Why the Bitcoin rules can't change (reading time ~5min)

[tvbcof]

For my part, I feel more comfortable with a solution in which decisions are made in a transparent manner even if it means that people who have different points of view than my own have a voice.  I think it is possible to hold a variety of competing points of view without being 'dishonest'.

Hiding or discounting one point of view and decision making processes would create the most doubt in my mind.

Every one of your posts exudes such a slimy malignancy that I want to take a shower after reading them.

Well as I always say "If it feels good, do it."

Oh, and don't forget about the 'ignore' button over on the left there.

[acoindr]

Since Hazek isn't very explicit, ill summarise his point.

Hazek wants to ensure that he can run a full node with whatever hardware/network he posses at that time without having to fork out additional funds or make special group funding arrangements to host a node in a datacenter.

Personally i can accept a Block size increase as long as a low end commodity PC with an average broadband connection can run as a full node(Should cover 90% of people with PCs). Note that miners will need a much faster connection than full nodes to reduce orphan rates + investment in ASICs.

CPU wise, this is not a problem today, nor will it ever be. You could run a full node on your average low-end smartphone CPU. Even if we were to scale block size by 100x, your average CPU could handle it(Not for mining, as speed is paramount in mining). Plus if need be, much of the intensive processing could be offloaded to the GPU.

Disk Space wise, an increase of block size by 100x would not be an issue once pruning is implemented. Big Bitcoin businesses and block chain explorers could run archival nodes which provide all blocks.

Network wise,  256Kbps could keep up with a 10MB block size if it were running 24/7. For a 1Mbps connection, you would need to run it 6hrs a day.

This.

Thank you! That's what I wanted to see, some explicit numbers and general conclusions, including compartmentalizing issues.

As we can see the network connection is the biggest hindrance to increasing block size, but a 10MB block size would still allow the majority of people to run full nodes. The biggest problems will be faced by miners who will not be able to compete with the bigger miners/pools due to connection speed. Which after some thinking, is fine as long as we have people running full nodes to "vote"(not relaying) against any predatory changes that large mining pools may introduce.

Yes, network connection is the one potential sticking point I see for home-based full nodes too.

[ldrgn]

Every one of your posts exudes such a slimy malignancy that I want to take a shower after reading them.

Project much?

[bg002h]

...if the bitcoin rules don't change, Bitcoin will be renamed to UselesslySlowCoin, and all the dolts thinking they're gonna cash in on transaction fees for UselesslySlowCoin will realize they have killed their cash cow via artificial scarcity...

So much for the revolutionary thinking that makes Bitcoin Bitcoin...I feel rather harsh disdain for those who wish to kill society's chance at having an open and robust payment network that can actually replace banks because they think they scan squeeze a few more dollars out of someone. A limit of 7 transactions per second will mean BITCOIN WILL NEVER REPLACE BANKS!  In fact, it will never replace PayPal.

[markm]

I thought we'd got that. Is there anyone who still didn't?

Since then its been more about how fast to raise it than about whether to raise it,

plus me saying any increase should never be undone (barring global catastrophe).

(Thus, no adaptive reduction; adapt up if you must, but down is a no-no.)

(Thus be careful with increase: never increase more than you are ready and willing to handle 24/7 every block thereafter forever.)

-MarkM-

[tvbcof]

...if the bitcoin rules don't change, Bitcoin will be renamed to UselesslySlowCoin, and all the dolts thinking they're gonna cash in on transaction fees for UselesslySlowCoin will realize they have killed their cash cow via artificial scarcity...

I don't know about others, but I don't stand to make a dime on whatever transaction fees and what-not you are talking about.  I'll only mine if I feel the system is at risk, and in that case I expect to absorb the cost out-of-pocket.

I do stand to lose quite a bit monetarily if the system is subverted and I do care about that both for personal and for philosophical reasons.

So much for the revolutionary thinking that makes Bitcoin Bitcoin...I feel rather harsh disdain for those who wish to kill society's chance at having an open and robust payment network that can actually replace banks because they think they scan squeeze a few more dollars out of someone. A limit of 7 transactions per second will mean BITCOIN WILL NEVER REPLACE BANKS!  In fact, it will never replace PayPal.

Firstly, anyone thinking about 'replacing the banks' and 'replacing PayPal' better come prepared.  I think that most in the Bitcoin community are anything but prepared for what lies down that road.  We are likely starting to appear on the distant horizon for these folks, and certain other tangential threats are starting to appear.  This is why I care so deeply about the subject at this time.

