New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
You can route between multiple parties like IP routing. There can be many hops between you and the party you are transacting with.
Yes but you can commit more to the channels if you receive payments from other people.
User B is a seller and expects only ever to receive funds. So they as a seller do not commit any BTC to open a channel. But in order for the hub to actually transfer funds to user B, the hub needs to commit some BTC, which cannot be accessed or used for anything else while the channel is open. In order for user A to send 1BTC to user B, the funds committed by the hub to the channel with user B must be 1BTC or greater.
Actually such transfers are atomic in nature so there is no credit or loaning of funds. If payment is being routed from A-> B via C then the transaction that pays out B pays pays out C at the same time. It's called a hash timelocked contract https://en.bitcoin.it/wiki/Hashed_Timelock_Contracts
And you are compensated from transaction fees so it's not like you are getting nothing out of it. If you don't want to facilitate payments to third parties don't. No one is forcing you to do it. People who have lots of bitcoin may choose to help route payments in exchange for transaction fees. It's a business like any other with cost, risk and reward.
It'll scale to millions of transactions per second compared to the current limit of 3 per second.
All the funds in your hot wallet will be on lightning. It's the most useful place to keep them.
IRL people living in first world countries choose to keep most of their money in banks and make payments from their accounts there using credit cards, debit cards, bank transfer and so on. No one calls that money "blocked" or wasted. OTOH in developing countries people keep more cash around because cash has greater utility than money in the bank. Note lightning is not a bank account (it is trustless). I'm just trying to explain it with an analogy you might understand.
Which requires even MORE funds locked up in channels.
If A wants to send 1BTC to B, and there are 5 hops between them, the channels between A and B each need to have at least 1BTC committed to each channel. 5 hops = 5BTC required in channels for a 1BTC transaction.
Which requires a block chain transaction.
Read what I wrote. You cannot exceed the amounts committed to each channel. This is a limiting factor. I make no mention of credit or loans.
Which will make it about as expensive as credit cards. Capital can fairly easily generate 5% /year. so the fees need to be competitive. Not to mention, there are still blockchain transaction fees you need to deal with whenever things get cleared.
It's more like 7 second, on the block chain. And no it wont acheive millions per second unless funds are just going back and forth on the same channels. Which it wont. To be fair, the actual transaction volume it can handle will be difficult to estimate given there are different bottlenecks at different places. Some places of the network will be near infinite, but other areas not at all.
Sure, for buyers, it's fine. But sellers it's not. Sellers will deal with larger volumes than buyers, and those channels will require very significant amounts of BTC comitted to it. Which you seem to not think is an issue.
What? That's not even remotely comparable. Credit cards do not block any funds. Neither to bank accounts.
A better comparison would be when you put a downpayment on something. For example, you wanna buy a Tesla Roadster. You need to put down $30k. That's $30k you cannot spend. You do not 'lose' that $30K, because you get the car later. But you cant do anything with it. And there are opportunity costs locking up funds.
You need to look into this more. You seem to not recognise the key limitations here.
a) Each of the hops are getting a cut of the transaction fees so they are not doing it for free.
b) Transactions are instantaneous so you can use those funds for other transactions very quickly.
Only if the person sending you the funds hasn't already got an open channel.
Which bank account gives you 5% a year for bitcoin deposits? Even for inflationary USD deposits you don't make more than 1%. Bitcoin is a deflationary currency so it appreciates on its own.
It won't be as expensive as CC. Even on-chain payments aren't that expensive currently and lightning promises to increase capacity many times that.
It's not blocking funds. You can use your payment channels to send and receive bitcoin cheaper than on-chain so it saves you money even if you choose not to route for third parties.
The fact that they earn a fee is not relevant. The point is that you need a massive amount of deposits in order to facilitate transactions.
What? Use where? For what? How? You have an idealistic view and ignore the pitfalls.
What?
I did not say BTC earns 5%. I said capital earns 5%, and quite easily. BTC being deflationary introduces a whole bunch of other issues that make BTC not suitable for a currency at all. But that's not relevant to Lightning Network so we need not discuss it.
