Topic: The Barry Silbert segwit2x agreement with >80% miner support. - page 23. (Read 120029 times)

It is correct to say LN does not scale bitcoin as it only creates centralization for micropayments at the second layer. But it also allows atomic cross chain transactions.

Right, pretty much... There's not really any such thing as a backwards compatible hardfork, only some SPV clients might not validate enough information that they get hijacked.

Read as:
The "backwards compatible softfork" that is designed to be followed by a "backwards compatible hardfork" isn't backwards compatible with the old clients that will no longer work after the fork anyway.
 

Here Jeff Garzik is explaining how the requirement for a >1MB block worked exactly as predicted saying it's not a problem at all...

https://lists.linuxfoundation.org/pipermail/bitcoin-segwit2x/2017-July/000094.html

Peter Todd asks the valid question of why the hard fork bit wasn't used.

https://lists.linuxfoundation.org/pipermail/bitcoin-segwit2x/2017-July/000098.html

Important quote

And basically the question is shrugged off and he is referred back to the git pull request citing tl/dr:

https://lists.linuxfoundation.org/pipermail/bitcoin-segwit2x/2017-July/000096.html

Bingo. For the past year, altcoins have been used precisely for all of LN's stated use cases - that's why the altcoin market cap has exploded. Many more people trust and use altcoins. Due to the network effect, many of those who accepted bitcoin for payment last year now accept altcoins this year. Investors see growth potential in alts now that bitcoin has stagnated. All of this is due to the scaling roadblock. Vitalik wakes up each day and thanks the Core devs for supressing growth of bitcoin.

Altcoins are absolutely superior to LN in every way: No theoretical new second-layer model is needed. LN is basically pointless. But Blockstream has a $75 million bet on LN.

The only problem with using altcoins is that often there is a dependency on notoriously-untrustworthy exchanges...

Q 14.1: A standard bitcoin Tx is dependent on confirmations in the blockchain… So, is it really fair to claim that a Lightning Tx is the same as a normal bitcoin Tx?

A:

This is a valid point, they are not the same…

A Lightning Tx is a zero-confirmation Tx. But if it is broadcasted to the bitcoin network; it will be just as valid as any “on-chain” zero-confirmation Tx.

Both types of Tx will eventually be mined into the bitcoin blockchain if they pay a sufficient fee.

However, a LN-Tx has a different security model that makes it much more reliable when compared to a standard zero-confirmation Tx.

A Lightning Tx is only indirectly secured by Proof of Work. This is due to fact that a Lightning Network will be completely dependent on the underlying bitcoin network (see Q12)

Within an open Lightning channel; there is a different set of game-theoretical mechanisms that provide a different type of security model.

Lightning will extend the capabilities of bitcoin without the need for a trusted third party.

But the tradeoff is that you must monitor the bitcoin network by the operation of a full-node.

This monitoring can be outsourced, but in that case you must trust an external server to actually do its job. Your money will still not be routed through this server. The only role of the server is to monitor the bitcoin network, and to broadcast a so-called Penalty Transaction when necessary.

Note that the use of this service is an option, in case you don’t want to run your own full-node.

It will not be possible for this third party to steal money from a Lightning channel.

Also note that the LN is intended as a platform for low-value-transfer (sub $100)

All LN transactions are multi-signature and both participants in a channel must sign for a Tx to become valid. A traditional double-spending attack is therefore made extremely difficult.

However, there is a risk that someone can broadcast an obsolete Lightning Tx to the bitcoin network.

An obsolete Lightning Tx is a Tx that does not represent the latest state of its channel.

The above mentioned risk is the reason that you (or a service that you trust) must operate a “Watcher Node”.

This node will monitor all the transactions that are broadcasted to the bitcoin network.