Secondly, trying to compete head-on is, in my opinion, not the way to do it.  I'll bet most of my BTC (literally) that we end up smashed like bug if we try, and not only that but there will be massive collateral damage as well.

Thirdly, I have no interest in Bitcoin replacing VISA or PayPal if the results are even less desirable that what these folks already supply.  That also strikes me as a very real possibility.

[westkybitcoins]

Some of my view:

Actually many people here are just hoarding coins, there are not really a lot of transactions happening everyday, especially miners who connected to a pool. If we remove satoshi-dice, I think at least in the latest 2 years it will still work fine. Satoshi-dice is just an execellent example of no matter how big the block size is, there will be applications flood that space with meaningless transactions

And now imagine Satoshi-dice translated in 100 languages and used daily by 100 million people. And now imagine imagine another 100 Satoshi-dice type of service. That's going to be a problem, no matter how big the block size is.

Huh. I think this just nailed it for me.

The blocks will fill up, whatever size they are. This is simply human nature's approach to computing technology. Rewriting the rules to expand the size in the hopes of mitigating that is a task that will never end; there will be new calls for a larger block size every year, and before we know it, we'll be outpacing Moore's Law and slowly turning the blockchain into an unusable artifact.

Fortunately, Bitcoin already has a built-in mechanism to counteract overcrowded blocks: transaction fees.

When the blocks do start becoming consistently full, it'll be the ones with the highest priority (as in, smallest size and largest transaction fee) that get first priority, as it should be. If this means that SatoshiDice can't compete, and has to quit since it can't send tons of transactions for little to no fee any longer, why in the world should I or anyone else (except the owner of SatoshiDice... sorry bud, but you should have seen this coming) shed a tear over it? Pushing for a core rule change to Bitcoin to accommodate one current, and multiple future companies that want to fill blocks with their own transactions and not pay for doing so strikes me as just a bad idea.

So does that mean there should never be a block size increase? I'm going to say yes, there should not be one. Maybe it was originally intended, but I don't think it's necessary, and really, it's a pointless endeavor, if allowing "enough" transactions is the goal. (If instead the goal is a set, ideal number of transactions per second, say to match PayPal, and it's intended to be the only increase ever, I'd be more sympathetic... but I doubt that'll ever be the proposal.)

That's not to say that there shouldn't be some way to allow more, smaller-value transactions to flow. It's been pretty clear for a while now (to me, anyway) that Bitcoin needs a silver to its gold. We need a second cryptocurrency--one with larger blocks; smaller, simpler transactions, possibly doing away with scripting and only allowing simple spends; faster difficulty adjustments; a less-demanding full node; and somewhat faster confirmation times.

There actually already is another chain that could be perfect for running as a parallel chain- Terracoin. It's been around a few months and is pretty much a straight copy of bitcoin, with shorter block times (2 minutes), faster diff retarget, lower reward that halves every 4 years (20TRC per block). It's updated by adding the new patches for bitcoin, so it keeps the same consistency with bitcoin as well in terms of security and reliability. The address format is even the same- so there is no need to migrate addresses, your Private key for bitcoin yields the same private key for Terracoin. Although at first some people thought this would lead to confusion, the result has been the opposite. One Public key works for both TRC and BTC, making it very convenient.

The chain is young enough that an increased block size could probably be implemented without much trouble, or so many users to disturb. There is even talk about having it merge-mined to boost its security.

Currently it's traded at Vircurex, Bitparking and Zaptos exchanges. There are several Satoshi Dice like sites, and even an 'instawallet' that is in beta development. It has it's own forums as well: www.terracointalk.org (because the alt-forum is a bit of a cesspool).

Perhaps terracoin would be fertile soil to test some of these ideas? The user base is very open to testing new ideas.

Hmm. Worth looking into.

[solex]

Fortunately, Bitcoin already has a built-in mechanism to counteract overcrowded blocks: transaction fees.

Which is why this type of proposal is being offered...

https://bitcointalk.org/index.php?topic=145754.msg1551072#msg1551072

[Rogue Star]

Reading over the technical/economic arguments, I think the max block size should be whatever the p2p network (full nodes) can handle while allowing smallish miners to be competitive. The limit should be expanded to meet consumption as long as the p2p network can support it. If the change is done in one or more hard forks, pressure to develop better solutions shall and are emerging to make use of the available space more efficiently. We do not need a kitchen sink solution where everyone and their toaster can run a full node. What we do need is a Bitcoin where everyone that puts in a minimal amount of effort can run a full node, although running an archive node might not always be simple.

