Hacker News new | ask | show | jobs
by kbody 3064 days ago
Wow, I guess it's dead then /s

It's an evolving software still in pre-alpha state. But the core idea is solid and if you follow the development you'll see the strides of progress that is being done on all fronts.

To your points.

1. Depending on the options you set to your channels, you could have a window of days (or anything the two parties agree to), to act in case the other party tries to broadcast outdated TXs i.e. steal funds. Also you don't need to have a computer constantly online, you can offload the task to Watchtowers, software dedicated to do that for multiple users concurrently. (You also can have multiple Watchtowers as backups). Also, I imagine it could even be a mobile app. So this is a non-issue.

2. Routing will keep being improved but it's already pretty good and the way you present it screams FUD.

3. Currently crazy low fees aside, LN basically enables the multiplication of on-chain TXs population. I think your assumption is that the retailer will close and reopen a channel for every LN payment they receive which it makes no sense whatsoever from any point of view.

4. If by huge you mean $100 of which the fee is $0.5, sure. You can add any amount you want on an LN channel. Basically it's like a hot-wallet, e.x you can add $1500 for $0.5 fee, and casually spend it for coffees and other small expenses with lower TX fees than a satoshi (since LN supports fees at the millisatoshi = 1/1000 satoshi level).

P.S: This FUD-wave is pretty pitiful, at least get some real arguments. I hope even more people start to recognize that a lot of people have incentives to attack new tech in order to push their narratives for their altcoins or w/e.

2 comments

> It's an evolving software still in pre-alpha state

Now imagine a start-up in this state was hocking shares to college students and grandmas at a $140 billion valuation.

I was talking about Lightning Network nodes, not Bitcoin. Bitcoin at one point was at this stage but through iterations and continuous improvement it evolved to being able to secure all those funds.

The point is that it's disingenuous to deem a new tech dead, while the engineers working on it sill say that it's not yet fully-ready for production.

The implementation and the protocol still evolves, but the core idea behind it is rock solid and the most promising way forward to help with scaling AND enable use-cases that would be otherwise impossible (see microtransactions & realtime millisecond scale monetary transactions etc.)

> I was talking about Lightning Network nodes, not Bitcoin

The latter being useless as a currency, per the article, without the former.

I fail to see how this moves the discussion forward, we're talking about how effective LN could be regarding helping with scaling.

Also, note the core idea of LN (updatable non-broadcasted Bitcoin TXs with incentive-based security and eventual broadcast to the blockchain network) is a powerful idea and as we explore this scheme and more engineers get accustomed I am sure we will see even more exciting things.

Do you happen to know where I can find the code or white paper for the current, working, routing algorithm?
Sure.

- Onion Routing Protocol: https://github.com/lightningnetwork/lightning-rfc/blob/maste...

- P2P Node and Channel Discovery https://github.com/lightningnetwork/lightning-rfc/blob/maste...

This repo contains specs for a lot of parts of the LN, although probably there may be changes I imagine.

As for code, one of the implementations, LND (Go): https://github.com/lightningnetwork/lnd/tree/master/routing but there are at least 3 other implementations in the works.

https://github.com/ElementsProject/lightning (clang) https://github.com/ACINQ/eclair (Scala) https://github.com/mit-dci/lit (Go)