If your Watcher Node discovers an obsolete Tx; it will (as a countermeasure) broadcast a “Penalty Transaction”

The Penalty Transaction gives you the power to confiscate all the money within your channel (including the money that belongs to your counterpart)

However, a penalty Tx can only become valid after the discovery of a broadcasted obsolete Tx.

Your ability to broadcast a Penalty Transaction makes it very risky for your counterpart to broadcast an obsolete Tx.

Another security/privacy feature, is that all Lightning Tx will be end-to-end encrypted between the participants.

***********

The above in red needs more research. "The Penalty Transaction gives you the power to confiscate all the money within your channel (including the money that belongs to your counterpart)" That doesn't look good. This looks like a backdoor for government to come in and steal. Scammers to take advantage of. “Watcher Node” - looks like a centralise entity and isn't trustless.

For anyone still confused about my stance on LN:

• I am in favor of LN for what it is
• Bitcoin is part of LN
• LN is not part of Bitcoin
• An analogous example of the above would be: "all squares are rectangles, but not all rectangles are squares".
• Since LN is not part of Bitcoin, LN has no useful place in the conversation of Bitcoin protocol changes
• Since LN is not part of Bitcoin, LN has no useful place in the conversation of Bitcoin scaling
• Since LN is not part of Bitcoin, saying that LN is "part of a scaling solution for Bitcoin" is saying "the solution for scaling Bitcoin is to not use Bitcoin"

Tx again - so think about all the possible attacks that are still unknown (I piss at ANY code reading and testnets - that's NO production proof / even SW in LTC is a proof of no-use)) ? How many options you had attacking Bitcoin over its first 1-3 years?

We should really better get back 'conservative' and get Bitcoin stronger and risk-free with highest reputation than ever!  Scale the thing up to the sky and pay the miners for our all security -  or scam them and you pay later by loosing your coins.

And if you still think you need new toys -> get scammed in the altcoin sections!


Listen to expert Matt Corallo at 35Min to a question of a friend of mine about the security issues - dodgy...

https://www.youtube.com/watch?v=GMvWDg9dtcw

Can anybody say : "The DAO"   

Older LN channel state is cryptographically valid the same way as newer state. The only incentive not to trying to add older LN channel state to Bitcoin blockchain is the other party with newer state can punish it by getting all such channel coins for themselves - but only if the tansaction can be added to Bitcoin blockchain in timelly manner (you can set the time how long your fine to have stuck your LN coins when opening the channel - this corresponds to the time available to react for having confirmed tansaction in Bitcoin blockchain should this scenario happen). This has theoretical attack vector possible with big LN hubs and congested Bitcoin - like with exchange hacks, it is only matter of time when such attack become reality one day.

Well, I'm a simple fellow on all this. Just an observation...not that it is gonna effect this agreement or anything else....

The most telling thing to me in all this ...is

1) the spam and all the full block stuff went away as soon as consensus was reached on seg witness being done anyway by the 2mb camp (non-core) that
    flies even if the 2mb becomes an issue in the future after August. If this had never happened, their would have been no incentive for doing anything
    other than just implementing seg witness promptly ..and work on this stuff later (as supposedly is agreed to now ..again imho I think that is how it is setup)

2) who has incentive to FORCE this issue with the spam attacks that now mysteriously disappeared...what is the agenda (I'll not speculate on this just tossing
    it out there as weird as all hell) .......was it just sandbox control of the dump truck by a bunch of 2-year-olds..that is the end result of 2 years of cluster?

3) Ego and control..no matter how you look at this in hindsight or how you go about seeing this whole thing since the failed Hong Kong agreement.....nobody involved
    looks in any way or shape 'clean' after this cluster....everyone seems to want to toy for themselves and screw the community.

Just something that should be mentioned as a "was it really just a play for control" the whole charade of high transaction fees and slow confirm times....
just a lever to make some kind of ego protecting 'stand' with seg witness the starting point..and a jump off point for more clusterf'ing around in the future?

anyway, just looking at it from that point of view, it makes me sad as hell, we have drifted that far down, since the attempt at the Hong Kong agreement...we
are a long way imho on both sides from principle and logic determining progress....and I fear it will become worse the next 'go around' end of this fall 2017.


