No Bitcoin For You
13 messages, 14 to 26 May 2015. Started by Tom Harding.
Jim Phillips (5), Thy Shizzle (3), gabe appleton (2), Tom Harding, Ryan X. Charles, Mike Hearn
A recent post, which I cannot find after much effort, made an excellent point.
If capacity grows, fewer individuals would be able to run full nodes. Those individuals, like many already, would have to give up running a full-node wallet :(
That sounds bad, until you consider that the alternative is running a full node on the bitcoin 'settlement network', while massive numbers of people *give up any hope of directly owning bitcoin at all*.
If today's global payments are 100Ktps, and move to the Lightning Network, they will have to be consolidated by a factor of 25000:1 to fit into bitcoin's current 4tps capacity as a settlement network. You executing a personal transaction on that network will be about as likely as you personally conducting a $100 SWIFT transfer to yourself today. For current holders, just selling or spending will get very expensive!
Forcing block capacity to stay small, so that individuals can run full nodes, is precisely what will force bitcoin to become a backbone that is too expensive for individuals to use. I can't avoid the conclusion that Bitcoin has to scale, and we might as well be thinking about how.
There may be a an escape window. As current trends continue toward a landscape of billions of SPV wallets, it may still be possible for individuals collectively to make up the majority of the network, if more parts of the network itself rely on SPV-level security.
With SPV-level security, it might be possible to implement a scalable DHT-type network of nodes that collectively store and index the exhaustive and fast-growing corpus of transaction history, up to and including currently unconfirmed transactions. Each individual node could host a slice of the transaction set with a configurable size, let's say down to a few GB today.
Such a network would have the desirable property of being run by the community. Most transactions would be submitted to it, and like today's network, it would disseminate blocks (which would be rapidly torn apart and digested). Therefore miners and other full nodes would depend on it, which is rather critical as those nodes grow closer to data-center proportions.
I agree with this analysis. I'm not sure if we will increase 1 MB block size or not, but with a block size that small, it is all but impossible for most people on the planet to ever own even a single utxo.
At 7tps, how long would it take to give 1 utxo to all of the 7 billion people currently alive? It would take 1 billion seconds, or about 32 years.[1] So for all practical purposes, at 1 MB block size, far less than 1% of people will ever be able to own even a single satoshi. Unless those people are willing to wait around 30 years for their lightning network to settle, they will either not use bitcoin, or they will use a substitute (such as a parallel decentralized network, or a centralized service) that lacks the full trust-minimized security guarantees of the main bitcoin blockchain.
I can't speak for most people, but for me personally, the thing I care most about as an individual (besides being able to send bitcoin to and from anyone on the planet) is being able to validate the blockchain. With a pruning node, this means I need to download the blockchain one time (not store it), and maintain the utxo set. The utxo set is, roughly speaking, 30 bytes per utxo, and therefore, at one utxo per person, about 7*30 billion bytes, or 210 GB. That's very achievable on the hardware of today. Of course, some individuals or companies will have far more than one utxo. Estimating an average of ten utxos per person, that will be 2.1 TB. Also very achievable on the hardware of today.
I don't think every transaction in the world should be on the blockchain, but I think it should be able to handle (long-term) enough transactions that everyone can have their transactions settled on a timescale suitable for human life. 30 years is unsuitable, but 1 day would be pretty good. It would be great if I could send trillions of transactions per day on networks built on top of bitcoin, and have my transactions settle on the actual blockchain once per day. This means we would need to support about 1 utxo per person per day, or 7 billion transactions per day. That translates to about 81 thousand transactions per second [2], or approximately 10,000 times the current rate. That would be 10 GB per ten minutes, which is achievable on current hardware (albeit not yet inexpensively).
Using SPV security rather than pruning security makes the cost even lower. A person relying on SPV would not have to download every 10 GB block, but only their transactions (or a small superset of them), which is already being done - scaling to 7 billion people would not require that SPV nodes perform any more computation than they already do. Nonetheless, I think pruning should be considered the default minimum, since that it what is required to get the full trust-minimized security guarantees of the blockchain. And that requires 10 GB blocks (or thereabouts).
The number of people on the planet will also grow, perhaps to 14 billion people in the next few decades. However, the estimates here would still be roughly correct. 10 GB blocks, or approximately so, allows everyone in the world to have their transactions settled on the blockchain in a timely manner, whereas 1 MB blocks do not. And this is already achievable on current hardware. The most significant cost is bandwidth, but that will probably become substantially less expensive in the coming years, making it possible for everyone to inexpensively and securely send and receive bitcoin to anyone else, without having to use a parallel network with reduced security or rely on trusted third parties.
[1] 10^9 / 60 / 60 / 24 / 365 ~= 32.