Yes it is.
Not necessarily.
Okay let me pose a very simple hypothetical scenario to you.
Lets assume that there are 3 types of entities. (A) buyer, (B) hub and (C) merchants.
When buyers open channels with hubs, they commit BTC to the channel. Hub operators does not need to commit anything (in theory).
So I'm a merchant and I want to use the hub operator to process BTC payments on the Lightning Network. I expect to be able to process 10BTC per year. I don't plan to spend my BTC, I want to clear it to my block chain wallet.
How much does the hub operator commit to the channel?
How many merchant channels can the hub keep open?
What if a merchant needs more than 10BTC in payments?
The very fact that that the the transaction limits are determined by the amounts 'deposited' into the channel limits the scalability beyond a completely unrealistic and idealistic use case.
The more hops you add, the more ridiculous because you need even more resources committed, and you add cost. You also introduce bottlenecks. Yes, hub operators get fees, but that doesn't magically solve the problem does it.
For other transactions on the lightning network. Your funds are not "blocked" for long is what i mean.
You can route payments to anyone on the network. So if you want to receive money from someone else and they already have a channel open to anyone else on the network they can send you money without having to open another channel with you directly i.e. without having to resort to an on-chain channel opening transaction.
You don't use a hub to process payments for you. You connect to any node on the network you like. The advantage of connecting to a well networked node, i.e. what you call a hub, is that you have fewer hops to get to other people and save on transaction fees. But then the hub will also have more liquidity otherwise it wouldn't be able to keep everyone happy and people would move to another well connected node.
If there is no routing path available with sufficient liquidity then each of your customers opens a separate channel with you directly. They are still able to pay you multiple times with that one channel at the cost of one channel opening transaction and one channel closing one. It's still an improvement over on-chain transactions.
Capital that earns 5% isn't available on demand. You have to lock it up for a year. You have to bear greater risk. Bank accounts that let you withdraw whenever you like only give you 1%.
Legal: Artifical Intellect, some kind of global data blockchain systems, construction of mathematical models, star evolution, biological researches, genetic modeling and other
Illegal: Building AI for hacking and intellegence, self-learning (D)DoS mechanism
Yeah but that's not how it works is it. Again.. this is a very idealistic view and not realistic in terms of how it actually works. Look at Paypal for example. Most merchants will receive far more via Paypal than they make payments with. Yes you can keep money in Paypal, and yes money can flow around freely in Paypal and yes that's what paypal would prefer. But in reality merchant will take out the majority of their earnings and only leave a small amount.
So you make an assumption that the funds will continuously cycle, even though there is no reason to assume that would ever be the case.
What? You connect to a network so you can recieve payments via a hub. That's what I mean by processing. So yes, you connect to a hub to process payments for you. That's exactly what it's for.
The more channels the hub (node) has, the more BTC required to maintain those channels. More channels technically reduces liquidity. Having more buyer channels also does not in any way add liquidity.
There is literally no point in opening a direct channel between buyer and merchant. I think it's pretty obvious why.
It's not in any way an improvement. If I'm a buyer, and I open a channel with a seller, basically, lets say I commit 1BTC to that channel, that's 1BTC I cannot spend anywhere else. Why would I do that? It's basically like 'adding funds' to a WHMCS billing account, but with the possibility of clawing back what I don't spend. If I put $100 into a WHMCS billing account with a provider, that $100 is locked away in the same way BTC is locked into the channel. I cannot spend that BTC anywhere else, just as I cannot spend that $100 anywhere else. So the ONLY benefit is that I can get the $100 back if I force close the channel. But opening and closing a channel requires a block chain transaction, which cost money.
Actually a very large number of blue chip stocks pay around 5% dividend. They are readily available and highly liquid. They are also far more stable than BTC in terms of pricing. But this is not really an issue.
The point I really wanted to make was that you need an enormous amount of capital to maintain all the channels. And the more people you have on the network, the more capital is required. And the amount of capital required to maintain channels between hubs/nodes will be ridiculously huge.
There are far too many limitations that will make Lightning Network impractical to use.