 

I was not aware of that. These kinds of fine technical details are why I have trust that eventually some smart person will sort it out. Is it a core requirement of how LN functions that any time state of the channel can be included in the Bitcoin blockchain or is it just because LN is not yet fully developed. Hopefully a clever solution can be found to fix this flaw. You are right though, it is a niche technology for now and not likely to be touched by Joe bitcoin user. My gut feeling tells me that btc will not be competing with other payment systems on a sheer transaction volume capability basis; at least not for a long time to come. I think most average people / investors who are aware of bitcoin see it as a sort of digital gold, and nothing more.

And therein lies the main problems:

• You read what you want to read, not what's on the screen in front of you.

• You misread the multiple times that I said I'm all for LN (as it's intended to be used).
• You can't grasp that this thread is about Bitcoin.
• You can't grasp that, in a conversation about Bitcoin, things that are not the Bitcoin protocol don't matter to the Bitcoin protocol.
• You can't grasp that LN isn't part of Bitcoin, LN has no relevance to the Bitcoin protocol, and LN has no useful place in this conversation.

Thx - but for the small brainers (< 1MB) you should drop some link why....

LN flaw is any older state of your channel can be added to Bitcoin blockchain. Your only defence is try to include the latest state of your channel to Bitcoin blockchain in timelly manner, but you cant guarantee your onchain transaction going to be included to Bitcoin blockchain before is too late. LN might find its niche, but its too risky and complicated for average Joe. Bigger blocks would make using LN actually safer.

Thing is we are running Bitcoin since 2009 now and lots' of people not even trust in this (and created 2k of altcoins to try out on that as well) - and we have limited space and resources to invest on the Bitcoin trust & scale.

Why should we not really concentrate on that very thing =on-chain ?

Diluting trust and resources to (not on-chain) things bankers and scammers can do way better and is NOT scope of pure Bitcoin ?

Tell people to read 2n order solutions, NOT Bitcoin?

Test fancy redundant networks, NOT Bitcoin?

Getting scared about sell chills, NOT Bitcoin?

Putting finally Bitcoin reputation ON RISK ?

- Sorry - I like the Satoshi Game & Nash equilibrium and second the code and try to understand this better FIRST Place. And miners = big users might do same.  Second place see FIRST.

Yes. However if a trader is doing arbitrage trading between two exchanges, why use LN?

Lets say i want to sent BTC from exch A to exch B, btc takes so long. I just look at another coin like Stellar and if safe and profitable, buy STR, send to exch B, convert to BTC, make the usual profit. Stellar already does the job securely and very fast, less than 30 seconds; so why would anyone use LN?

I don't have to trust sales arguments because smarter people than me can read the code and find flaws. Besides the system can be tested before a single coin is ever risked. It's all open source, read the code and make your own mind if you should or should not trust it.

  Yeah this dogmatism I see from some people is pretty amusing. It's a mix of stubbornness and self interest me thinks. Jihan likely just cares about Bitmain's bottom line not necessarily about new and innovative technologies. They fell into a very lucrative niche as a virtue of an accident of geography and they intend to squeeze this bitcoin thing for all its worth lol .

Well, I guess 20 years down the road they can have their 1 GB blocks cause we'll all be using holographic crystal "hard drives" and we'll all have 1 TB fibre connections. This kind of thinking is akin to coming across a mountain in your way and instead of finding a way to avoid it you just smash your head against its cliffs with ever increasing power in the hope of tunneling through. Elegant and innovative solutions to problems are always more beautiful than brute force solutions.

I could be wrong about this but this idea probably stems from a misunderstanding of how LN works. If they wanted to attack the LN nodes they'd have to attack the BTC blockchain itself.