about the utility of LN in regards to peoples false ideals of the fee's they expect.
anyone also notice how blockchain.info removed the "days destroyed" stat.
this is because if you analyse that stat against other stats, it reveals that people DONT on average do several transactions a day or week.
many peoples funds are used once every 5 days. some peoples funds sit for months and yes years. thus something like LN with a 10day n-lock is useless utility for these people who do 2 or less tx per channel period.
so dont expect "cheap" tx fee's as the norm because LN isnt designed for people that only do 2 or less tx's per LN channel period.
next we have the CLTV which when confirmed(settled) stops funds from being respent for the CLTV period (laymans: much like banks/paypals 3-5business day delay on ''funds available", or like bitcoins blockrewards 100confirm maturity)
so people who do desire 'regular' spending wont want to have funds maturing for a few days.
however, LN have a scheme/plan for that too.
an LN hub can circumvent the CLTV maturity by revoking the funds to a destination in the CSV
revoke commitment to push the funds into a new LN channel. meaning an LN hub can lock peoples funds permanently into LN contracts.(laymens: bank chargeback/reclaim)
LN actually initially wanted to penalise users by taking all their funds to the LN hubs non multisig address(fraud penalty).. as the ultimate penalty.. but morally they should only use the CLTV and CSV features to reset a new LN contract* to then get the customer to agree a new honourable agreement of who owes what. suggesting that due to CLTV its better to just use LN than resettle again.
*https://github.com/bitcoin/bips/blob/master/bip-0112.mediawiki#lightning-network
To allow a commitment transaction to be effectively revoked, Alice and Bob have slightly different versions of the latest commitment transaction. In Alice's version, any outputs in the commitment transaction that pay Alice also include a forced delay, and an alternative branch that allows Bob to spend the output if he knows that transaction's revocation code.
so LN hubs can actually hold funds to ransom and even if you broadcast your own copy, they can CSV revoke to honourable put funds back into a new LN agreement.. or maliciously allow the hub owner to take the funds into his own private ownership (non multisig).
yes i left the best for last because i know people are itching to say they will save on costs if they need to do more than 2 transactions per channel period...
lastly there are the internal penalty fee's too.
LN have come up with many penalties. such as, if there is a multihop transaction happening.
you can be penalised for not promptly agreeing to it, as it would cause a extra delay on the other commitments agreements.
you can be penalised for not signing your agreement
you can be penalised for attempting to spam/DDoS the hub with excess data
you can be penalised for sending the wrong agreements to be signed by the hub
they actually envisioned a 0.006btc fee as a minimum entry into a channel ($4-$7 at this weeks volatile price) to cover their expectant fee's over time. yet again rules out micro payments of developing countries where $4 is 2weeks wage (cuba, georgia, etc).. $4 is 1 week(Kyrgyzstan, gambia, ethiopia, etc)
imagine having to deposit a week or 2 wage just to cover fee's for a 10 day lock. before you factor in how much you would deposit ontop to actually spend.
the problem with the devs, both calculating the onchain fee war and the LN fee penalty acceptability. is they are thinking with a western world mindset of being only slightly cheaper than paypals 20-30cent fee. before the devs would even consider lowering their 'spam' limiter down. they should spend a week outside their basements and travel the world and experience currency usage outside of the corporate america setting.
overall summary
LN is not the ultimate scaling solution. it is and should be just a niche/side voluntary service much like third party exchanges are.. for things like gamblers, faucet raiders and day traders who usually spam the blockchain frequently.
however malicious spammer who want to spam the blockchain will avoid LN because LN defeats their malicious purpose. thus LN wont mitigate intentional spammers.
also infrequent bitcoin users (under a few times a week) wont need/want LN due to the settlement delays and the costs of setting up/settling would be more then just paying someone direct onchain infrequently
