how are miners going to get paid when block reward goes away?
tx fees years ago we just kinda shrugged this off saying "one day there will be several orders of magnitude more TX and the fees will pay the miners"
at 1 or 2 MB there's just no way fees can replace block reward.
By the time of the next 2-3-4 halvings, our blocksize limits could be in the hundreds of megabytes. |
|
|
|
I think it's a good analogy, which makes the issue easier to understand. If there are flaws in the analogy, I'm happy to discuss them. The criticism of AlexGR that people on a bus are real, and Bitcoin transactions in a block (according to him) are not, is just childish.
His other remark that it's cheap to buy all the space in a block (because blocks are quite small and transactions cheap), doesn't discredit the analogy. If someone wants to take out the high speed train from Amsterdam to Paris, he can buy all the train tickets. It's doable, and effectively a DOS attack preventing any other people from using the train. If anything, it shows my analogy is a correct one.
Not there yet. When you broadcast a tx, where you "pay", say, 1 satoshi per byte, what you are *really* doing is that you are stating your intention that if you get included in some block then you will pay the said amount. You don't actually pay anything beforehand. The payment is only done upon inclusion. If you get the service, you get paid. If someone else pays more than you, then HE gets it, not you. In that scenario, where he got in and you didn't, the only party paying is him, not you. You haven't paid anything. You only said that you were willing to pay a trivial amount, which was less than him, and the miner said ok, you aren't paying me that much, so I'm going to process that other guy who pays me more. I can imagine that a flood of tx without fee can be considered spam. But nowadays, only mining pools use zero fee tx to payout their miners. There are hardly any zero fee tx issued otherwise. (800 in the past 24 hours, source: https://bitcoinfees.21.co/ ) But if a tx carries a fee, how to decide if it's spam or not? What is your definition of spam? Likewise, remember, you haven't paid anything until inclusion. You are just broadcasting your intentions, for free. In this sense, broadcasting a billion txs and "paying a fee" of 1 satoshi/byte doesn't mean anything. The broadcasting of this billion txs is free. If a few thousands of these txs go in, if at all, I may have to pay 1 satoshi/byte. This is precisely why the mempool backlog is a useless metric. It could be 50 gigabytes of backlogged txs, which even the sender isn't really expecting to get included.
Personally, I'm not anxious for all that spam to be "cleared" by getting included in the blockchain.
It would be like having the street outside your house filled with trash and considering collecting them inside your house to "clean the street".
Nice analogy, it illustrates how we disagree. Going with the trash theme, IMO the current setup is akin to saying everything people throw away should be dumped and never touched again. Never mind that due to changing market conditions and innovation, some trash may actually be valuable to people if they can get their hands on it - recycling metals for example. If you remove, by dictate, the ability to process "trash" beyond some arbitrary treshold, you remove the markets ability to correctly value that "trash". Well, physical garbage do tend to have more valuable stuff in them compared to digital... Anyway, in the above analogy, the street garbage is the mempool, and your house is the blockchain. Now, the mempool can be spammed A LOT because it is practically free to do so (broadcasting intentions to get included, either for free, or with a trivial fee) and you can decide whether you will do what the spammer wants, or not. The market always has an ability to value the trash: If the sender of the tx doesn't care for his tx (broadcasting zero and near-zero cost txs) he is betraying his intentions. When you see "blocks are full" and fees topping out at 4-5-6 cents for 1st block inclusion, without activating any serious fee competition, you know right away what is going on: Junk transactions are flowing in. If, say, I want to send my coins to an exchange and sell right away there are various costs: -Exchange commissions -Bank wire/withdrawal costs -Conversion fees (for non-$ users) ...which might add up to tens of dollars. Now why would I be "cheap" when I'm sending the tx?... to save what? 0.03$? |
|
|
|
Why do I feel you are deliberately choosing not to see the other POV? Let's forget for a minute that you seem to have the ability to tell apart a user transaction from a spam transaction and suggest that actually the daily demand for transactions rises above 250,000 consistently. Tell me how exactly users who are unable to get a transaction confirmed (they just sit there ad infinitum) are not 'disrupted'?
Since when did failure of transactions to be written to the chain become a feature? It is a sign of failure.
