Bitcoin Forum

So-called lightning network “watchtowers” — a much-anticipated next step for securing the network — are coming soon.

Watchtowers have long been considered a missing piece of bitcoin’s lightning layer, as they’re crucial for squashing fraud on the experimental, off-chain network that could make bitcoin payments faster and more scalable.

Visit now at www.coindesk.com/fraud-fighting-watchtowers-are-coming-with-the-next-big-lightning-release for more information

This seems to be a good move but the problem is that they are solving a problem that does not exist yet. Most part of Bitcoin users never had a deal with LN. It is not even implemented anywhere (considering that we need LN for "offline" payments).
So in this case I guess that no one really cares about the security of their money on the system they don't use.

LN not implemented anywhere??Are you kidding me?
It's true that most bitcoin users never used LN,but with increased adoption of btc,offchain solutions like LN will increase their popularity as well.The "Watchtowers" are a good move for the LN dev team.

Some traders are having a second thought of trading in LN because of the risk of being scammed. If LN would implement this so-called "watchtowers"  I'm sure that it will change the perspective of some traders. It will also be an advantage if it could make transaction even faster and better. I hope that other exchanges would also focus on the development of their security system.

LN works have been going on with the utmost privacy since the very beginning! I haven't liked this privacy since the beginning. Therefore, I feel confident that there will be a problem.

I found a relatively good "ELI12"-style explanation of watchtowers (https://www.reddit.com/r/Bitcoin/comments/94u9se/how_do_the_watchtowers_work_in_the_lightning/). Overall, watchtowers seem to be an interesting concept and do not put in danger the decentralized model of LN. (I don't really like the name, though - I already see tinfoil hat rage incoming)

What's to be determined is the incentive to run watchtowers, as they seem to be data-intensive. The linked article seems to imply that Lightning Labs is considering to create a kind of private altcoin (they call it an "Ecash (https://en.wikipedia.org/wiki/Ecash)" variant) to pay watchtowers. If this "coin" is distributed via an ICO or something similar then I don't know if I like that ("Ecash" seems to be a pretty centralized concept involving banks ... again, fodder for tinfoil hat guys).

Then when it comes to this it is just another way of promoting their alt-coin,
And making sure that there would be a demand for their crypto .
But if it would make the Bitcoin transaction much more faster then it would be great.

Fraud is everywhere and it's good to know that there's a service being created to stop it. It's another step to lessen the risk of crypto transactions and to implement the security to get rid of scams. If watchtowers could make instant transactions, I think it's also the best concept for other platforms to apply.

If I understand the concept correctly the technology allows for someone to monitor and identify "scam" transactions, while you are not there to identify them? The drawback for me about the LN is that you have to constantly over see the whole process and that it needs to be online 24/7. <Most of us sleep for 7 to 8 hours in a day>  :P

The on-chain Blockchain transactions does not need 24/7 babysitting, so I would much rather opt for a system that can run on it's own.  ::)

But that's exactly what watchtowers are trying to solve.

I'll try to re-phrase the linked Reddit post (https://www.reddit.com/r/Bitcoin/comments/94u9se/how_do_the_watchtowers_work_in_the_lightning/) to make it even more simple:

• As a "normal Lightning user", you may not want to be online 24/7.
• So you can subscribe to a "watchtower" - or to the number of watchtowers you want. Any Bitcoin user that's online 24/7 (e.g. using a VPS) can be a "watchtower". They don't need to run Lightning nodes, probably they don't even need a full node, so an electrum instance on a VPS should be enough for the watchtower software.
• Every time you update your channels with an "update transaction" (which is not broadcasted to the blockchain), you do the following:
  
  • You send half of the Transaction ID to the watchtower
  • You encode the transaction which contains the "punishment" for an offending channel counterparty, which returns all the funds of the channel to you, using the other half of the transaction as the key and send it as a "punishment BLOB" to the watchtower, too.
• The watchtower continuously observes the Bitcoin mempool
• If it sees a transaction that matches the half Transaction ID you gave it, it begins its "punishment operation":
  
  • It takes the other half of the complete transaction ID it found on the mempool. If your channel counterparty has really broadcasted one of the channel states, for which you sent half of the transaction ID to the watchtower, then the other half is the key for the "punishment BLOB".
  • It decodes the punishment BLOB and broadcasts this transaction to the blockchain. All the funds are returned to you.

Maybe you can even subscribe to all existing wachtowers via a broadcasting mechanism, which wouldn't be a bad idea, as also watchtowers can in theory go offline.[/list]

I'm going to be unfashionable here and say that nothing I've read or come across so far has convinced me lightning networks are anything other than an interesting wee curio.

