Topic: So we tried out Lightning tonight at ROOM77 (Read 2568 times)

In my opinion, that's just the use case I would also use LN for: small transactions in real-life or brick-and-mortar businesses. So I encourage these experiments. When I'm again in Berlin maybe I'll try it out in your bar - hopefully with real coins and Segwit activated

For larger transactions, like my salary or a remittance, I wouldn't like to use LN. In these cases I would prefer other second-layer solutions like extension blocks and side/drivechains.

They're going to have to fill more than 1 block. Unless they ALL try and cash out ON the block the time lock expires.. They won't..


Until SegWit is activated - BTC ain't going nowhere.. and LN too.. so I'm assuming this all happens after that.

That an oldie.   
https://www.youtube.com/watch?v=iNLXxDMxe18



 ,
Let's Say 100 other users Time Locks are expiring at the same time as yours , and all drawing out at least $100.
Total amount is now 100users*$100=$10,000
Average # of transactions per block ~2000.
At $1 each , it cost me $2,000 bucks to spam the block locking out all of your transactions and including mine, that lets me steal the entire $10,000 .
So My Profit is $8,000  , not bad for less than 30 minutes worth of work.




Until segwit is activated, you have to trust them.  (Even LN devs warn you of this.)

If segwit is ever activated,
And a Hacker Hacks an LN Node,
he could have it Lie to you about the route or fees or maybe even corrupting the preimage , triggering a roll back of the payment.
http://bitfury.com/content/5-white-papers-research/whitepaper_flare_an_approach_to_routing_in_lightning_network_7_7_2016.pdf

Slickest thing for an Hacker to do is Trick the Neighborhood Discovery
to forward the preimage to himself so he can steal the funds directly.
By updating one LN, Network Discovery Paths, you update all of them.
http://bitfury.com/content/5-white-papers-research/whitepaper_flare_an_approach_to_routing_in_lightning_network_7_7_2016.pdf
Page 15 , goes on about Network Discovery

 

Ok Kiklo - let's dance.


This I agree with  


As easy as mounting a 51% attack !? yeah - real easy.. (And this attack applies to EVERYTHING crypto-coin based - not just LN)


Nah.. Let's say I want to cash out $100. All I need to do is pay a fee higher than the LOWEST the spammer pays.. Sooo.. If I'm prepared to pay $1, the spammer would have to spam the network, with fees OVER $1 for 100's / 1000's of TXNS to prevent mine from getting on.. no way is that worth it.


Nope.. AT NO STAGE am I having to TRUST the LN node to sign anything on my behalf. All the clever crypto-shit is done by the client, on his machine, and he certainly isn't going to sign a txn that has 'funny' timelock schedules.. The worst the hackers can do is force me to cash out, if the LN node starts spewing rubbish.

The Lightning Shit Network is unnecessary. It has four elements that it claims will solve problems with Bitcoin.

Instant payments without waiting for long confirmation times = my bitcoin transactions have always shown up instantly on the network. For security purposes you have to wait for multiple confirmations. So what, Bitcoin works best for international money transfer (buying online long distance). If you want to buy a beer or a cup of coffee use the ACH/EFT system or cash. Both are instant and have successfully worked for decades. If you're buying drugs on a street corner use cash. If you're buying drugs online use Bitcoin, before the dealer can package and ship your order the btc transfer will have 6 confirmations.

It's immediately capable of scaling to huge demand = bitcoin doesn't have a huge demand yet. It will be decades before bitcoin is overcome by demand and that can be corrected within bitcoin. There's no need for an outside fix. Bitcoin should never be used to grocery shop, buy a soda or a candy bar. That's not where its most useful. Bitcoin will never replace every payment system in the world. It certainly doesn't need to try to do it within the first decade of its life. Let's pretend bitcoin is only seven years old and will not be required to replace ACH/EFT tomorrow. We have decades to work on this problem.

