Author

Topic: Thoughts on the early bugs discovered by Peter Todd in LN? (Read 395 times)

RNC
newbie
Activity: 42
Merit: 0
I'm "trying to watch" Paul Sztorc's overview and misconceptions on drivechains, https://www.youtube.com/watch?v=gUbGT70wy5k

I started to watch the link in an effort to learn something new but the guy just goes on and on and we nearly got to the point were
he wanted to say that drive-chain could bring about world peace before I hit the stop button.

Shame because I am looking for new concepts and I suspect this was about hashing private ledgers and putting the SHA256
in the main block-chain but i could not bear to listen anymore so maybe you would like to spell it out for me in under fifty lines
of text because I like KISS as do most developers unless they are cutting code.
RNC
newbie
Activity: 42
Merit: 0
I am not a technical person nor an expert

Then you probably won't attract many technically skilled people to comment before you amend the thread title to something less sensational. Most of Peter's gripes are with specific implementations, and his only issue with the protocol itself is version specific (not to mention just a general worry, Peter Todd has not found any actual attack vector)

Not everyone here is out to collect merits like you seem to be and he raises some good points but you keep
getting away with being rude towards people and when they go back out you they are kicked out the forum.

Team leader Mr P Noon on the LN project won't even answer his emails and each hub is a single machine
so how centralized would you like to get with the new design because professionals like myself would never
build something like this and we are being silenced here.
legendary
Activity: 2898
Merit: 1823
Quote
As for the Lightning protocol, I'm willing to predict it'll prove to be vulnerable to DoS attacks in it's current incarnation, both at the P2P and blockchain level.

While bad politics, focusing on centralized hub-and-spoke payment channels first would have been much simpler.
I think that if The Lightning Network will be enough decentralized, the DoS attacks will be ineffective. The focusing on the huge "banking-hubs" is not a properly way for decentralized Bitcoin.

But having a hub and spoke model would be more efficient, especially in the beginning. I hope all the hubs would open channels and connect with each other though.

Quote
Anyway, the "sidechain" technologies are needed right now to decrease the current Bitcoin-transaction count which are made, for example, by the big crypto exchange markets, and as a result to reduce transaction fees. Now they can create their own channels for payments. I agree that it is necessary to develop a federated sidechains, The Fantom Network and other solutions outside of the blockchain of Bitcoin. The Lightning Network is in the first stage and is growing yet.

I'm "trying to watch" Paul Sztorc's overview and misconceptions on drivechains, https://www.youtube.com/watch?v=gUbGT70wy5k

Maybe that's something the more technical people here would like to watch and talk about as well.
legendary
Activity: 2618
Merit: 2304
Quote
As for the Lightning protocol, I'm willing to predict it'll prove to be vulnerable to DoS attacks in it's current incarnation, both at the P2P and blockchain level.

While bad politics, focusing on centralized hub-and-spoke payment channels first would have been much simpler.
I think that if The Lightning Network will be enough decentralized, the DoS attacks will be ineffective. The focusing on the huge "banking-hubs" is not a properly way for decentralized Bitcoin.

Anyway, the "sidechain" technologies are needed right now to decrease the current Bitcoin-transaction count which are made, for example, by the big crypto exchange markets, and as a result to reduce transaction fees. Now they can create their own channels for payments. I agree that it is necessary to develop a federated sidechains, The Fantom Network and other solutions outside of the blockchain of Bitcoin. The Lightning Network is in the first stage and is growing yet.
jr. member
Activity: 56
Merit: 115
Lowest EVER interest lending! (Use escrow always)
All good technologies suck ass in their former stages.
legendary
Activity: 2898
Merit: 1823
I am not a technical person nor an expert

Then you probably won't attract many technically skilled people to comment before you amend the thread title to something less sensational.

I was not trying to make it look "sensational". It was how I felt at the time I made the thread.

Quote
Most of Peter's gripes are with specific implementations, and his only issue with the protocol itself is version specific (not to mention just a general worry, Peter Todd has not found any actual attack vector)

It is still disappointing though, as excited as I am on the Lightning Network.

But thinking about it more, and after reading theymos' post, maybe JonaldFyookball was right. Some centralization might happen as a consequence as users would more likely open payment channels to services than their peers.
administrator
Activity: 5222
Merit: 13032
The minor software bugs will eventually get worked out. It's still new software. C isn't the greatest choice of language, but it can be made to work with enough persistence.

I agree that it is in some ways unfortunate that most people are focusing entirely on LN rather than other methods of fast & cheap off-chain payments like hub-and-spoke, federated sidechains, or blinded bearer certificates. While LN is the most decentralized off-chain solution that's been proposed AFAIK, it's also by far the most complicated. Even though you can sort of use it now, I'm not so sure that we're going to be seeing ordinary people using it regularly in the next year. And even if we do, it'll only be to pay established merchants, since no solution has yet been devised for LN payments where the recipient is offline; in practice, this will prevent peer-to-peer LN payments (eg. forum trades) from being practical.

LN will probably be the long-term solution to Bitcoin payments, but in the short-term I've long advocated for federated sidechains, which could be an ultra-simple drop-in replacement for on-chain Bitcoin transactions, or blinded bearer certificates, which immediately provide perfect anonymity as a side benefit. Both of them are much simpler than LN on a theoretical level, though both have the cost vs LN of increased centralization. I was really hoping that RSK would provide a federated sidechain usable for fast & cheap payments in this way, but so far I've been disappointed there. And nobody is working on blinded bearer certificates AFAIK.

(But I'm not working on any of this stuff, so I can't complain too much...)
legendary
Activity: 3514
Merit: 1963
Leading Crypto Sports Betting & Casino Platform
Not all criticism is bad. Take for example these two comments :

1. I think Bitcoin sucks, because the fees are too high. <Negative criticism without value>

2. I think Bitcoin can be more popular, if they can reduce the fees. <Constructive criticism with value>

If I analyse his comments, based on the two examples above, then I would categorize them as being constructive criticism. Peter's opinion highlights certain issues, which can be addressed, before the shit hits the fan. We are in the testing phase and people are encouraged to report on issues that might be exploited in the future.  
legendary
Activity: 3430
Merit: 3080
I am not a technical person nor an expert

Then you probably won't attract many technically skilled people to comment before you amend the thread title to something less sensational. Most of Peter's gripes are with specific implementations, and his only issue with the protocol itself is version specific (not to mention just a general worry, Peter Todd has not found any actual attack vector)
legendary
Activity: 2898
Merit: 1823
Making this topic for the purpose of starting a discussion among the more technically skilled people of Bitcoin in the forum. I am not a technical person nor an expert but I would like to listen to all your thoughts and understand them with all of my very limited "brain" power. Hahaha.

https://twitter.com/peterktodd/status/968190530294337538

Quote
Initial impressions of Lightning on testnet: c-lightning segfaults a lot, and when it's not crashing payments fail more often than not. Writing it in C - a notoriously dangerous language - doesn't strike me as a good idea.

Quote
The Android wallet Eclair is equally unreliable, and I got it into a state where it crashes on startup, effectively losing funds.

Quote
Haven't tried lnd yet, although the decision to use an alt implementation backend (btcd) is obviously a bad idea given how critical consensus compatibility is for Lightning.

Quote
As for the Lightning protocol, I'm willing to predict it'll prove to be vulnerable to DoS attacks in it's current incarnation, both at the P2P and blockchain level.

While bad politics, focusing on centralized hub-and-spoke payment channels first would have been much simpler.
Jump to: