SEGWIT & LN KILLING OFF the OnChain Miners (Better start looking for new Jobs)

[kiklo]

crypto is about freedom and freedom of choice... is it not !

Funny , you claim freedom , for crypto

But in the same post want to censor my freedom of speech , because it differs from yours,
and sorry to put it so bluntly, I am speaking the truth whereas you are marketing lies. 


Segwit & LN are very very bad for the crypto world and must be stopped.


 

[1715388408]

1715388408

[1715388408]

1715388408

[IadixDev]

My fear with LN is rather the opposite: that propagating "waves of panic" will overwhelm the block chain with transactions, because the amount of transactions pending on the LN network can in principle be orders of magnitude larger than what a block chain can handle (that's its main idea !).  So if a block chain can handle, say, 100 000 transactions per hour, and the LN network has 10 million transactions pending in 10 minutes, and there's a panic wave going through the network, those 10 million transactions will need to go on-chain which will create a backlog of 100 hours, often passing the safety time limit of regularisation, and huge opportunities to scam.

Your fears are confirmed , article from  Jul 5, 201612:28 PM EST by Kyle Torpey
https://bitcoinmagazine.com/articles/here-s-how-bitcoin-s-lightning-network-could-fail-1467736127/

And also what would happen in this scenario if the locks on the main chain expire before the tx from LN can be pushed back to the main chain, it would be a bit like double spending issue no ?

[kiklo]

My fear with LN is rather the opposite: that propagating "waves of panic" will overwhelm the block chain with transactions, because the amount of transactions pending on the LN network can in principle be orders of magnitude larger than what a block chain can handle (that's its main idea !).  So if a block chain can handle, say, 100 000 transactions per hour, and the LN network has 10 million transactions pending in 10 minutes, and there's a panic wave going through the network, those 10 million transactions will need to go on-chain which will create a backlog of 100 hours, often passing the safety time limit of regularisation, and huge opportunities to scam.

Your fears are confirmed , article from  Jul 5, 201612:28 PM EST by Kyle Torpey
https://bitcoinmagazine.com/articles/here-s-how-bitcoin-s-lightning-network-could-fail-1467736127/

And also what would happen in this scenario if the locks on the main chain expire before the tx from LN can be pushed back to the main chain, it would be a bit like double spending issue no ?

@IadixDev,
Nice,  you see the problems with LN also. 


How to Steal LN Funds from the LN WhitePaper itself.

https://lightning.network/lightning-network-paper.pdf
Page 49 thru 51 

Improper Timelocks
Participants must choose timelocks with sucient amounts of time.  If insuf-
 cient time is given, it is possible that timelocked transactions believed to
be invalid will become valid, enabling coin theft by the counterparty.  There
is a trade-o  between longer timelocks and the time-value of money.  When
writing wallet and Lightning Network application software, it is necessary
to ensure that sucient time is given and users are able to have their trans-
actions enter into the blockchain when interacting with non-cooperative or
malicious channel counterparties

9.2    Forced Expiration Spam
Forced expiration of many transactions may be the greatest systemic risk
when using the Lightning Network.  If a malicious participant creates many
channels and forces them all to expire at once, these may overwhelm block
data capacity, forcing expiration and broadcast to the blockchain.  The re-
sult  would  be  mass  spam  on  the  bitcoin  network.   The  spam  may  delay
transactions to the point where other locktimed transactions become valid

9.3    Coin Theft via Cracking
As parties must be online and using private keys to sign, there is a possibility
that, if the computer where the private keys are stored is compromised, coins
will  be  stolen  by  the  attacker.   While  there  may  be  methods  to  mitigate
the threat for the sender and the receiver, the intermediary nodes must be
online and will likely be processing the transaction automatically. For this
reason,  the  intermediary  nodes  will  be  at  risk  and  should  not  be  holding
a  substantial  amount  of  money  in  this  \hot  wallet."   Intermediary  nodes
which have better security will likely be able to out-compete others in the
long run  and  be able to  conduct greater transaction volume due to  lower
fees.  Historically, one of the largest component of fees and interest in the
 nancial system are from various forms of counterparty risk { in Bitcoin it
is possible that the largest component in fees will be derived from security
risk premiums.
A Funding Transaction may have multiple outputs with multiple Com-
mitment Transactions, with the Funding Transaction key and some Commit-
ment Transactions keys stored oine.  It is possible to create an equivalent
of a \Checking Account" and \Savings Account" by moving funds between
outputs  from  a  Funding  Transaction,  with  the  \Savings  Account"  stored
oine and requiring additional signatures from security services.