LN is so cheap to use that we can start allowing dust payments again, yippee = this is just goofy. Why would anyone care if you can buy a piece of buttered toast or tip someone a penny with bitcoin. If you really feel the need to buy one slice of buttered toast with crypto then go start butteredtoastcoin and use that. Make a copy of a script coin like Litecoin and buy your toast and tip pennies all fucking day long. Leave bitcoin out of it. Stop being so fucking cheap and pay a fee. Up until now people have been paying $25-$50 for international wire transfers that took a week. Pay your fucking dollar to get same hour service.

Hurrah, cross-chain atomic swaps are possible with two coins that use the same cryptographic hash function. With Lightning Shit Network dogiedickcoin, titcoin, Litecoin, creamcheesecoin and bitcoin can all be lovers. =  I don't want bitcoin to have any strange lovers and catch HIV from some perverted blockchain. Leave bitcoin alone.

If that is real that the clients does not need any more to wait for confirmation then that is a very good news to all of us bitcoin users. It maybe time for the merchants to come in and accept bitcoin without having to worry slow confirmation of transactions which very unpleasant. But the users also hope that with this automatic confirmation the fees will not sting us and instead enjoy lower fees compared to the fees at this moment.

Thanks for quoting my explanation of how LN works. The more people understand LN, the less they fear it.
Now LN is being tested on the testnet. You can check your claims there. The more attacks you try - the better, testnet exists exactly for that. Probably you can even convince people to conduct a 51% attack.

LN are not transacting actual BTC, get that thru your head.
LN Notes are a representation of a BTC or a LTC or whatever segwit infected altcoin blockchain they lock up the coins on.
LN is nothing more than a Locking/Unlocking mechanism which gives 1 or 3 parties the right to move portions of it once the time locks are up.
Example: If those time locks expired due to a hack or network spamming, when you go to redeem your LN note, it will Fail.
Then your so called LN Notes prove to have been worthless.

Offchain Verification can not be trusted,
If they let you check the Onchain BTC address the actual BTC is locked up in, you can use a block explorer to confirm they are onchain.
This would inform you if someone 51% attacked and counterfeit LN, but you have to know the Onchain BTC address.
(Does LN even show the Onchain BTC Address so you can check it?)
But if someone evades the time locks so that the BTC can be stolen before you can redeem it, then above check was useless.
It is also useless if the 51% attack occurs after you checked the BTC address.

 

FYI:
LN security is not secured by the Coin's Blockchain,
LN security is nothing more than a series of complicated Time Locks.

This is the only argument against LN. Such study would be very difficult to do, too few people would trust it. Because too many parameters are involved. If we can't reliably measure all those parameters, it's garbage in - garbage out study. People are different. Some will be OK transacting via an excessively big hub. Some won't be doing it, some will be running small low fee hubs on principle.
My opinion is that entry barrier is very low, so there will be enough hubs.

putting LN like this make it sound like sidechain isn't it? sidechin are also worthless until the proper amount of bitcoin are locked, for that particular chain and you cna use that value on that chain now

are they abandoned the initial idea of sidechain for LN? what i don't like about the LN is the centalized HUB, they want to get rid of possible centralization that major block would bring but then they want to go with LN...

Even those which which are not part of the hub's channels?
There are no LN notes. LN will be transacting BTC.

Verification is very easy for channel participants. It's as easy as checking an address in a blockexplorer today.

Say Bob and Carol have a channel with current funds distribution Bob - 3BTC, Carol - 2BTC. Let's represent this as
Bob-3:2-Carol
Alice wants to pay Carol 1 BTC, but opens a channel with Bob (he plans to build a hub, charges no fee for now), she loads 1 BTC to the channel. So now our small piece of LN looks like:
Alice-1:0-Bob-3:2-Carol
There are 2 channels: Alice-Bob with 1 BTC in it and Bob-Carol with 5 BTC. Bob owns 3 BTC: 0 in his channel with Alice and 3 in his channel with Carol. Alice owns 1 BTC, Carol - 2 BTC.

