This is a brilliant analysis of the Lightning Network.
It is great at scaling the number of transactions but it can only achieve this for a limited number of people.
LN is like a prepaid debit card. Funds are always online in a hot wallet. How many funds would you be comfortable to luck up? For me, whatever I am comfortable keeping in my regular wallet. But not 6 months worth of expenses (how many people even have that?)
Read for yourself
http://codesuppository.blogspot.com/201 ... t.html?m=1
So in summary LN is a brilliant innovation. It allows a micropayment type of transaction on top of the bitcoin protocol that was not possible before. It will increase the number of on chain transactions because of it. It is a lot of great things but it is not a scaling solution.
We need bigger blocks in order to get the most out of Lightning.
It would be best to keep only a weeks worth of expenses in a non-cold wallet.
I agree. But then LN does not really add all that much to scaling.
It is a great innovation, and certainly has its place in the bitcoin ecosystem. It may allow the transfer of small value for low fees. I could pay my coffee with it. But I would probably choose an on chain transaction for my rent.
But if we accept that people would use LN with 1 weeks of cash tied up in it, we must conclude that LN will not contribute to scaling bitcoin in any significant way.
If we also accept that LN facilitates a new kind of transaction (affordable micropayments) and that this would result in transactions between people that would normally not have happened, then we must conclude that there are on chain transactions created to facilitate these micropayment channels.
Thus while LN will make using bitcoin more valuable, it will not fix the scaling issue.
This means that both LN and SegWit are great additions to Bitcoin. I am grateful that they have been invented. But neither should be mistaken for something that they are not. Neither should be given qualities that they do not have. Neither are a scaling solution. One is scaling in nr of tx not nr of users, and the other is an accounting trick that actually increases the total block size more than full on chain transactions would.