[2] 7*10^9 / 24 / 60 / 60 ~= 81018
On 05/14/2015 08:22 AM, Tom Harding wrote:
Quoted text (53 lines)
> A recent post, which I cannot find after much effort, made an > excellent point. > > If capacity grows, fewer individuals would be able to run full > nodes. Those individuals, like many already, would have to give up > running a full-node wallet :( > > That sounds bad, until you consider that the alternative is running > a full node on the bitcoin 'settlement network', while massive > numbers of people *give up any hope of directly owning bitcoin at > all*. > > If today's global payments are 100Ktps, and move to the Lightning > Network, they will have to be consolidated by a factor of 25000:1 > to fit into bitcoin's current 4tps capacity as a settlement > network. You executing a personal transaction on that network will > be about as likely as you personally conducting a $100 SWIFT > transfer to yourself today. For current holders, just selling or > spending will get very expensive! > > Forcing block capacity to stay small, so that individuals can run > full nodes, is precisely what will force bitcoin to become a > backbone that is too expensive for individuals to use. I can't > avoid the conclusion that Bitcoin has to scale, and we might as > well be thinking about how. > > There may be a an escape window. As current trends continue toward > a landscape of billions of SPV wallets, it may still be possible > for individuals collectively to make up the majority of the > network, if more parts of the network itself rely on SPV-level > security. > > With SPV-level security, it might be possible to implement a > scalable DHT-type network of nodes that collectively store and > index the exhaustive and fast-growing corpus of transaction > history, up to and including currently unconfirmed transactions. > Each individual node could host a slice of the transaction set with > a configurable size, let's say down to a few GB today. > > Such a network would have the desirable property of being run by > the community. Most transactions would be submitted to it, and > like today's network, it would disseminate blocks (which would be > rapidly torn apart and digested). Therefore miners and other full > nodes would depend on it, which is rather critical as those nodes > grow closer to data-center proportions. > > > > ------------------------------------------------------------------------------ > > > >
One dashboard for servers and applications across Physical-Virtual-Cloud
Quoted text (8 lines)
> Widest out-of-the-box monitoring support with 50+ applications > Performance metrics, stats and reports that give you Actionable > Insights Deep dive visibility with transaction tracing using APM > Insight. http://ad.doubleclick.net/ddm/clk/290420510;117567292;y > _______________________________________________ Bitcoin-development > mailing list Bitcoin-development@lists•sourceforge•net > https://lists.sourceforge.net/lists/listinfo/bitcoin-development >
Quoted text (3 lines)
> > If capacity grows, fewer individuals would be able to run full nodes. >
Hardly. Nobody is currently exhausting the CPU capacity of even a normal computer currently and even if we did a 20x increase in load overnight, that still wouldn't even warm up most machines good enough to be always on.
The reasons full nodes are unpopular to run seem to be:
1. Uncontrollable bandwidth usage from sending people the chain 2. People don't run them all the time, then don't want to wait for them to catch up
The first can be fixed with better code (you can already easily opt out of uploading the chain, it's just not as fine-grained as desirable), and the second is fundamental to what full nodes do and how people work. For merchants, who are the most important demographic we want to be using full nodes, they can just keep it running all the time. No biggie.
Quoted text (4 lines)
> Therefore miners and other full nodes would depend on > it, which is rather critical as those nodes grow closer to data-center > proportions. >
This meme about datacenter-sized nodes has to die. The Bitcoin wiki is down right now, but I showed years ago that you could keep up with VISA on a single well specced server with today's technology. Only people living in a dreamworld think that Bitcoin might actually have to match that level of transaction demand with today's hardware. As noted previously, "too many users" is simply not a problem Bitcoin has .... and may never have!
On Mon, May 25, 2015 at 1:36 PM, Mike Hearn <mike@plan99•net> wrote:
This meme about datacenter-sized nodes has to die. The Bitcoin wiki is down
Quoted text (7 lines)
> right now, but I showed years ago that you could keep up with VISA on a > single well specced server with today's technology. Only people living in a > dreamworld think that Bitcoin might actually have to match that level of > transaction demand with today's hardware. As noted previously, "too many > users" is simply not a problem Bitcoin has .... and may never have! > >
... And will certainly NEVER have if we can't solve the capacity problem SOON.
In a former life, I was a capacity planner for Bank of America's mid-range server group. We had one hard and fast rule. When you are typically exceeding 75% of capacity on a given metric, it's time to expand capacity. Period. You don't do silly things like adjusting the business model to disincentivize use. Unless there's some flaw in the system and it's leaking resources, if usage has increased to the point where you are at or near the limits of capacity, you expand capacity. It's as simple as that, and I've found that same rule fits quite well in a number of systems.
In Bitcoin, we're not leaking resources. There's no flaw. The system is performing as intended. Usage is increasing because it works so well, and there is huge potential for future growth as we identify more uses and attract more users. There might be a few technical things we can do to reduce consumption, but the metric we're concerned with right now is how many transactions we can fit in a block. We've broken through the 75% marker and are regularly bumping up against the 100% limit.