Now Alice pays 1 BTC to Carol via Bob:
Alice-0:1-Bob-2:3-Carol
As expected, Bob still owns 3 BTC: 1 in his channel with Alice and 2 in his channel with Carol. Alice couldn't pay more than 1 BTC. But now she can use her channel with Bob for receiving up to 1 BTC. For the whole lifetime of a channel, this channel's BTC never leave it.
If now, due to a 51% attack, Alice-Bob channel is no more, this directly affects only Bob, he looses 1 BTC, Alice gains it and can spend again, effectively performing a doublespend, just like with good old onchain transactions. Carol isn't affected, she has all info about her channel with Bob. Bob's problems aren't her problems.

@joecooin, you never answered my question.



 

No one really has done a study on whether LN will be practical to do as a true P2P network  Vs. only big hubs/spoke-and-wheel/paypal2.0

LN's Code is nothing but a fancy series of Time Locking scripts.  

An LN Hub can place a time lock on any Segwit Infected Blockchain coins.
Meaning if BTC & LTC vote yes for segwit (doubt it) , you would be able to lock BTC or LTC in place on their respective chains.

LN is nothing but some code that lets you time lock the coins Onchain so they can not be moved.
LN Notes are like Bank Notes, completely worthless unless someone is willing to trade you something for it.
LN transfers happen Offchain , so their is no verification by the actual blockchain, that what you are trading for even exists, until you cash out.
When you cash out , then and only then do you find out if your coins were counterfeit or stolen before you tried to remove them.
Until that point you are trading Fake Notes, because they only prove to be real notes, if their is a payout of the value it is representing. 


 

Except lightning will not be working at its full capacity if segwit isnt put in place and that wont happen without a fork because BU wants to make a point rather than actually use good tech.

OK, we are getting somewhere.
First of all, I'd like to note, that when I mention LN without explicitly stating which version, I mean segwit LN.
Secondly, with CLTV and CSV (already adopted on Bitcoin mainnet), permanent channels are possible even without malleability fix: https://bitcointalksearch.org/topic/permanent-lightning-channel-without-malleability-fix-1711460 Their opening is a bit cumbersome and involves temporarily locking additional (to what is going to be put in a channel) funds, but after all, permanent channels are possible even without fixing malleability.

LN isn't a separate blockchain. LN BTC are real, same BTC as BTC you are transacting today. Say Alice and Bob have a bidirectional channel. For outside world the channel looks like a 2 of 2 multisignature output. Both participants of the channel, as well as other interested parties, can very easily check existence of that output. If the output is gone, so is the channel with all it's funds.

Hard for you & me , yes it is.

But the rich guys that can collude to 51% attack or the rich guys that can spam the network until your timelocks expire, it is easy as pie.
* The only poor people able to pull it off , are the Hackers that learn to Hack the LN Nodes, and modify the timelocks ahead of schedule.*

 

Elaboration


LN Does Needs Segwit so that LN can be trust less.
Otherwise LN requires a separate Trust system in place or very long extensive Time Locks , Both of which LN Devs don't want.

https://www.reddit.com/r/Bitcoin/comments/5eqm2b/can_ln_work_without_segwit/?st=izzovrzk&sh=b2fe8b0a

In Short , without Segwit any version of LN is going to be crap.

 

That is true for any 'refundable' time locked txn, like Atomic Cross Chain Transfers.

If I can stop you from claiming the coins before the time lock expires, i can refund them back to myself (after I have taken your coins aswell that is).. Just have to prevent your TXN from getting on chain.

This is much harder to do than I think you are implying.

Just quoted this for posterity. In the case the author later decides to delete/edit this reply.

I'm trying to look at LN from your perspective. Care to elaborate? More about timelocks? How an attack can fund a channel?