23:32 <Taek> Do people in this channel often consider altcoins as a method for scalability? Most altcoins have transaction fees that are very much below the transaction fees of Bitcoin

23:32 <luke-jr> speaking for myself, no.

23:32 <Taek> and, while the security margins are definitely lower, if you are placing $200 into an altcoin for small transactions, is it really a big deal?

23:32 <kanzure> altcoins inherit the same problems

23:32 <kanzure> and/or create new problems for themselves

23:33 <Taek> kanzure: sure, but if your altcoin of choice is different from my altcoin of choice, we've basically expanded the chain substantially

23:33 <Taek> *

23:33 <kanzure> and for altcoins that have different security models, it's already known that different security models can be much more highly scalable than bitcoin

23:33 <Taek> *I am assuming there's an easy way to do cross chain payments. Right now you'd need to use an exchange like shapeshift

23:33 <kanzure> insert snappy comment about DMT here

23:35 <Taek> luke-jr: you would be opposed to using a non-Bitcoin currency to make your small payments?

23:36 <Taek> let's even assume that the software is a full clone of Bitcoin, but with a different PoW algorithm + genesis block, fewer users, lower fees.

23:37 <Taek> and maybe a smaller max blocksize, since the assumption is that it would live alongside Bitcoin as a method for cheaply making transactions which don't require as much security

23:38 <Taek> I guess the full assumption is that there would be many such chains

23:38 <Taek> and each person would be on 1-2 of them

23:39 <Taek> and this way you could get "on-chain" scalability

23:40 <Taek> To trade with eachother, you would use something like cross-chain lightning

23:40 <Taek> going through nodes that are on both chains

23:40 <Taek> (sender's preferred chain and receiver's preferred chain)

23:40 <midnightmagic> not the best place to be talking about that, really.

23:41 <kanzure> cross-chain lightning was already proposed quite a while ago

23:41 <gmaxwell> Taek: I consider them less of a scalability tool than microtransaction services. Two primary reasons: you take exchange rate risk which is less predictable than fraud by a microtransaction server, and the other is that they're generally subject to the same scalablity problems as Bitcoin more or less. Same reason I generally say that sidechains aren't themselves much of a scalablity tool (even th

23:41 <gmaxwell> ough they eliminate the exchange rate risk).

23:41 <kanzure> Taek: https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-September/010909.html

23:42 <gmaxwell> One argument that they (both altcoins and potentially sidechains) are a scalablity tool is that they can hit different points of the scale vs decenteralization trade-off.

23:42 <gmaxwell> E.g. you could exchange into insecure coin for small payments, but keep savings in Bitcoin.

23:43 <gmaxwell> But I'm not very convinced by that argument, since if you're willing to accept insecure coin, you'd probably be happy with a microtransaction server which could have a more clear security model AND better scalability.

23:43 <Taek> yes, that's exactly the advantage I'm proposing - they allow you make a tradeoff further along on the security/usability gradient

23:43 <Taek> if you don't require full Bitcoin-grade security, you should not need to pay full Bitcoin-grade fees

23:43 <kanzure> yes, mostly it's a slippery slope to off-chain centralization land (--- which isn't a bad thing necessarily, some users have different preferences for their money)

23:44 <kanzure> there's not much of a gradient on the altcoin side though

23:45 <gmaxwell> Taek: right but part of the problem is "security risk distribution". Blockchains have a generally unfavorable risk spectrum. I'm not sure that there really is such a thing as a kind of secure blockchain. There are secure blockchains and blockchains that are insecure but you don't know it yet. Same goes for things like microtransaction servers, but with things like remote attest and provable

23:45 <gmaxwell> solvency I think the risks of something like that are easier to measure than a blockchain's risk.. So if your goal is only kinda-secure a blockchain may not be the best tool. But I guess I would say that I wouldn't argue much with someone who disagreed with me on this point.

23:51 <Taek> You can measure the amount of work burying a given transaction, surely that counts for something. Even if a 51% attack is only a small bribe away re-mining the chain at all had a measurable cost

23:51 <Taek> *has

23:52 <Taek> I'm not sure what else you might mean by "not secure but you don't know it", I can only think of the 51% attack right now, given the same codebase that is