It is time to stop debating this and take action to expand capacity. The only questions that should remain are how much capacity do we add, and how soon can we do it. Given that most existing computer systems and networks can easily handle 20MB blocks every 10 minutes, and given that that will increase capacity 20-fold, I can't think of a single reason why we can't go to 20MB as soon as humanly possible. And in a few years, when the average block size is over 15MB, we bump it up again to as high as we can go then without pushing typical computers or networks beyond their capacity. We can worry about ways to slow down growth without affecting the usefulness of Bitcoin as we get closer to the hard technical limits on our capacity.
And you know what else? If miners need higher fees to accommodate the costs of bigger blocks, they can configure their nodes to only mine transactions with higher fees.. Let the miners decide how to charge enough to pay for their costs. We don't need to cripple the network just for them.
Nah don't make blocks 20mb, then you are slowing down block propagation and blowing out conf tikes as a result. Just decrease the time it takes to make a 1mb block, then you still see the same propagation times today and just increase the transaction throughput. ________________________________ From: Jim Phillips<mailto:jim@ergophobia.org> Sent: 26/05/2015 12:27 PM To: Mike Hearn<mailto:mike@plan99.net> Cc: Bitcoin Dev<mailto:bitcoin-development@lists.sourceforge.net> Subject: Re: [Bitcoin-development] No Bitcoin For You
On Mon, May 25, 2015 at 1:36 PM, Mike Hearn <mike@plan99•net> wrote:
This meme about datacenter-sized nodes has to die. The Bitcoin wiki is down
Quoted text (7 lines)
> right now, but I showed years ago that you could keep up with VISA on a > single well specced server with today's technology. Only people living in a > dreamworld think that Bitcoin might actually have to match that level of > transaction demand with today's hardware. As noted previously, "too many > users" is simply not a problem Bitcoin has .... and may never have! > >
... And will certainly NEVER have if we can't solve the capacity problem SOON.
In a former life, I was a capacity planner for Bank of America's mid-range server group. We had one hard and fast rule. When you are typically exceeding 75% of capacity on a given metric, it's time to expand capacity. Period. You don't do silly things like adjusting the business model to disincentivize use. Unless there's some flaw in the system and it's leaking resources, if usage has increased to the point where you are at or near the limits of capacity, you expand capacity. It's as simple as that, and I've found that same rule fits quite well in a number of systems.
In Bitcoin, we're not leaking resources. There's no flaw. The system is performing as intended. Usage is increasing because it works so well, and there is huge potential for future growth as we identify more uses and attract more users. There might be a few technical things we can do to reduce consumption, but the metric we're concerned with right now is how many transactions we can fit in a block. We've broken through the 75% marker and are regularly bumping up against the 100% limit.
It is time to stop debating this and take action to expand capacity. The only questions that should remain are how much capacity do we add, and how soon can we do it. Given that most existing computer systems and networks can easily handle 20MB blocks every 10 minutes, and given that that will increase capacity 20-fold, I can't think of a single reason why we can't go to 20MB as soon as humanly possible. And in a few years, when the average block size is over 15MB, we bump it up again to as high as we can go then without pushing typical computers or networks beyond their capacity. We can worry about ways to slow down growth without affecting the usefulness of Bitcoin as we get closer to the hard technical limits on our capacity.
And you know what else? If miners need higher fees to accommodate the costs of bigger blocks, they can configure their nodes to only mine transactions with higher fees.. Let the miners decide how to charge enough to pay for their costs. We don't need to cripple the network just for them.
But don't you see the same trade-off in the end there? You're still propagating the same amount of data over the same amount of time, so unless I misunderstand, the costs of such a move should be approximately the same, just in different areas. The risks as I understand are as follows:
20MB:
1. Longer per-block propagation (eventually) 2. Longer processing time (eventually) 3. Longer sync time
1 Minute:
1. Weaker individual confirmations (approx. equal per confirmation*time) 2. Higher orphan rate (immediately) 3. Longer sync time
That risk-set makes me want a middle-ground approach. Something where the immediate consequences aren't all that strong, and where we have some idea of what to do in the future. Is there any chance we can get decent network simulations at various configurations (5MB/4min, etc)? Perhaps re-appropriate the testnet?
