58 !wolfe.freenode.net [freenode-info] if you're at a conference and other people are having trouble connecting, please mention it to staff: http://freenode.net/faq.shtml#gettinghelp
19:21 < BlueMatt> so my research advisor commented that people dont want to do research on bitcoin because it is "the biggest ponzi scheme in history" :(
19:21 < BlueMatt> ofc I didnt have time to correct him then, so I had to put it in an email :(
19:24 <@gmaxwell> But it is the worlds first ponzi scheme based on CRYPTOGRAPHY!
19:24 <@gmaxwell> I usually find that it's best to embrace the skepticism.  "Sure it's crazy and going to lose people money— but look how _interesting it is_"  .. a week later they're asking me how to buy bitcoins.
19:25 < BlueMatt> all I had time to do (lecture had already started) was say that "ehh, the system is freaking awesome, even if you dont care about the currency"
19:25 <@gmaxwell> yea, thats the general approach.
19:26 <@gmaxwell> Once someone has grasped the idea its often hard to talk them _out_ of buying into it. Convincing technical people that bitcoin isn't a scam is not a limiting factor in my expirence.
19:26 < BlueMatt> problem is, I highly, highly doubt he has time to read into the system
19:27 < BlueMatt> hes "director of undergraduate studies" so...
19:28 < petertodd> indeed, I spend most of my time talking to my coworkers about how many serious problems bitcoin has
19:28 < BlueMatt> what production system doesnt?
19:28 < petertodd> and the other day I had two of them asking me what's the best ASIC company to buy mining equipment from...
19:28 < warren> I already told this story (here?).  2.5 weeks ago a MBA student asked if I had bitcoins to sell him.  I refused to sell to him and told him to do his own risk/reward research. This guy doesn't know how to use his own computer.
19:29 < BlueMatt> heh
19:30 < petertodd> ouch...
19:30 < warren> perhaps interest in ASIC's will be lower now.
19:30 < petertodd> maybe, although I will say, one of those co-workers wanted to buy his 17 year old son an ASIC miner, because he figured he'd either learn about crypto, or business, specifically risk
19:31 < BlueMatt> heh, thats a fun gify
19:31 < BlueMatt> t
19:31 < petertodd> I support repeatable life-lessons :)
19:31 < BlueMatt> he should have given him a bitcoin a week ago
19:31 < petertodd> ha, for sure
19:35 < petertodd> so, wizard topic: I'm thinking it might be worthwhile to make the mother of all alt-coin scams, er, I mean trusted ledgers. basically addresses in this scheme would look like 1AGUP2MVtTqMHysb9gS3upedZmSvaV1f5h@3CK4fEwbMP7heJarmU4eqA3sMbVJyEnU3V and the logic is pretty much as you would expect "pay to the first bit, and the transaction is guaranteed and recorded in some ledger by the second address"
19:35 < petertodd> the advantage being you can re-use all the existing transaction infrustructure, in particular, existing and future hardware wallets and similar things
19:36 < petertodd> the logic also becomes nice: pay to the person holding these public keys, but guarantee the transaction using whomever is running the ledger service with the identity this set of pubkeys
19:36 < petertodd> s/pubkey/seckeys/
19:36 <@gmaxwell> so basically the 3CK4fEwbMP7heJarmU4eqA3sMbVJyEnU3V identifies the ledger in question?
19:37 < petertodd> exactly, how you determine how to communicate with that is another matter, but it's that signature that is important
19:37 < petertodd> (hence my post on the -dev list about multisig signmessage)
19:38 <@gmaxwell> I'd worry about cross ledger signature rebinding in that.
19:38 < petertodd> what do you mean exactly?
19:39 <@gmaxwell> E.g. you make a transaction X in ledger 1  and I replay that transaction in ledger 2.
19:41 < petertodd> right, yeah, I'm thinking the ledgers should act like blockchains, which means when coins enter the ledger, apply something like HMAC(on-chian-txid, ledger-identity) to the txid's to ensure that they don't make sense in the context of another ledger.
19:44 < petertodd> also, from a ui perspective, you need to ensure the user can determine where coins are being sent, so I'd just take the ugly hack of interleaving txouts the user wants, with txouts specifying what ledger (or the blockcahin itself) each txout is going to, heck, those txouts can contain the fees...
19:45 < petertodd> basically, the philosophy being leverage existing software as much as possible for v0.1
19:46 <@gmaxwell> I think it's too abstract for me to say anything more.
19:46 < petertodd> well, sounds like it's not inherently a bad idea, so worth a prototype
19:46 < petertodd> what's nifty about it, is a core bit of the trust would be the exact same merkle-sum utxo tree that Bitcoin itself might have one day
--- Log closed Mon Apr 15 21:30:49 2013