I did not even answer because I found it so strange. For a start, a FPGA devotes ANYTHING because either customers in general want it (e.g. fast serial IO, memory interface) or because a given customer needs it and hence puts it in his image.
And the very idea of creating an ASIC based on eliminating major blocks of a GPU just looks weird and unrealistic to me. Btw them widespread CPUs and GPUs are extremely complex designs and I doubt that some small design company would even get at the necessary designs to create e.g. a modern Nvidia GPU but without the multipliers.
Looking from the crypto perspective most algos are based on simple and fast bit operations and on basic arithmetic of which only multiplication, division/mod are expensive. Pretty any modern CPU is insanely fast with bit operations and one major reason for desiring custom chips of some kinds is fast basic arithmetic, typically either in the form of building blocks (e.g. Montgomery ladder) or in the general but strongly optimized form and often with wider registers of 256 or even more bits.
So @willies post just sounded strange to me.
Exactly. I won't say a lot about mining but for (usually institutional players) that's what's done and it's often done ad extremum. Very specialized ASICS that only do one thing but insanely fast are the desired results. Power is less of a concern but binding many, many of them into large computational arrays is, so often one uses the approach of creating "mainboards" with 1 normal CPU (housekeeping, control & communication) and 8 or 16 (or even more) of those ASICs.
Yes and no. Principally yes but smaller batches (which those applications typically have) might have to wait quite some time to be processed/produced.
Yes (cost:value), and: 20 nm is already the higher end for ASICS. Even CPUs even in relatively large quantities are often still produced in 45 (and even more) nm. It's much cheaper, it'll use much less power, be easier to work with in board design, and be plenty fast enough.
Looking again from the crypto perspective: My goal isn't to gain another relatively small advantage (2-5: 1) by having my ASIC in 20 nm and running at insane clock rates. My goal is to gain an advantage of 1000+:1 and more by having the algorithm run on specialized hardware and to be able to run thousands of those (which is MUCH easier and cheaper with chips that are not pushed to their limits).
Side note: No, RAM is usually not a problem. The simple reason is that I often have none for the ASIC. Remember what I said above about those boards with 1 CPU and a group of ASICs. The housekeeping, controlling, communicating CPU needs and has RAM. The ASICs however don't. They have the right amount and size of on die "registers".
Just think of an ASIC that does for example AES. (a) It doesn't need RAM because there is no "programm" and because the algorithm doesn't need but very little memory (on die). (b) Look at the access times. RAM is about 100 times slower than on-die memory plus it makes my chip design more complex.
At the same time this is why hashing with large state size, especially PHFs (e.g. Argon) and KDFs are a PITA and a good choice for crypto currencies. Their very nature forces us to have way considerably more complex and at the same time limited ASICS and boards. Plus of course loads of RAM which take up space and are in one way or the other slow and expensive.
And that's the major factor why I love PHFs/KDFs. With even a modest m factor of say 8 MB you can considerably drive up the cost and even complexity of NSA and friends. The t factor unfortunately still is largely just "run fewer or more rounds of the algo" but if we create "erratically behaving" "chameleon" algos and use that in t we can make the lives at NSA quite miserable.
That said this at the same time is a strong argument to doubt the "democratization of currencies" (or to even call that BS). Simple reason: We COULD very seriously limit the advantage rich massive players have over normal people - but they didn't. Probably for a reason.
@Abdussamad
Sorry but to me that sounds just like the (not at all) good old bankers game of binding funds and keeping them in their system as long as possible.
Well that's not exactly what's happening. The funds that get locked up do cannot be 'stolen' per se as transactions still need to be signed.
But the point is that with all this locking up of funds, and the fact that transaction sizes are limited to as much as has been committed to any given channel limits this system enormously. It creates many many bottlenecks, adds cost and ultimately does not scale in real world situation.
The only way it can scale transaction volume is if money is going back and forth between users in a totally balanced way. But without complete balance, it will be absolutely necessary to clear to go back to the block chain.
To assume that absolute balance is achievable is idealistic at best.