On Mon, May 25, 2015 at 10:30 PM, Thy Shizzle <thyshizzle@outlook•com> wrote:
Quoted text (82 lines)
> Nah don't make blocks 20mb, then you are slowing down block propagation > and blowing out conf tikes as a result. Just decrease the time it takes to > make a 1mb block, then you still see the same propagation times today and > just increase the transaction throughput. > ------------------------------ > From: Jim Phillips <jim@ergophobia•org> > Sent: 26/05/2015 12:27 PM > To: Mike Hearn <mike@plan99•net> > Cc: Bitcoin Dev <bitcoin-development@lists•sourceforge•net> > Subject: Re: [Bitcoin-development] No Bitcoin For You > > > On Mon, May 25, 2015 at 1:36 PM, Mike Hearn <mike@plan99•net> wrote: > > This meme about datacenter-sized nodes has to die. The Bitcoin wiki is > down right now, but I showed years ago that you could keep up with VISA on > a single well specced server with today's technology. Only people living in > a dreamworld think that Bitcoin might actually have to match that level of > transaction demand with today's hardware. As noted previously, "too many > users" is simply not a problem Bitcoin has .... and may never have! > > > ... And will certainly NEVER have if we can't solve the capacity problem > SOON. > > In a former life, I was a capacity planner for Bank of America's > mid-range server group. We had one hard and fast rule. When you are > typically exceeding 75% of capacity on a given metric, it's time to expand > capacity. Period. You don't do silly things like adjusting the business > model to disincentivize use. Unless there's some flaw in the system and > it's leaking resources, if usage has increased to the point where you are > at or near the limits of capacity, you expand capacity. It's as simple as > that, and I've found that same rule fits quite well in a number of systems. > > In Bitcoin, we're not leaking resources. There's no flaw. The system is > performing as intended. Usage is increasing because it works so well, and > there is huge potential for future growth as we identify more uses and > attract more users. There might be a few technical things we can do to > reduce consumption, but the metric we're concerned with right now is how > many transactions we can fit in a block. We've broken through the 75% > marker and are regularly bumping up against the 100% limit. > > It is time to stop debating this and take action to expand capacity. The > only questions that should remain are how much capacity do we add, and how > soon can we do it. Given that most existing computer systems and networks > can easily handle 20MB blocks every 10 minutes, and given that that will > increase capacity 20-fold, I can't think of a single reason why we can't go > to 20MB as soon as humanly possible. And in a few years, when the average > block size is over 15MB, we bump it up again to as high as we can go then > without pushing typical computers or networks beyond their capacity. We can > worry about ways to slow down growth without affecting the usefulness of > Bitcoin as we get closer to the hard technical limits on our capacity. > > And you know what else? If miners need higher fees to accommodate the > costs of bigger blocks, they can configure their nodes to only mine > transactions with higher fees.. Let the miners decide how to charge enough > to pay for their costs. We don't need to cripple the network just for them. > > -- > *James G. Phillips IV* > <https://plus.google.com/u/0/113107039501292625391/posts> > > *"Don't bunt. Aim out of the ball park. Aim for the company of immortals." > -- David Ogilvy * > > *This message was created with 100% recycled electrons. Please think > twice before printing.* > > > > ------------------------------------------------------------------------------ > One dashboard for servers and applications across Physical-Virtual-Cloud > Widest out-of-the-box monitoring support with 50+ applications > Performance metrics, stats and reports that give you Actionable Insights > Deep dive visibility with transaction tracing using APM Insight. > http://ad.doubleclick.net/ddm/clk/290420510;117567292;y > _______________________________________________ > Bitcoin-development mailing list > Bitcoin-development@lists•sourceforge•net > https://lists.sourceforge.net/lists/listinfo/bitcoin-development > >
I wouldn't say same trade-off because you need the whole 20mb block before you can start to use it where as a 1mb block can be used quicker thus transactions found in tge block quicker etc. As for tge higher rate of orphans, I think this would be complimented by a faster correction rate, so if you're pumping out blocks at a rate of 1 per minute, if we get a fork and the next block comes in 10 minutes and is the decider, it took 10 minutes to determine which block is the orphan. But at a rate of 1 block per 1 minute then it only takes 1 minute to resolve the orphan (obviously this is very simplified) so I'm not so sure that orphan rate is a big issue here. Indeed you would need to draw upon more confirmations for easier block creation but surely that is not an issue?
Why would sync time be longer as opposed to 20mb blocks? ________________________________ From: gabe appleton<mailto:gappleto97@gmail.com> Sent: 26/05/2015 12:41 PM To: Thy Shizzle<mailto:thyshizzle@outlook.com> Cc: Jim Phillips<mailto:jim@ergophobia.org>; Mike Hearn<mailto:mike@plan99.net>; Bitcoin Dev<mailto:bitcoin-development@lists.sourceforge.net> Subject: Re: [Bitcoin-development] No Bitcoin For You
But don't you see the same trade-off in the end there? You're still propagating the same amount of data over the same amount of time, so unless I misunderstand, the costs of such a move should be approximately the same, just in different areas. The risks as I understand are as follows:
20MB:
1. Longer per-block propagation (eventually) 2. Longer processing time (eventually) 3. Longer sync time
1 Minute:
1. Weaker individual confirmations (approx. equal per confirmation*time) 2. Higher orphan rate (immediately) 3. Longer sync time
That risk-set makes me want a middle-ground approach. Something where the immediate consequences aren't all that strong, and where we have some idea of what to do in the future. Is there any chance we can get decent network simulations at various configurations (5MB/4min, etc)? Perhaps re-appropriate the testnet?