You said: The best bit is that as the network becomes increasingly congested actually performing a flooding attack to completely disrupt the network becomes trivially cheap to employ.
And I just noted that the network cannot be disrupted (let alone "completely"), it will continue to process its 250k txs per day, with the highest-paid-tx-gets-included queue. The idea that every tx has to be included and the network must upgrade to compensate is wrong because it is a self-feeding loop that tends to infinite spam and abuse. If I am a spammer and you give me 1mb to fill, and I fill it, and you give me 10mb and I do the same, will you keep giving me 100mb, 1gb, 10gb blocks etc etc, where I fill them all? Do you think this kind of self-defeating system is some kind of "success"? Are there many blockchains, beyond bitcoin, where you can do just that and where the devs haven't taken action to save their blockchain - typically through sharper fees to act as a deterrent to the attackers? The case of having 250.000 legit txs and zero spam, and them having to compete for that 250k tx space is just theoretical. Why? Because actual txs are way below that point and by the time actual txs double or triple (which could take >1 year), we'll be 1.7 or even more (which is right ahead). So rejoice, more spam will be able to be included to the blockchain compared to our 700-800kb per block right now. |
|
|
|
The actual problem comes next if transactions continue to ramp up as they have progressively leading to a backlog which never clears.
Personally, I'm not anxious for all that spam to be "cleared" by getting included in the blockchain. It would be like having the street outside your house filled with trash and considering collecting them inside your house to "clean the street". When the network is spammed, two things can happen, or both of them to some degree. 1) the blockchain will get bloated with spam 2) the mempool will get bloated with spam Then it doesn't matter what price your wallet chooses, it will not guarantee inclusion by a miner, as the next person putting a transaction in will out bid you (yay fee market!).
This is only theoretical and provided all 1mb is legit use. Practice shows that even at "extreme" load, fees don't rise above 5-6 cents. The answer is simple: Legit txs are way below 1mb and the rest is topped off with spam. So there is no reason for fee competition at the top, as legit txs always outbid the spammers (provided their software is not crap). What you say would happen, if indeed there was sufficient legit demand to cover the 1mb (which, unfortunately, there isn't). The time when this will be so is coming but by the time legit activity hits 1mb in a year or so, or later, we'll be at 1.7mb or more, so, again, it won't matter. Being economically naive you would hurrah! at a working fee market. But actually by enforcing limited transactional scarcity with a blocksize cap what you actually do is break bitcoin for most people. Suddenly stuck transactions are widespread and sure you can still use bitcoin but fees will simply spiral upwards until they are ridiculous.
This is FUD (explained above why it has never worked that way). The best bit is that as the network becomes increasingly congested actually performing a flooding attack to completely disrupt the network becomes trivially cheap to employ.
The network will still process the best paying 250.000 txs per day, undisrupted, no matter how much one floods/spams it. |
|
|
|
I doubt anyone is suggesting small blocks forever, but at the moment there doesn't seem to be sufficient evidence of an emergency need to take some kind of drastic measures to increase the blocksize limit on an emergency basis.. and in that regard, the impression and/or creation of an emergency seems to be largely fabricated with a bunch of loud and whiny voices.
I think it was something like a year ago when Hearn predicted that urgent action must be taken or Bitcoin would ...crash. Yes, he said crash. The rationale was something like "ohh nodes will fill their mempools and nodes will start crashing and ohhhh, we need bigger blocks". For a programmer it's very suspect why he didn't simply write a patch and say "here guys, this will fix nodes from crashing through a parameter that sets mempool size"... Why promote a hard fork for 8-20mb increases (which would be 20-50 times the legit activity of the network) instead of simply patching the software with a mempool limit, as 0.12 does? The whole crisis / urgency scenario was bullshit, as was the "problem=>reaction=>solution" - in terms of ...proposed "solution". For some reason your answer entirely misses the fact that blocks are filling up. A cheap transaction flooding attack brought the network to it's knees just last week. You are that dog cartoon saying 'everything is fine' as the house burns down around you. "We can't have 1mbforevah" "We want 1cent-txsforevah" (and if we get to 2 cents we say the network is down to its knees because we didn't pay the 2-3-5 cent fee) Newsflash: The system is not based on the premise of 1 cent txs forevah. Fees are dynamic, based on the load. You don't pay the fees => you get in line. And wait. The network operates normally for those that follow proper tx pricing. If they don't, it's usually because their wallet makes wrong assumptions about fees. They should upgrade to core 0.12. But then they'd have to download 10gb txs and 60gb spam... ooops. The whole crisis / urgency scenario was bullshit, as was the "problem=>reaction=>solution" - in terms of ...proposed "solution".
