In real world use case I don't think average users will want to create a new channel for every single time they trade with someone new. It will be a hassle. They will want to just use 'the lightning network'. Hub and spoke will arise naturally as most efficient and fast way of routing through the network.
So there you have: If most people use LN for their small transactions/payments more block space there is for the "important" transactions if you are willing to pay the fee. You can choose which method you use. What's bad about that?
It's not that you need to chose one way or another, you will be able to decide for each transaction you decide to make.
The bit I don't like is everyone else using 'central' hubs to put their transactions through.
Where do you see the problem of mining centralisation when everyone can just buy their own miner?
Because they don't have an incentive to go through all that trouble.
Yes that illustrates my point LN users wont be incentivised to go to all that trouble either.
Thanks, at least you guys are giving me some food for thought. Mainly responses have been trollish before.
It's not that users wont be incentivised to go to... Which trouble? Standard BTC transactions will be exactly the same trouble as they are now, nothing changes. But yes, users will be incentivised to use LN for several reasons:
- Lower fees
- Almost INSTANT transactions <- This is a really important point. No blocksize increase could ever reduce a minimum of 10 mins for the first confirmation.
- Real scalability: No more TPS limit. LN networks could process thousands, maybe even millions of transactions per second.
The future usage I envision for myself is something like this:
- For moving my BTC in non trivial ammounts -> BTC Blockchain
- For buying "coffee" or who knows what other smallish stuff -> LN
- For things between those two use cases -> LN or BTC Blockchain depending on the circumstances.
I am not saying that a Block size increase won't be needed in the future. In fact I am sure it will. But what we need first is Segwit. After that, and if it is possible to achieve an enormous consense (95%) for a hardfork to increase the blocksize, then I am ok with that too... a fixed blocksize increase, not a joke like BU.