On Mon, May 25, 2015 at 10:30 PM, Thy Shizzle <thyshizzle@outlook•com> wrote:
Quoted text (82 lines)
> Nah don't make blocks 20mb, then you are slowing down block propagation > and blowing out conf tikes as a result. Just decrease the time it takes to > make a 1mb block, then you still see the same propagation times today and > just increase the transaction throughput. > ------------------------------ > From: Jim Phillips <jim@ergophobia•org> > Sent: 26/05/2015 12:27 PM > To: Mike Hearn <mike@plan99•net> > Cc: Bitcoin Dev <bitcoin-development@lists•sourceforge•net> > Subject: Re: [Bitcoin-development] No Bitcoin For You > > > On Mon, May 25, 2015 at 1:36 PM, Mike Hearn <mike@plan99•net> wrote: > > This meme about datacenter-sized nodes has to die. The Bitcoin wiki is > down right now, but I showed years ago that you could keep up with VISA on > a single well specced server with today's technology. Only people living in > a dreamworld think that Bitcoin might actually have to match that level of > transaction demand with today's hardware. As noted previously, "too many > users" is simply not a problem Bitcoin has .... and may never have! > > > ... And will certainly NEVER have if we can't solve the capacity problem > SOON. > > In a former life, I was a capacity planner for Bank of America's > mid-range server group. We had one hard and fast rule. When you are > typically exceeding 75% of capacity on a given metric, it's time to expand > capacity. Period. You don't do silly things like adjusting the business > model to disincentivize use. Unless there's some flaw in the system and > it's leaking resources, if usage has increased to the point where you are > at or near the limits of capacity, you expand capacity. It's as simple as > that, and I've found that same rule fits quite well in a number of systems. > > In Bitcoin, we're not leaking resources. There's no flaw. The system is > performing as intended. Usage is increasing because it works so well, and > there is huge potential for future growth as we identify more uses and > attract more users. There might be a few technical things we can do to > reduce consumption, but the metric we're concerned with right now is how > many transactions we can fit in a block. We've broken through the 75% > marker and are regularly bumping up against the 100% limit. > > It is time to stop debating this and take action to expand capacity. The > only questions that should remain are how much capacity do we add, and how > soon can we do it. Given that most existing computer systems and networks > can easily handle 20MB blocks every 10 minutes, and given that that will > increase capacity 20-fold, I can't think of a single reason why we can't go > to 20MB as soon as humanly possible. And in a few years, when the average > block size is over 15MB, we bump it up again to as high as we can go then > without pushing typical computers or networks beyond their capacity. We can > worry about ways to slow down growth without affecting the usefulness of > Bitcoin as we get closer to the hard technical limits on our capacity. > > And you know what else? If miners need higher fees to accommodate the > costs of bigger blocks, they can configure their nodes to only mine > transactions with higher fees.. Let the miners decide how to charge enough > to pay for their costs. We don't need to cripple the network just for them. > > -- > *James G. Phillips IV* > <https://plus.google.com/u/0/113107039501292625391/posts> > > *"Don't bunt. Aim out of the ball park. Aim for the company of immortals." > -- David Ogilvy * > > *This message was created with 100% recycled electrons. Please think > twice before printing.* > > > > ------------------------------------------------------------------------------ > One dashboard for servers and applications across Physical-Virtual-Cloud > Widest out-of-the-box monitoring support with 50+ applications > Performance metrics, stats and reports that give you Actionable Insights > Deep dive visibility with transaction tracing using APM Insight. > http://ad.doubleclick.net/ddm/clk/290420510;117567292;y > _______________________________________________ > Bitcoin-development mailing list > Bitcoin-development@lists•sourceforge•net > https://lists.sourceforge.net/lists/listinfo/bitcoin-development > >
Frankly I'm good with either way. I'm definitely in favor of faster confirmation times.
The important thing is that we need to increase the amount of transactions that get into blocks over a given time frame to a point that is in line with what current technology can handle. We can handle WAY more than we are doing right now. The Bitcoin network is not currently Disk, CPU, or RAM bound.. Not even close. The metric we're closest to being restricted by would be Network bandwidth. I live in a developing country. 2Mbps is a typical broadband speed here (although 5Mbps and 10Mbps connections are affordable). That equates to about 17MB per minute, or 170x more capacity than what I need to receive a full copy of the blockchain if I only talk to one peer. If I relay to say 10 peers, I can still handle 17x larger block sizes on a slow 2Mbps connection.
Also, even if we reduce the difficulty so that we're doing 1MB blocks every minute, that's still only 10MB every 10 minutes. Eventually we're going to have to increase that, and we can only reduce the confirmation period so much. I think someone once said 30 seconds or so is about the shortest period you can practically achieve.