 

[IadixDev]

Presented like this it remind me of the pb with buffered io when the io is too slow downstream, and adding a buffer will just end up with the same blocked io latter with a bigger buffer. Buffered can just help smoothing the ups & down,  and it would be a bit like putting a time out on the write io, without really checking the all the data actually been sent at the end of the time out ;p knowing that most likely the throughput of LN is much higher than blockchain, it still lead to the question if the tx can be really processed on the chain before the lock expire,  and if the goal is to write them on chain at the end, if it's really going to do this faster.

Idk to me LN doesn't seem too bad in itself, but I just find there is something a bit askew with the mechanism of locking & parallel processing.

[mining1]

And this is one of the many reasons pow sucks. You need a system similar to POS to make it long term. But bitcoin and others will never change because centralized miners in china don't wanna kill the golden goose.

[dinofelis]

Presented like this it remind me of the pb with buffered io when the io is too slow downstream, and adding a buffer will just end up with the same blocked io latter with a bigger buffer. Buffered can just help smoothing the ups & down,  and it would be a bit like putting a time out on the write io, without really checking the all the data actually been sent at the end of the time out ;p knowing that most likely the throughput of LN is much higher than blockchain, it still lead to the question if the tx can be really processed on the chain before the lock expire,  and if the goal is to write them on chain at the end, if it's really going to do this faster.

Idk to me LN doesn't seem too bad in itself, but I just find there is something a bit askew with the mechanism of locking & parallel processing.

I think you've misunderstood the principle of the LN.  The idea of the LN is NOT to have a "buffered" list of transactions that have to go on-chain.  If that were the case, it wouldn't have any purpose.  The danger with the LN is that "by panic" it becomes such a system at a certain point.  But normally, the LN has more or less the following principle.

Consider first a toy LN with 2 nodes, Alice and Bob.  I even think that this embryonic proposal was already in Satoshi's paper.

Alice and bob want to pay one-another several times.  They could simply do that each time on-chain.  But they could also put each of them, say, 100 coins in an escrow-type of wallet (on chain).  Now, suppose that Alice wants to pay 20 coins to Bob.  She could send a *transaction signature* to Bob that would allow Bob to obtain 20 coins from Alice's wallet ; only, the idea is that Bob doesn't broadcast that transaction, but simply keeps the signature on his disk.  If later, Bob wants to pay 20 coins to Alice, he sends HER a transaction signature, which would override the signature HE obtained from Alice IF EVER he broadcasted it.  And if Alice now pays 20 coins to Bob again, she can send HIM a transaction signature that would override Bob's signature that would have overridden Alice's first signature if ever it was broadcasted.

The whole idea is that people send one-another transaction signatures that override previously obtained transaction signatures.

In other words, if at the end of the day, those 20 coins went 50 times to Bob, and 51 times to Alice, Alice and Bob have simply exchanged signatures amongst themselves, and NOTHING HAPPENED ON THE BLOCK CHAIN.   When they both want to quit, they can opt to:
1) send eachother the final signatures to release the coins to one another in the right amounts of the final balance, if they cooperate
2) DUMP THEIR WHOLE LIST OF TRANSACTIONS on the chain, which will then result in 50 transactions from Alice to Bob, and 51 transactions from Bob to Alice, ending up doing the same.

If the LN link "finishes cleanly" with mutual cooperation, then both of them could have exchanged even 1000 signatures, at the end of the day, only one or two transactions make the final balance on chain.  The idea is that there's no way to cheat, if you try to cheat, the partner dumps the whole list of transactions and obtains in any case what was the final balance.  And you're motivated NOT to do that, because you'd have to pay 1000 transaction fees, while now, you can do with only one.