Amen. "the malicious miner DoS limit from 2010 is still in place, delays & fees up, investor confidence in future capacity growth down => listen to technical experts: to increase a constant would be a harsh blow to muh decentralization, non-mining rasb pi node on LJR internet is sad => let me tell you about highly connected and well funded Blockstream pre-paid scrip payments lightning hubs, out in 2 weeks." Once "XT" appeared to be dead and buried the market went from low 200's to 500's. Is this lack of "investor confidence"? When did it go back down? When we had the coordinated threat of forking by Gavin + the FUD by Hearn. When did it go back up? When Classic was buried. The only thing the market cares is contentious hard forks. The capacity issue is priced in as are the dev responses to it (which, contrary to the propaganda, is not 1mbforevah). |
|
|
|
Το πλήθος των νομισμάτων που περιλαμβάνονται σε μια συναλλαγή, επηρεάζει το μέγεθος του μπλοκ περισσότερο ή λιγότερο σε σχέση με τον αριθμό των συναλλαγών;
Δηλαδή μπορεί μια συναλλαγή να περιλαμβάνει τόσα πολλά νομίσματα που να καταλάβει το σύνολο του μπλοκ;
Οχι, δεν εχει να κανει με το ποσα νομισματα διακινουνται. Βασικα αυτο ακριβως ειναι και το τεραστιο "αντιβαρο" στο θεμα του scaling του bitcoin. Οτι η διακινουμενη αξια μπορει να ειναι τεραστια, με μηδενικη επιβαρυνση για το συστημα. Δλδ το συστημα πες μπορει να κανει 250.000 συναλλαγες τη μερα αυτη τη στιγμη. Μπορει να κανεις 250.000 συναλλαγες του 0.001 btc ή 250.000 συναλλαγες των 10 btc. Δεν αλλαζει κατι στο φορτο του συστηματος, αλλα στη μια περιπτωση εχεις κανει 250btc ημερησιο ογκο (105.000$), στην αλλη εχεις κανει 2.5εκ btc ημερησιο ογκο (>1 δις $). Ενω δλδ το συστημα δε μπορει να κανει -αυτη τη στιγμη- απεριοριστο scaling στον αριθμο των συναλλαγων, μπορει να κανει τρελο scaling στη διακινουμενη αξια. Αυτη τη στιγμη ειμαστε γυρω στα 150εκ-230εκ $ τη μερα, το οποιο βγαινει 55-84 δις $ - μεγεθη συγκρισιμα με τη Western Union. |
|
|
|
I doubt anyone is suggesting small blocks forever, but at the moment there doesn't seem to be sufficient evidence of an emergency need to take some kind of drastic measures to increase the blocksize limit on an emergency basis.. and in that regard, the impression and/or creation of an emergency seems to be largely fabricated with a bunch of loud and whiny voices.
I think it was something like a year ago when Hearn predicted that urgent action must be taken or Bitcoin would ...crash. Yes, he said crash. The rationale was something like "ohh nodes will fill their mempools and nodes will start crashing and ohhhh, we need bigger blocks". For a programmer it's very suspect why he didn't simply write a patch and say "here guys, this will fix nodes from crashing through a parameter that sets mempool size"... Why promote a hard fork for 8-20mb increases (which would be 20-50 times the legit activity of the network) instead of simply patching the software with a mempool limit, as 0.12 does? The whole crisis / urgency scenario was bullshit, as was the "problem=>reaction=>solution" - in terms of ...proposed "solution". |
|
|
|
We're at 15GB. I think I heard the Ethereum blockchain is at 10GB already. And nobody actually uses it for anything yet. I say good luck to 'em.
The data are a bit rusty for btc (and the numbers for visa are also calculated wrong)... Right now we are closing to 70gb with btc. You really think miners are complete idiots, then. Chomping at the bit to bloat blocks to infinity with free spam...