Indeed Jim, your internet connection makes a good reason why I don't like 20mb blocks (right now). It would take you well over a minute to download the block before you could even relay it on, so much slow down in propagation! Yes I do see how decreasing the time to create blocks is a bit of a band-aid fix, and to use tge term I've seen mentioned here "kicking the can down the road" I agree that this is doing this, however as you say bandwidth is our biggest enemy right now and so hopefully by the time we exceed the capacity gained by the decrease in block time, we can then look to bump up block size because hopefully 20mbps connections will be baseline by then etc. ________________________________ From: Jim Phillips<mailto:jim@ergophobia.org> Sent: 26/05/2015 12:53 PM To: Thy Shizzle<mailto:thyshizzle@outlook.com> Cc: Mike Hearn<mailto:mike@plan99.net>; Bitcoin Dev<mailto:bitcoin-development@lists.sourceforge.net> Subject: Re: [Bitcoin-development] No Bitcoin For You
Frankly I'm good with either way. I'm definitely in favor of faster confirmation times.
The important thing is that we need to increase the amount of transactions that get into blocks over a given time frame to a point that is in line with what current technology can handle. We can handle WAY more than we are doing right now. The Bitcoin network is not currently Disk, CPU, or RAM bound.. Not even close. The metric we're closest to being restricted by would be Network bandwidth. I live in a developing country. 2Mbps is a typical broadband speed here (although 5Mbps and 10Mbps connections are affordable). That equates to about 17MB per minute, or 170x more capacity than what I need to receive a full copy of the blockchain if I only talk to one peer. If I relay to say 10 peers, I can still handle 17x larger block sizes on a slow 2Mbps connection.
Also, even if we reduce the difficulty so that we're doing 1MB blocks every minute, that's still only 10MB every 10 minutes. Eventually we're going to have to increase that, and we can only reduce the confirmation period so much. I think someone once said 30 seconds or so is about the shortest period you can practically achieve.
I don't see how the fact that my 2Mbps connection causes me to not be a very good relay has any bearing on whether or not the network as a whole would be negatively impacted by a 20MB block. My inability to rapidly propagate blocks doesn't really harm the network. It's only if MOST relays are as slow as mine that it creates an issue. I'm one node in thousands (potentially tens or hundreds of thousands if/when Bitcoin goes mainstream). And I'm an individual. There's no reason at all for me to run a full node from my home, except to have my own trusted and validated copy of the blockchain on a computer I control directly. I don't need to act as a relay for that and as long as I can download blocks faster than they are created I'm fine. Also, I can easily afford a VPS server or several to run full nodes as relays if I am feeling altruistic. It's actually cheaper for me to lease a VPS than to keep my own home PC on 24/7, which is why I have 2 of them.
And as a business, the cost of a server and bandwidth to run a full node is a drop in the bucket. I'm involved in several projects where we have full nodes running on leased servers with multiple 1Gbps connections. It's an almost zero cost. Those nodes could handle 20MB blocks today without thinking about it, and I'm sure our nodes are just a few amongst thousands just like them. I'm not at all concerned about the network being too centralized.
What concerns me is the fact that we are using edge cases like my home PC as a lame excuse to debate expanding the capacity of the network.
Incidentally, even once we have the "Internet of Things" brought on by 21, Inc. or whoever beats them to it, I would expect the average home to have only a single full node "hub" receiving the blockchain and broadcasting transactions created by all the minor SPV connected devices running within the house. The in-home full node would be peered with high bandwidth full-node relays running at the ISP or in the cloud. There are more than enough ISPs and cloud compute providers in the world such that there should be no concern at all about centralization of relays. Full nodes could some day become as ubiquitous on the Internet as authoritative DNS servers. And just like DNS servers, if you don't trust the nodes your ISP creates or it's too slow or censors transactions, there's nothing preventing you from peering with nodes hosted by the Googles or OpenDNSs out there, or running your own if you're really paranoid and have a few extra bucks for a VPS.
Sync time wouldn't be longer compared to 20MB, it would (eventually) be longer under either setup.
Also, and this is probably a silly concern, but wouldn't changing block time change the supply curve? If we cut the rate in half or a power of two, that affects nothing, but if we want to keep it in round numbers, we need to do it by 10, 5, or 2. I feel like most people would bank for 10 or 5, both of which change the supply curve due to truncation.