The LN blows this scale up: instead of having 2 nodes exchanging coins back and forth, a node can open such 1-1 links to several partners, making up a mesh.  Suppose that Alice is connected to Bob and Claire, and Bob is connected to Joe and Jack.  If Alice wants to pay Jack, she can send the money to Bob who can send it to Jack (with a safeguard that Bob cannot keep it).  A lot of payments can hence travel across a lot of 1-1 links without a single on-chain transaction.  A node has to go "on chain" if his balance drops to 0 of course.  Each time the partners cooperate, they can just do one or two transactions to get to the final balance ; if not, they dump ALL their transactions on chain and obtain the same result but with much more transaction fees.

Note: I simplified.  The LN is somewhat more involved.  But you get the idea.

[IadixDev]

Yes I didn't look in depth yet into it, so I might have a wrong idea I know im missing many things about it

I can understand the part with merging transactions to spare some from being mined, the case you present is case where its useful,but as the tps ratio of LN is still much higher, do you have a rough idea of how many tx can really be saved this way, and it can garantee the tx from a locked tx will be updated to the blockchain before the lock expire ? Like there will never be a flow of merged tx who would take more time than the lock to expire ?


Cause LN as no way to notify the other node or force them to extend the lock if that happen right ?

For me it's still the pb that the volume of merged tx to be synchronized back on the chain must be kept under control, otherwise it's bit like fractional reserve on the tx processing power. With the same potential issue if everyone want to get their tx back on the main chain all at once.


It's why to me it would be somehow cleaner to mark those tx with a special marker to say they are indefinitely locked  on LN, until another tx from LN make them available again.

That would mean probably slight upgrade  of the protocol because such tx would have no point in an "intra chain" view, as it would make the bitcoin just virtually disapear entierely from the chain, but that would be somehow more clear, and would solve the expire issue.

Because the only thing that I see a bit weird is that it still make the btc state in a sort of Split, they are marked as locked on the chain, meaning there is no way anyone on the network can move those coins before the lock expire, whereas they are effectively moved on the LN, without there is anyway to know it can even possibly happen from the chain.

And it seem dangerous to put a non reiterable timeout on the lock when you cant have a very good predictibility of if the state can be fully synchronized back with the blockchain in this time.


But that's mostly the concern id have with it, otherwise it seem a good idea in the principle, clearly something to be tried.

 

[dwgscale11]

Kiklo the biggest shill that you've seen thus far.

 

[kiklo]

Kiklo the biggest shill that you've seen thus far.

Dwgscale11 the biggest Crybaby in the forums.  



 

[Sumo]

Now they will have to support segwit or suffer from falling prices lol On the flipside Segwit support will make their coins worth more.

[kiklo]

Now they will have to support segwit or suffer from falling prices lol On the flipside Segwit support will make their coins worth more.

Exactly who has put that into writing, that LTC will be worth more?
Groestlcoin activated segwit, and they are being dumped at the moment.


 

[Sumo]

Now they will have to support segwit or suffer from falling prices lol On the flipside Segwit support will make their coins worth more.

Exactly who has put that into writing, that LTC will be worth more?
Groestlcoin activated segwit, and they are being dumped at the moment.


 

Groestlwho? Never heard of it

[kiklo]

Groestlwho? Never heard of it

http://coinmarketcap.com/currencies/groestlcoin/

https://bitcointalk.org/index.php?topic=525926.0

Any problems with segwit or everything went smooth?

No problems at all, all is working flawlessly as designed.

[Ayers]

And this is one of the many reasons pow sucks. You need a system similar to POS to make it long term. But bitcoin and others will never change because centralized miners in china don't wanna kill the golden goose.

pos suck even more, who buy more coins on the market win, because he can stake more quickly and dump faster, pow is more fair, you can't just go out and buy tons of gpu, as they require space maintenance, electricity ,cooling etc...and last but not last pos is more prone to be double spended, remember mintcoin and its exchange?

[kiklo]

And this is one of the many reasons pow sucks. You need a system similar to POS to make it long term. But bitcoin and others will never change because centralized miners in china don't wanna kill the golden goose.

pos suck even more, who buy more coins on the market win, because he can stake more quickly and dump faster, pow is more fair, you can't just go out and buy tons of gpu, as they require space maintenance, electricity ,cooling etc...and last but not last pos is more prone to be double spended, remember mintcoin and its exchange?