~Come with me, reader, into the mind of the economic central planner... where "capitalism" is one broken production quota away from abject devastation and horror.~
If miners didn't mine txs => "Ohhh those bad miners are not mining txs... we must hard fork to show them a lesson" If miners do mine txs => "Ohhhh those bad miners are mining junk txs... we must give them a lesson" Whatever they do, it can be used for political friction. |
|
|
|
define "centralization"
I use these terms in a relative fashion, not absolute. This a p2p protocol which requires an extensive "p2p mesh" so to speak where a lot of people are running as peers, not clients dependent on some server. Decentralization = you want the network to be more on the p2p side rather than the client/server side. Centralization = the network tends to gravitate to a consolidated state where equal peers drop off and a few main central players are left standing. you place way too much weight on spam. think of spam as a placeholder for legit TX, we welcome the placeholder because it pays some miner fees
Spam is the most wasteful symptom. However, given ample supply and low actual demand, the extremely cheap use of bitcoin's blockspace will not attract only spam, but also other systems on top of bitcoin that will use its transaction space as a ...cheap distributed storage system, where the costs burden the network nodes. Thing is, if that happens, and as Satoshi said about the need to avoid DNS data being stored on the blockchain: Piling every proof-of-work quorum system in the world into one dataset doesn't scale.
...now Satoshi understood the problem, however BTC is a data-agnostic protocol. Right? It doesn't know whether you are creating your own system and using it for storage, whether you are transacting, spamming, etc etc... So how can you prevent the scenario where BTC's blockchain becomes a distributed-storage system that "doesn't scale", or, worse yet, a spam dumpster? The only two ways are limiting block size and raising tx price. As for now: If the fees were helping the network to sustain itself economically, which will be the case in some years, then we'll probably need a lot of txs. However this is not the case right now, since subsidy is like ...50-100 times larger than fees and mining can still be profitable as it is, or even ...more profitable with 0-tx blocks that avoid orphaning. |
|
|
|
AlexGR, do you allow for the possibility that maybe small blocks isn't a good idea?
Timing is crucial. Even 1TB per year blocks (20mb/block) will have its time when 20mb/block will be "alright". Upgrade too soon, you'll have 10gb txs and 990gb spam. Upgrade on time, you'll get 800-950gb txs and 50-200gb spam. |
|
|
|
Lol? The problem is that this is not an "argument by those who reject a block-size increase". It's simply the reality of the matter, whether it's BTC or ...Ethereum. https://github.com/ethereum/wiki/wiki/White-PaperScalability One common concern about Ethereum is the issue of scalability. Like Bitcoin, Ethereum suffers from the flaw that every transaction needs to be processed by every node in the network. With Bitcoin, the size of the current blockchain rests at about 15 GB, growing by about 1 MB per hour. If the Bitcoin network were to process Visa's 2000 transactions per second, it would grow by 1 MB per three seconds (1 GB per hour, 8 TB per year). Ethereum is likely to suffer a similar growth pattern, worsened by the fact that there will be many applications on top of the Ethereum blockchain instead of just a currency as is the case with Bitcoin, but ameliorated by the fact that Ethereum full nodes need to store just the state instead of the entire blockchain history. The problem with such a large blockchain size is centralization risk. If the blockchain size increases to, say, 100 TB, then the likely scenario would be that only a very small number of large businesses would run full nodes, with all regular users using light SPV nodes. In such a situation, there arises the potential concern that the full nodes could band together and all agree to cheat in some profitable fashion (eg. change the block reward, give themselves BTC). Light nodes would have no way of detecting this immediately. Of course, at least one honest full node would likely exist, and after a few hours information about the fraud would trickle out through channels like Reddit, but at that point it would be too late: it would be up to the ordinary users to organize an effort to blacklist the given blocks, a massive and likely infeasible coordination problem on a similar scale as that of pulling off a successful 51% attack. Again, this is from ...Ethereum, not BTC. So if Ethereum seriously considers centralization as a problem, and most if not all altcoins do too, then how is this relevant only for "btc smallblockers"? |
|
|
|
So the definition of spam is something that is bought cheap?