Again, it's a trivial concern, but probably one that should be addressed. On May 25, 2015 11:52 PM, "Jim Phillips" <jim@ergophobia•org> wrote:
Quoted text (207 lines)
> Incidentally, even once we have the "Internet of Things" brought on by 21, > Inc. or whoever beats them to it, I would expect the average home to have > only a single full node "hub" receiving the blockchain and broadcasting > transactions created by all the minor SPV connected devices running within > the house. The in-home full node would be peered with high bandwidth > full-node relays running at the ISP or in the cloud. There are more than > enough ISPs and cloud compute providers in the world such that there should > be no concern at all about centralization of relays. Full nodes could some > day become as ubiquitous on the Internet as authoritative DNS servers. And > just like DNS servers, if you don't trust the nodes your ISP creates or > it's too slow or censors transactions, there's nothing preventing you from > peering with nodes hosted by the Googles or OpenDNSs out there, or running > your own if you're really paranoid and have a few extra bucks for a VPS. > > -- > *James G. Phillips IV* > <https://plus.google.com/u/0/113107039501292625391/posts> > <http://www.linkedin.com/in/ergophobe> > > *"Don't bunt. Aim out of the ball park. Aim for the company of immortals." > -- David Ogilvy* > > *This message was created with 100% recycled electrons. Please think > twice before printing.* > > On Mon, May 25, 2015 at 10:23 PM, Jim Phillips <jim@ergophobia•org> wrote: > >> I don't see how the fact that my 2Mbps connection causes me to not be a >> very good relay has any bearing on whether or not the network as a whole >> would be negatively impacted by a 20MB block. My inability to rapidly >> propagate blocks doesn't really harm the network. It's only if MOST relays >> are as slow as mine that it creates an issue. I'm one node in thousands >> (potentially tens or hundreds of thousands if/when Bitcoin goes >> mainstream). And I'm an individual. There's no reason at all for me to run >> a full node from my home, except to have my own trusted and validated copy >> of the blockchain on a computer I control directly. I don't need to act as >> a relay for that and as long as I can download blocks faster than they are >> created I'm fine. Also, I can easily afford a VPS server or several to run >> full nodes as relays if I am feeling altruistic. It's actually cheaper for >> me to lease a VPS than to keep my own home PC on 24/7, which is why I have >> 2 of them. >> >> And as a business, the cost of a server and bandwidth to run a full node >> is a drop in the bucket. I'm involved in several projects where we have >> full nodes running on leased servers with multiple 1Gbps connections. It's >> an almost zero cost. Those nodes could handle 20MB blocks today without >> thinking about it, and I'm sure our nodes are just a few amongst thousands >> just like them. I'm not at all concerned about the network being too >> centralized. >> >> What concerns me is the fact that we are using edge cases like my home PC >> as a lame excuse to debate expanding the capacity of the network. >> >> -- >> *James G. Phillips IV* >> <https://plus.google.com/u/0/113107039501292625391/posts> >> <http://www.linkedin.com/in/ergophobe> >> >> *"Don't bunt. Aim out of the ball park. Aim for the company of >> immortals." -- David Ogilvy* >> >> *This message was created with 100% recycled electrons. Please think >> twice before printing.* >> >> On Mon, May 25, 2015 at 10:02 PM, Thy Shizzle <thyshizzle@outlook•com> >> wrote: >> >>> Indeed Jim, your internet connection makes a good reason why I don't >>> like 20mb blocks (right now). It would take you well over a minute to >>> download the block before you could even relay it on, so much slow down in >>> propagation! Yes I do see how decreasing the time to create blocks is a bit >>> of a band-aid fix, and to use tge term I've seen mentioned here "kicking >>> the can down the road" I agree that this is doing this, however as you say >>> bandwidth is our biggest enemy right now and so hopefully by the time we >>> exceed the capacity gained by the decrease in block time, we can then look >>> to bump up block size because hopefully 20mbps connections will be baseline >>> by then etc. >>> ------------------------------ >>> From: Jim Phillips <jim@ergophobia•org> >>> Sent: 26/05/2015 12:53 PM >>> To: Thy Shizzle <thyshizzle@outlook•com> >>> Cc: Mike Hearn <mike@plan99•net>; Bitcoin Dev >>> <bitcoin-development@lists•sourceforge•net> >>> >>> Subject: Re: [Bitcoin-development] No Bitcoin For You >>> >>> Frankly I'm good with either way. I'm definitely in favor of faster >>> confirmation times. >>> >>> The important thing is that we need to increase the amount of >>> transactions that get into blocks over a given time frame to a point that >>> is in line with what current technology can handle. We can handle WAY more >>> than we are doing right now. The Bitcoin network is not currently Disk, >>> CPU, or RAM bound.. Not even close. The metric we're closest to being >>> restricted by would be Network bandwidth. I live in a developing country. >>> 2Mbps is a typical broadband speed here (although 5Mbps and 10Mbps >>> connections are affordable). That equates to about 17MB per minute, or 170x >>> more capacity than what I need to receive a full copy of the blockchain