If someone wants to keep their sovereignty on a 56K modem, they can convince AOL to send out block chain coasters to everyone (using CDs) every week (tongue in cheek solution). If miners can't afford to include free transactions, they won't. Node relay rules don't determine what will be included in the block, the miners decide. I can't see inclusions for txns being much different than the market for domain registrars, if anything it is already a much freer market. Let the free market do its work in that regard. Bitcoin doesn't promise that anyone can validate the block chain in real time, for free, forever.

Like a few people in the other threads said, the market for including transactions is exactly like electricity prices. If marginal costs are not covered, miner won't do it any longer than they would if electricity prices are too high. People can use overlay networks as much as they want, the block chain should support whatever the network can handle. Arguing about max block size within the bounds of what is pragmatic is like arguing that txn relay fees should never change. I see any non-technical limit on transactions as anti-Bitcoin and not the Bitcoin I signed up for.

[HBBZ]

Well, as a bitcoiner without programming background, I simply smelled something important. That is to run a full node, the Satoshi client and configure it to run at startup. It's something like my computer is a branch office of a bank that can perform full services to the community as much as headquarters do. 

[da2ce7]

I see the block size (and frequency) as much as economic rules as the total number of coin is.  I disagree with Satoshi that it is just something that we can 'change.'

I'm part of that sticky group of people that WILL NOT accept a larger block size.  I doesn't matter what other people say; or the reasons that they give.  Frankly, I would reject (with my full node, and SPV nodes) any non-bitcoin chain.  Any block breaking the 1mb/10min avg limit will not be a bitcoin for me.


However I am not worried.  I know that Bitcoin will be just fine with a 1mb block limit; as MarkM so eloquently put is, we need to get over having only one chain, but rarther think of having a ecosystem of hundreds of merge-mined chains that all fullfill different roles.


One of the reasons, (there are many), that I work on Open Transactions is that I believe that we need secure and cryptographically auditable off-chain transactions.  Making the need for more than 7tx/s on the Bitcoin block chain much (completely?) negated.

The best part is that I am in control of this problem.  Even if every other person moves to larger (or more frequent) blocks, I know my bitcoins are safe, and for-me bitcoin will keep-on working. (well other than the attack of a drastic lower hash-rate maybe)

So yes, I think that there is AT LEAST a key 10% of the Bitcoin user population (including me) that will veto any such changes.

[solex]

So yes, I think that there is AT LEAST a key 10% of the Bitcoin user population (including me) that will veto any such changes.

Phew, for a moment you had me worried, but with 90% following one side of a fork - then worries over.

da2ce7, please consider that nearly everyone on this forum has bitcoin's interests at heart. That it why thousands of posts debate the minutiae of any software change. Then consider how much progress there would be if 100% agreement was needed before any action took place...

[da2ce7]

So yes, I think that there is AT LEAST a key 10% of the Bitcoin user population (including me) that will veto any such changes.

Phew, for a moment you had me worried, but with 90% following one side of a fork - then worries over.

da2ce7, please consider that nearly everyone on this forum has bitcoin's interests at heart. That it why thousands of posts debate the minutiae of any software change. Then consider how much progress there would be if 100% agreement was needed before any action took place...

But, for such a change, even 1% can veto it... That is what makes Bitcoin so great!  The economic problems caused by 1% forking makes it not worth for the 99% to implement the change.

[justusranvier]

I'm part of that sticky group of people that WILL NOT accept a larger block size.  I doesn't matter what other people say; or the reasons that they give.  Frankly, I would reject (with my full node, and SPV nodes) any non-bitcoin chain.  Any block breaking the 1mb/10min avg limit will not be a bitcoin for me.

If they developers commit to scaling the network as far as the demand for transaction requires so that it can be the replacement for PayPal, and for Visa, and maybe even national currencies that we've all been promised, then I'll start up an extra full node to replace yours.

[tvbcof]

...

So yes, I think that there is AT LEAST a key 10% of the Bitcoin user population (including me) that will veto any such changes.

I'll not 'veto' an effort to have Bitcoin evolve to a high footprint system which natively competes head-on with mainstream solutions, but I expect that it will either fail outright or eventually be perverted to the point where it is worse than nothing.  All the same, I'll hope that I am wrong and wish it well.