The point is that it doesn't matter if it's free or paid, if what you pay is near-zero cost. It definitely cannot be a factor of ruling out spam if the fees are too cheap and aren't an adequate deterrent. And even "adequate deterrent" is wrong, as a term, because a script kiddie might not afford a good spam attack but a deep-pocketed adversary may not be deterred by the costs, because by attacking in this fashion he is getting side-benefits by harming BTC. From these graphs I conclude that about 2/3rd of all tx are part of chains longer than 10. So what? When you receive a payment, do you often send your coins to another address, to another address, to another address, to another address, in a chain of hundreds or thousands? Are we serious? Do you want people to just sit and pretend that all this is somehow "natural growth" and that there is a "problem" when the majority of txs are junk - done by god knows who - and we should give him/them the tools to continue doing that at ever lower prices, and with more space to spam? If you have an attack vector, or a weakness, do you amplify it? Is this rational behavior? But anyway, this issue is ..."solved" because either classic or core, 1mb is dead and we go to 1.7 to 2. Question is, what should the minimum fee be at a given block size limit to have enough headroom for a consistent user experience? Recycling an arbitrary limit as a capacity quota is not a good answer.
Things like that aren't simple. Someone could counter-argue that the central planning of fees isn't much different than central banking policies. |
|
|
|
Suppose a bus company starts running a service in 2010. At first, you'd only see one or two people in the bus. As the bus service becomes more known, more people travel on the bus. Nowadays, busses are often completely occupied. It's common that people are left at the bus stops having to wait for the next one, esp. if they bought the cheapest tickets.
The difference between buses and blockchains is that buses can only take onboard real passengers, not ...fake ones. Blockchains can be burdened with millions of transactions just because one guy, that didn't even want to make one legitimate transaction, wants to have fun. I can imagine that a flood of tx without fee can be considered spam. But nowadays, only mining pools use zero fee tx to payout their miners. There are hardly any zero fee tx issued otherwise. (800 in the past 24 hours, source: https://bitcoinfees.21.co/ ) But if a tx carries a fee, how to decide if it's spam or not? What is your definition of spam? Someone "pays" for 1mb of spam with 1 satoshi per byte. Thus for occupying 1mb of space, he only needs 1 million satoshi = 0.01 btc = 4.2$. He can actually occupy the entire's day blockspace of 144mb by "paying" 604.8$. Even at 2mb, he can buy a block at 8$ and a day's worth of blocks at 1200$. Does this mean that he is not a spammer because he "paid" a ridiculously low amount of money? Fact: the number if tx per day is close to the limit of 250,000. We recently touched that twice. (sources: https://blockchain.info/charts/n-transactions?timespan=2year&showDataPoints=false&daysAverageString=1&show_header=true&scale=0&address= , http://www.coindesk.com/data/bitcoin-daily-transactions/ ) We crossed the 20,000 tx/day in June 2012 We crossed the 50,000 tx/day in August 2013 We crossed the 100,000 tx/day in March 2015 We crossed the 200,000 tx/day in January 2016 And the hard limit is a little over 250,000... So, do you think 250,000 tx/day is sufficient for Bitcoin to be successful? Do you think 400,000 tx/day is sufficient by the end of this year, when SW is rolled out and most other software is updated to take advantage of it? Do you think 400,000 tx/day will be enough until LN comes into existence? Compare these: Number of transactions: https://blockchain.info/charts/n-transactionsvs Number of Transactions Excluding Chains Longer Than 10: https://blockchain.info/charts/n-transactions-excluding-chains-longer-than-10Number of Transactions Excluding Chains Longer Than 100: https://blockchain.info/charts/n-transactions-excluding-chains-longer-than-100Number of Transactions Excluding Chains Longer Than 1000: https://blockchain.info/charts/n-transactions-excluding-chains-longer-than-1000Number of Transactions Excluding Chains Longer Than 10,000: https://blockchain.info/charts/n-transactions-excluding-chains-longer-than-10000(long chains = moving money to next address => to next address => to next address => to next address => ...) |
|
|
|
Γελαει ο κοσμος... Εντωμεταξυ αν δεις το roadmap του classic: https://github.com/bitcoinclassic/documentation/blob/master/roadmap/roadmap2016.mdπρος τα κατω λεει Phase 3 (Q3-Q4)
Make the block size limit dynamic
Note: This phase will only happen when miners & companies confirm Phase 2 successfully addressed their blocksize concerns.