if >>> I only talk to one peer. If I relay to say 10 peers, I can still handle 17x >>> larger block sizes on a slow 2Mbps connection. >>> >>> Also, even if we reduce the difficulty so that we're doing 1MB blocks >>> every minute, that's still only 10MB every 10 minutes. Eventually we're >>> going to have to increase that, and we can only reduce the confirmation >>> period so much. I think someone once said 30 seconds or so is about the >>> shortest period you can practically achieve. >>> >>> -- >>> *James G. Phillips IV* >>> <https://plus.google.com/u/0/113107039501292625391/posts> >>> <http://www.linkedin.com/in/ergophobe> >>> >>> *"Don't bunt. Aim out of the ball park. Aim for the company of >>> immortals." -- David Ogilvy * >>> >>> *This message was created with 100% recycled electrons. Please think >>> twice before printing.* >>> >>> On Mon, May 25, 2015 at 9:30 PM, Thy Shizzle <thyshizzle@outlook•com> >>> wrote: >>> >>> Nah don't make blocks 20mb, then you are slowing down block >>> propagation and blowing out conf tikes as a result. Just decrease the time >>> it takes to make a 1mb block, then you still see the same propagation times >>> today and just increase the transaction throughput. >>> ------------------------------ >>> From: Jim Phillips <jim@ergophobia•org> >>> Sent: 26/05/2015 12:27 PM >>> To: Mike Hearn <mike@plan99•net> >>> Cc: Bitcoin Dev <bitcoin-development@lists•sourceforge•net> >>> Subject: Re: [Bitcoin-development] No Bitcoin For You >>> >>> >>> On Mon, May 25, 2015 at 1:36 PM, Mike Hearn <mike@plan99•net> wrote: >>> >>> This meme about datacenter-sized nodes has to die. The Bitcoin wiki >>> is down right now, but I showed years ago that you could keep up with VISA >>> on a single well specced server with today's technology. Only people living >>> in a dreamworld think that Bitcoin might actually have to match that level >>> of transaction demand with today's hardware. As noted previously, "too many >>> users" is simply not a problem Bitcoin has .... and may never have! >>> >>> >>> ... And will certainly NEVER have if we can't solve the capacity >>> problem SOON. >>> >>> In a former life, I was a capacity planner for Bank of America's >>> mid-range server group. We had one hard and fast rule. When you are >>> typically exceeding 75% of capacity on a given metric, it's time to expand >>> capacity. Period. You don't do silly things like adjusting the business >>> model to disincentivize use. Unless there's some flaw in the system and >>> it's leaking resources, if usage has increased to the point where you are >>> at or near the limits of capacity, you expand capacity. It's as simple as >>> that, and I've found that same rule fits quite well in a number of systems. >>> >>> In Bitcoin, we're not leaking resources. There's no flaw. The system >>> is performing as intended. Usage is increasing because it works so well, >>> and there is huge potential for future growth as we identify more uses and >>> attract more users. There might be a few technical things we can do to >>> reduce consumption, but the metric we're concerned with right now is how >>> many transactions we can fit in a block. We've broken through the 75% >>> marker and are regularly bumping up against the 100% limit. >>> >>> It is time to stop debating this and take action to expand capacity. >>> The only questions that should remain are how much capacity do we add, and >>> how soon can we do it. Given that most existing computer systems and >>> networks can easily handle 20MB blocks every 10 minutes, and given that >>> that will increase capacity 20-fold, I can't think of a single reason why >>> we can't go to 20MB as soon as humanly possible. And in a few years, when >>> the average block size is over 15MB, we bump it up again to as high as we >>> can go then without pushing typical computers or networks beyond their >>> capacity. We can worry about ways to slow down growth without affecting the >>> usefulness of Bitcoin as we get closer to the hard technical limits on our >>> capacity. >>> >>> And you know what else? If miners need higher fees to accommodate the >>> costs of bigger blocks, they can configure their nodes to only mine >>> transactions with higher fees.. Let the miners decide how to charge enough >>> to pay for their costs. We don't need to cripple the network just for them. >>> >>> -- >>> *James G. Phillips IV* >>> <https://plus.google.com/u/0/113107039501292625391/posts> >>> >>> *"Don't bunt. Aim out of the ball park. Aim for the company of >>> immortals." -- David Ogilvy * >>> >>> *This message was created with 100% recycled electrons. Please think >>> twice before printing.* >>> >>> >>> >> > > > ------------------------------------------------------------------------------ > One dashboard for servers and applications across Physical-Virtual-Cloud > Widest out-of-the-box monitoring support with 50+ applications > Performance metrics, stats and reports that give you Actionable Insights > Deep dive visibility with transaction tracing using APM Insight. > http://ad.doubleclick.net/ddm/clk/290420510;117567292;y > _______________________________________________ > Bitcoin-development mailing list > Bitcoin-development@lists•sourceforge•net > https://lists.sourceforge.net/lists/listinfo/bitcoin-development > >
I think all the suggestions recommending cutting the block time down also suggest reducing the rewards to compensate.