Spoken like a PoW miner, Completely Wrong, but hey your a PoW miner, thinking is not your strong suit. 

PoW is only fair to the Rich, that can afford a warehouse full of ASICS.

PoS is the only real game in town for the rest of the world.
PoS that incorporates Coin Age is better protected than PoW from Double spends.
Mintcoin never had any double spends because of Proof of stake to my knowledge, post a link if you have one.
(Are you talking about the security problems that was caused because they were at one time PoW & PoS, which is why they dropped the PoW.)

Mintpal & mintcoin had no affiliations.

 

[BillyBobZorton]

And this is one of the many reasons pow sucks. You need a system similar to POS to make it long term. But bitcoin and others will never change because centralized miners in china don't wanna kill the golden goose.

pos suck even more, who buy more coins on the market win, because he can stake more quickly and dump faster, pow is more fair, you can't just go out and buy tons of gpu, as they require space maintenance, electricity ,cooling etc...and last but not last pos is more prone to be double spended, remember mintcoin and its exchange?

Spoken like a PoW miner, Completely Wrong, but hey your a PoW miner, thinking is not your strong suit.  

PoW is only fair to the Rich, that can afford a warehouse full of ASICS.

PoS is the only real game in town for the rest of the world.
PoS that incorporates Coin Age is better protected than PoW from Double spends.
Mintcoin never had any double spends because of Proof of stake to my knowledge, post a link if you have one.
(Are you talking about the security problems that was caused because they were at one time PoW & PoS, which is why they dropped the PoW.)

Mintpal & mintcoin had no affiliations.

 

PoS is flawed. It has been proven to be insecure endless times due the nothing at stake problem which has never been solved.

No PoS coin will ever, ever, ever deliver the security of a network like bitcoin's. You aren't solving anything with PoS. PoS has more problems than PoW and is not as secure.

No one is going to hold value long term in PoS coins, sorry.

[kiklo]

And this is one of the many reasons pow sucks. You need a system similar to POS to make it long term. But bitcoin and others will never change because centralized miners in china don't wanna kill the golden goose.

pos suck even more, who buy more coins on the market win, because he can stake more quickly and dump faster, pow is more fair, you can't just go out and buy tons of gpu, as they require space maintenance, electricity ,cooling etc...and last but not last pos is more prone to be double spended, remember mintcoin and its exchange?

Spoken like a PoW miner, Completely Wrong, but hey your a PoW miner, thinking is not your strong suit.  

PoW is only fair to the Rich, that can afford a warehouse full of ASICS.

PoS is the only real game in town for the rest of the world.
PoS that incorporates Coin Age is better protected than PoW from Double spends.
Mintcoin never had any double spends because of Proof of stake to my knowledge, post a link if you have one.
(Are you talking about the security problems that was caused because they were at one time PoW & PoS, which is why they dropped the PoW.)

Mintpal & mintcoin had no affiliations.

 

PoS is flawed. It has been proven to be insecure endless times due the nothing at stake problem which has never been solved.

No PoS coin will ever, ever, ever deliver the security of a network like bitcoin's. You aren't solving anything with PoS. PoS has more problems than PoW and is not as secure.

No one is going to hold value long term in PoS coins, sorry.

You are one of those idiots that believ G.Maxwell Nothing at stake bullshit,
Read this forum , I destroy his lies:
https://bitcointalk.org/index.php?topic=1709776.0

PoW is a dead end, only the rich will mine.

PoS will defeat PoW, don't worry I don't care if you don't profit from it.  

 

FYI:
Your so called secured BTC, has been vulnerable to a 51% attack from the China for over a year.
It has no security!

Also BTC Core Dev refusal to fix the transactions issue , is hurting it.
Egad. 81,000 + Transactions Unconfirmed. Again. Ugh!   :  https://bitcointalk.org/index.php?topic=1799541.0
Transaction Fees are SPIKING !                : https://bitcointalk.org/index.php?topic=1801057.0

[IadixDev]

For me the issue of security with coins is not necessarily mostly about technical issues first.

It's what make bitcoin very successful is because it make in sort everyone participate get some reward somewhere. It's secure because 51% of users keep it so because they have an interest to do it.