And given it scales only when transactions are totally balanced, to say it scales is disingenuous, and a major misconception.
And yes, funds ARE BLOCKED while they are maintaining a channel in the same way that your deposit is 'locked' while waiting for your Tesla Roadster. Those are equivalent analogies.
For the merchant with an unbalanced channel:
https://www.reddit.com/r/Bitcoin/comments/8rgq0c/lightning_network_submarine_swaps/e0rmtfe/
Tbh i don't really follow any of this since i can't be arsed to care enough for all the technical details but from the overall impression it seems like duct tape and complicated duct tape at that. From my perspective ease of use of crypto currencies is already bad enough and stuff like that is not really going to increase adoption. Not saying anything about the actual practicability since as i said i don't care enough.
It's too complicated and too idealistic. In theory it could work if you could balance transactions across channels. But I don't it would work in practice. Especially since you need to add trust into the equation, which defeats the purpose of this solution.
If we are going to accept having to trust some parties, then you can get a much better solution in other ways.
As you seem to be determined to continue preaching, let me help you out by enhancing your quote somewhat:
"could", "hopefully" ... not exactly a solid position to evangelize.
Let's make a simple test. You transfer 10.000 (real) $ to me which I COULD and HOPEFULLY will transfer back. Deal?
The same thing that happened to all the video stores
I think the more important thing is that it requires trust to scale, which means the complexity of the lightning network is completely pointless.
Trust? The combination bitcoin+exchange? No way.
This whole thing is ugly and gets ever uglier.
Rule number 1: Keep it simple. Unnecessary complexity is the natural enemy of safety and security.
Rule number 2: A rotten house doesn't get better by adding rotten extensions.
Rule number 3: Never confuse math, software, and profit mechanisms and never make the formers solidity depend or bet on the latters solidity. If you do it anyway it will turn out to be a horrible loss mechanism for almost everybody. Rather have solid math plus solid software plus solid profit/business mechanisms.
Rule number 4: Adding trust as vital ingredient to a rotten mechanism is a confession of idiocy and failure.
Dont really know what you mean from points 1 2 or 3, but for point 4, I strongly disagree.
Though BTC and other crypto currencies can be considered 'trustless', there is always trust involved in everything.
Trust is a vital component of every aspect of society, including commerce and the internet.
Think of it this way, when you pay someone for a good or service, the buyer needs to trust that the seller will deliver after payment is complete. So there is still trust required in a transaction.
And, when it comes to the crypto world, most people will use exchanges extensively, and this requires complete trust in these institutions. And some of the bigger exchanges actually hold pretty substantial sums on behalf of their users as a lot of people don't feel comfortable being responsible for maintaining their own wallets.
If you're going to have an offchain solution to the scalability problem of BTC, and you introduce some trust into the system to actually make it work (as indicated and described by @Abdussamad) then Lightning network is needlessly complicated as all those channels and transaction signing is completely unnecessary. And I do mean COMPLETELY unnecessary.
Who said it requires trust? The swap itself is trustless. I mean you complained about unbalanced channels and I show you how that can be dealt with in a trustless off-chain manner. The quote itself says as much:
That comment is a list of possible use cases for new lightning swaps. That's why it says "could". Here's how this new technology could be used.
You say it's too complicated well here's why it has to be complicated.
At least point 1 is really really simple. It basically comes down to the fact that the harder it is to follow something the less likely errors and flaws will be spotted. Also more complexity has more room for flaws. The most secure things are often dumb to the point of hurting simply because there is little margin for flaws.
That's an awful analogy in my opinion. I think the basic functioning of a rocket could be explained way simpler than any of this.
Reading this thread makes me feel like reading a debate on politics.
Nobody agrees with anyone and no conclusion is ever reached along with odd daydreamers.
No it's not trustless. There must be some trust. If I'm wrong, feel free to provide an example of the money flows and how full trustlessness is achieved.
My point wasn't that trust is useless or bad.
My point was that adding (need for) trust to a rotten mechanism (in the context of "safe and secure" which certainly is a necessary and valid context here) is a confession of idiocy and failure.