Use a variation of Stephen Pair’s/BitPay proposal. Validation cost of a block must be less than a small multiple of the average cost over the last difficulty adjustment period.
Simplified version of Segregated Witness from Core, when it is available....εντωμεταξυ στο testnet για το segregated witness το testnet εγινε fork οταν ετρεχαν παλιες και νεες εκδοσεις του segwit ταυτοχρονα - δλδ δε σηκωνει ασυμβατες εκδοσεις... ο αλλος θελει να αλλαξει το segwit για να δημιουργησει και επιπλεον ασταθεια στο δικτυο αχαχαχχα.... και φυσικα απ'τα 2mb ξαφνικα πηγε στο "ατελειωτα megabyte" - δλδ dynamic limits... πχ καποιος σπαμαρει με τη ψυχη του και παιρνει 20-30-40mb block οπου το δικτυο αναπροσαρμοζει το οριο προς τα πανω αναλογα με τη ζητηση του "spammer"... δλδ ανευ οριου πρακτικα. |
|
|
|
if there junk why are miners including them in blocks? to get the fees the TX themselve are probably all junk tho. Actually, at levels like 0.2 - 0.3 btc per block, fees too are junk yes. Hence the presence of 0 tx blocks. Personally I'm currently mining with -blockmaxsize=0. My connectivity is not that good so it'd take me a couple secs to upload a full block and then it would take some extra time to propagate, so I can't risk losing 25 btc for peanuts. oh ic you must be F2Pool I'm not a pool and my hashrate is tiny so there is no guarantee I'll find blocks. I'm actually conducting a quantum experiment where consciousness interacts with hashing in order to ...generate hashing collisions and do so spectacularly against the odds. So if the experiment succeeds (in a universe of infinite possibilities it has already succeeded) it would be a tragedy if I found a block and have it get orphaned because of my slow upload speed or other propagation and verification delays. Maxblock 0 it is then... |
|
|
|
if there junk why are miners including them in blocks? to get the fees the TX themselve are probably all junk tho. Actually, at levels like 0.2 - 0.3 btc per block, fees too are junk yes. Hence the presence of 0 tx blocks. Personally I'm currently mining with -blockmaxsize=0. My connectivity is not that good so it'd take me a couple secs to upload a full block and then it would take some extra time to propagate, so I can't risk losing 25 btc for peanuts. |
|
|
|
The amount of dust we should normally be associated with at least of few weeks of usage in normal everyday environment (non server room) and we don’t think that these devices are tested in a very dusty room.
There is no "normal everyday environment" in China. The smog they are immersed in is visible from space (!) and people are wearing masks just to do everyday tasks. (a google of "china smog" photos is very revealing) Ugh, that's horrible! I thought it was bad in the 70's downtown and east Valley but never ever anything close to that! Yuck! It gets "better": http://edition.cnn.com/2016/02/08/world/fresh-air-britain-china-bottles/ |
|
|
|
Core Node Vs Classic Node
More like this => Core node Classic node |
|
|
|
The amount of dust we should normally be associated with at least of few weeks of usage in normal everyday environment (non server room) and we don’t think that these devices are tested in a very dusty room.
There is no "normal everyday environment" in China. The smog they are immersed in is visible from space (!) and people are wearing masks just to do everyday tasks. (a google of "china smog" photos is very revealing) |
|
|
|
By Moore's Law, we can expect hardware speed to be 10 times faster in 5 years and 100 times faster in 10. Even if Bitcoin grows at crazy adoption rates, I think computer speeds will stay ahead of the number of transactions.
Did it increase by tenfold in 5 years? Not even close. Satoshi did not have the adequate data here. There is a lot of processing power being untapped right now. This is typically found in GPUs though: The max radeon single card (not a 2x) of 2008 was, the 4870 doing 1.2 TFLOP (single) / 0.24 TFLOP (double) / 115gb/s ram The max radeon single card (not a 2x) of 2015 was R9 Fury 9 doing: 8.6 TFLOP (single) / 0.53 TFLOP (double) / 512 gb/s ram ---- Data from: https://en.wikipedia.org/wiki/List_of_AMD_graphics_processing_units--- And two imgs from NVIDIA (2-3 y. old but anyway)   |
|
|
|
|
Show Posts
