Post
Topic
Board Development & Technical Discussion
Re: Wondering out loud: Which should Chinese miners support - Core, Classic or another?
by
gmaxwell
on 28/01/2016, 10:53:34 UTC
In terms of revenue, you've mentioned private side chains, I assume either by subscription or fee? Some would infer that a somewhat expensive and crowded main chain environment might incubate such services, for transactions generally, not just those seeking confidentiality. Why are they wrong?

Gregory, I'm not trying to diminish the work you've done, which is much more than I will ever do. Some people just have some questions, and I sincerely appreciate your responses.
And by support and development contract.  Private sidechains can offer functionality that a global decentralized system cannot directly for fundamental reasons-- e.g. instantaneous transactions, not just thing they don't offer yet (like CT or smarter smart contracts); they're just a different trade-off.  

Regardless of what you think of crowding-- an increase in the blocksize does not guarantee its avoidance, as miners are free to impose further restrictions; and a single person with a while loop could produce unbounded amounts of 'load'. Even if I don't expect them to do so, or at least not often, they can and that's always a risk at any (plausible) size.

The kinds of block size increase that even people aggressively in factor of block size increases believe might be viable in the Bitcoin network are fairly modest, e.g. classic and "2 MB", which, as mentioned is pretty close to the proposed segwit capacity.  By contrast, a private system used between businesses, could run arbitrarily large assuming all its participants were willing and able to invest in the cluster computers and expensive bandwidth required at _every_ node. Oh the benefits of not being a permission-less decentralized system.  The space of possible systems which don't require the other benefits of a private system, don't require the benefits of a decentralized public system, and would not work in bitcoin now but would with the kind of realistically small blocksize increase, and would choose to do so.. is pretty narrow-- and isn't something that we've ever considered in business discussions.  More philosophically, the bitcoin blockchain is a precious global shared public resource with huge externalized costs that fall to all future users; it's good stewardship to not cram things that don't need to be in it, into it regardless of what the limits are.

For added color: I've never heard a prospective customer say something like "X TPS won't work but 8*x TPS will work"; they do say things like "X TPS works now, but we might need 100000*X TPS on short notice, can we have that with absolute confidence if we're willing to pour money at it?"  And that latter question can never be true when your only mechanism is dumping all your data into a in a worldwide _shared_ decentralized flooding system run by third parties whos costs you don't pay. So if then also we also avoid huge businesses setting themselves up for failure with the expectation of the 100000 fold increase peak loads that the system couldn't possibly take without a decentralization killing blocksize-bailout, I think that is good too.

But also, hold up a minute. I think we're both playing along with Mike Hearn's claim that it's all _me_ saying, or the techies (or blockstream) alone, saying that HF's are dangerous in principle  and/or that block-size matters. It's not (go check out those bitcoinocracy links-- or Jon Matonis, for an example)... and it's also not our choice (except as people who own Bitcoin, and in our individual capacity to decide what efforts we'll volunteer time on). I get targeted because I've stepped up and drawn fire so that other people can get some work done.

I'm sorry for taking things so off-topic here. This is a tangent, and I wouldn't have gone down it. But people from Bitcoin Core who were on a phone call with some Chinese miners last week told me that some of these claims about Blockstream came up multiple times.  Even if it took a "happily biased" poster here to call them out, I think some people were thinking about them and I think it's better to have addressed them head on. I hope we can talk some about actual business needs in later posts.