With pos ok some people could buy all the coins and screw the chain, but what does he win ? nothing, he just lose all his money dry. No point in doing this.

And i think for small chain, pos can be more secure because with pow there can always be the risk someone outside the network come with some asics and over power it, with pos it need someone who has coin and is on the network. For small chain i tend to think the risk is still smaller with pos than pow.

And there is something i find really nice with pos, is the authority on the chain can be distributed very easily, maybe it's not very useful in the perspective of coins, but if something like a "proof of authority" could be useful in some scenario, with pow it would need the node to compute a big thing to prove it has the pow, and it would need to do this each time it need to give a "proof of authority", and it would be a big waste, and harder to be sure it's distributed evenly between a certain number of entities, with staking just need to check the network weight. And it make it much easier to distribute the authority on the chain, in certain case it can be interesting.

Maybe some small proof of work should be added to the stake modifier computation to incite people keeping on a single chain to make it a bit more secure.

[kiklo]

Maybe some small proof of work should be added to the stake modifier computation to incite people keeping on a single chain to make it a bit more secure.

Actually adding PoW to a PoS coin has the exact Opposite effect, it adds attack vectors that do not exist without PoW included.
(Reason ZEIT & Mintcoin both dropped PoW from our coins.)

Security analysis of PoW/PoS hybrids with low PoW reward
https://bitcointalk.org/index.php?topic=551861.0

Attacks were Proved to Work by Rat4

Security analysis of PoW/PoS hybrids with low PoW reward

Low PoW reward doesn't attract miners. This leads to ridiculously low PoW difficulty.

A pair of examples:
Mintcoin scrypt diff 0.1 (vs Litecoin 5677)
SHACoin sha256 diff 1427 (vs Bitcoin 5006860589)

At such difficulty PoW blocks can be mined with speed of light.

Attack I

It is possible to build sequential chain of PoW blocks to confirm a transaction. Only 4 blocks for Mintcoin and 10 for SHACoin.

Is it hard to orphan the chain of PoW blocks?
One PoS block is enough. In both Mintcoin and SHACoin one PoS block may orphan a few millions of PoW blocks.
If at the same time the main chain will get a competing stake, attacker's chain can be enlarged with PoW.
This dramatically increases chance to success in comparison to pure PoW attack.

Ability to confirm a transaction and then orphan confirmations is ability to double spend.

Summary: double spend attack requires 1 PoS block and low hashing power.

Visualization: https://i.imgur.com/Pyrw75q.png

Attack II

Current implementation of stake miner gives up if median time of last blocks is in future.
This temporarily makes the whole network PoW-only and opens well known 51% PoW attack.

Attacker needs only 6 of 11 last blocks.

Successfully tested on Mintcoin: no PoS blocks from 203231 up to 203441, more than 1 hour of real time.

 

FYI:
Funny thing is people see this giant # for PoW has and think it is more secure.
When the fact is ,
I can have a      PoS coin with a  difficulty   of 1
and compare to  BTC with it's     difficulty      440,779,902,287 .

The PoS coin with the difficulty # of 1 is just as secure if not more so than the PoW coin with 440,779,902,287.

Because even with that ultra high difficulty, the Chinese Miners could overwrite the last 12 hours or more whenever they felt like it. ( PoW offers False Security)
On the other Hand,
No one has been able to overwrite even 1 hour on a PoS only chain with a decent staking difficulty ,
plus when someone stakes their coins become inactive for a period of time , blocking a continuous control of the network ,
unlike a PoW ASICS which can dominate a chain indefinitely.

[IadixDev]

Im saying adding pow on the stack modifier computation not on the block themselves to avoid the staking on multiple chains.

The way i understand it, the problem with POS chain is that for example in blackcoin there is granulosity of 16 sec on the time nonce used to compute the proof of stake, meaning for each output you are staking on, you have 16 sec to compute the hash for it, so it leave lot of free time, and it would be very possible for an attacker to keep in permanence different valid version of the chain, and push one or the other on the network at any time. It's what make pos coin more vulnerable to transaction maleability. Even if it's not that useable in practice i think, it can still be potentially a pb. Adding some pow on the stake modifier could avoid this to be exploited too much.