It's been said about countless other world beating things so I'll keep that under my hat until proven wrong.

watch towers are needed, because LN as just a 2 counterparty agreement just wont work.

if LN worked perfectly between 2 people. watchtowers wouldnt be needed. so watchtowers is big red neon sign that something is up

secondly. the onchain locked funds should actually be locked. thus broadcasting anything randomly day or night should get rejected right up until the lock is unlocked. watch towers subtly reveals this aint the case either.
EG if you put a 90 day lock on funds on the bitcoin network. you SHOULD live in peace for 89 days,23hours 50minutes without having to care what your counterpart might get upto.


meaning bitcoin utxo's used for LN are not as locked as you might think and that too opens up many attack avenues.

havnt people yet cottoned on that LN is and wont be as secure as bitcoin

and the last joke of the article.
even with LN payments. routing and onions. LN does not have a private way to pay a user on LN.. namely a watchtower. so now LN devs want to make a whole new token...

once people read the unicorn fluffy utopian promotion. have a coffee. sit back and then read things over properly. they start to see the real things behind the buzzwords

It's not that LN doesn't work, it's that the security model requires users to be online 24/7. Watchtowers are a solution to this obvious UX problem.


That wouldn't work very well. If all LN funds had different timelocks, it would destroy fungibility. Who wants funds that are stuck in LN for the next 90 days? These wouldn't be worth 1:1 to Bitcoin because they can't be used on-chain.

You literally disagreed with me and then assumed that your statement is wrong lol. If LN is not implemented anywhere and no one is using it then yes, it is not implemented anywhere.
The general picture is like this: BTC is an unpopular payment method, LN is only useful for offline transactions in different shops and none of them is using LN. So most people don't have to care about the LN fixes.
The success of LN is determined by the amount of people that are using it for their daily microtransactions and not by those huge success in their developement.

calling the flaw of needing to be online and needing to do certain things a 'security model' .. nah its a flaw
as for the 90 days. i personally think 30 is 'normal' as many live paycheque to paycheque and only plan/budget a month at a time
its others that love ln who keep thinking 90 in their scenarios. so i was just going with their flow

but that said. in a secure payment system. you shouldnt need middlemen, supervisees, security guards...
.. that was the whole point of bitcoin

having a system that without a middleman you have to worry about what happens when you sleep.. is not a good system

LN motto: "where sleeping makes you poor, unless you can afford a babysitter"

So Who is Watching the Watchers?  :D

http://www.marvunapp.com/Appendix6/uatu_watches-jla.jpg

nobody said payment channels were perfect. decentralization is hard. every solution has its trade-offs. the threat of watchtowers punishing fraud is a pretty damn strong incentive against defrauding LN users. is it as strong as bitcoin's on-chain security guarantees? no but can it be exponentially cheaper for users, which is pretty great.

your solution is some sort of socialist paradise where users don't need to pay the cost of their on-chain transactions. everything is free and miners don't need fee income. well sure, if long term viability doesn't matter, why should anyone pay for anything!?

back in the real world, people are trying to find off-chain (read: low cost) solutions for users that are actually long term viable.


and you can use bitcoin. but nobody said it would be free. pay the cost or use something cheap (like LN).

if you don't like watchtowers, put an LN node on a raspberry pi. It's easy. Problem solved

but wait. wasnt the whole point of LN because blockchains are too big for home users.
the mindset these days are blockchains are for "big blockers"

we have had 'banking' business models for decades. crypto/blockchains meant to be taking us away from middlemen not re-introducing them

but wait. wasnt the whole point of LN because blockchains are too big for home users.
but wait. wasnt the whole point of LN to just use smart phone to buy coffee in starbucks
now things are circling back to the whole 'home users need to be paid to be their own full node' greedy cries

the mindset these days are blockchains are for "big blockers"

we have had 'banking' business models for decades. crypto/blockchains meant to be taking us away from middlemen not re-introducing them


bitcoin is only getting more expensive. not due to technology. but due to human(dev) implemented limits under the guise of economic game theory.

LN cheap?
having a channel direct with a merchant(1hop) ~cheap. but routing =multiple fees. then add on watchtowers fee. then add on factories fee.
also people dont do 20tx a day normally. they do 1-2tx a day(even popular payment methods like visa show this). so if the guise of economic game theory is pushing onchain fee's up to match the (doomsday rhetoric game theory need for fee war) of no blockreward. then even 90tx over 3 months still would end up costing maybe $66*
crypto is becoming a system of needing to in the future pay like $22 a month just to use it.
which is still more than an hours labour in 3rd world countries per tx

*blockreward= ~$8k*12.5=100k...
  to compare pool income of future....100k/3000tx a block=$33 a tx..... open session+close session=$66

..
again i am not saying "free"tx's. but when payment systems start charging more than an hours labour for a 3rd world country. most of the ethoses and goals of blockchain currencies are lost. especially when they start needing middlemen to manage/counter-sign peoples funds.. it kinda starts defeating the point of the original inventions glamour, novelty, solution and hope

Although this look good but their function in the LN was much anticipated for and for them coming into the system, it will be better. But i had of their intention to create their own coin which will be use to pay them for the service rendering? wil this coin be decentralize or centralize? Hope things are done right as they come into the crypto-space.   

why need own coin specific for certain users
isnt using the pegged Msats(already utilised in LN) good enough?.. start wondering why is Msats not sufficient to pay watchtowers