I expect that there will be enough like-minded stakeholders to operate a light-weight solution which focuses on security.  If so, I'll be one.  As I said on one of my few OP threads, I feel that the resources needed to light-weight chain if security and credibility outweigh a profit motive are well within the reach of what a small number of enthusiasts can muster.  A chain maintained as an insurance policy against subversion of the 'real' Bitcoin does not necessarily need to spend it's gestational phase of life in highly profitable mode and a fairly small group of maintainers is something which has less potential to be subverted.

I've always specifically expected that all of my monetary input which acquired BTC would be a total loss.  A highly speculative adventure.  No matter what happens vis-a-vis the block size, I'll be re-claiming my initial investment one of these days.  Then the remaining BTC can sit for years.  So if the blockchain forks I'll have plenty of time to see how things progress.  I could have my hand forced if one chain forces a legacy spend of course, but I'll cross that bridge when I come to it.

[da2ce7]

I'm part of that sticky group of people that WILL NOT accept a larger block size.  I doesn't matter what other people say; or the reasons that they give.  Frankly, I would reject (with my full node, and SPV nodes) any non-bitcoin chain.  Any block breaking the 1mb/10min avg limit will not be a bitcoin for me.

If they developers commit to scaling the network as far as the demand for transaction requires so that it can be the replacement for PayPal, and for Visa, and maybe even national currencies that we've all been promised, then I'll start up an extra full node to replace yours.

That isn't the point either.  The fact is that my nodes will continue to run quite fine; rejecting your blocks.  I will be using bitcoin, and you will be using some alt-rule-chain that I don't consider to be bitcoin.

My coins will be spendable in both chains.  

The thing is that if 10% (or even 1%) of the Bitcoin USERS do not accept the change, the change will not happen: because the economic consequences will be too-large.


We are not working in a democracy here.  We are working with a "do what the you want, but I'm not going to change my code mkay."
For this change to be successful, the bar isn't 50%, or 90%, or 99%, but rather closer to 99.9%.

[justusranvier]

We are not working in a democracy here.  We are working with a "do what the you want, but I'm not going to change my code mkay."
For this change to be successful, the bar isn't 50%, or 90%, or 99%, but rather closer to 99.9%.

All you're saying is that to be successful the userbase just needs to grow by an order of magnitude, which is exactly what one would predict if the demand for transactions is exceeding what 1 MB blocks can provide.

[Rogue Star]

All you're saying is that to be successful the userbase just needs to grow by an order of magnitude, which is exactly what one would predict if the demand for transactions is exceeding what 1 MB blocks can provide.

He is also saying his fork will continue to exist regardless. That's fine, he can use his inconsequential fork as long as he has another node to connect to (even if it is his own). He can also use Open Transactions he is selling as a solution all he likes too.

I suspect most of us will not support a fork that doesn't scale to its potential. Zealotry is not a good reason to keep a crude network management rule that has served its purpose and was meant to be improved. Much of the protocol was meant to be upgraded and replaced as needed. The 1MB limit increase will likely be the second improvement of many. The first as I understand it is block V2, which is still phasing in. Others like the hashing and key algorithms appear to be fine for the foreseeable future and probably all our lifetimes, but they will change eventually if bitcoin doesn't die. The txn limit is actually much lower than 7 tps if bitcoin ever sees significant usage of any of it's coolest features, like smart contracts.

[da2ce7]

He is also saying his fork will continue to exist regardless. That's fine, he can use his inconsequential fork as long as he has another node to connect to (even if it is his own). He can also use Open Transactions he is selling as a solution all he likes too.

Except it isn't just me, there are hundreds of users (and owners of bitcoin) like me, who signed on to a system that we thought would have LOWER relative costs to verify in the future.  That one day every, even the cheapest computer, could become a full-node.

The thing is that these people don't post on the forums about this change.  Because they are not arguing FOR a change. They are quite happy with bitcoin is atm.

They don't even need to be convinced about the change.  Their Bitcoin; the current Bitcoin; the real Bitcoin.  Will continue to work just fine.

[justusranvier]

Except it isn't just me, there are hundreds of users (and owners of bitcoin) like me, who signed on to a system that we thought would have LOWER relative costs to verify in the future.  That one day every, even the cheapest computer, could become a full-node.

That expectation is insane, unless you expect Bitcoin to remain a tiny niche currency that a statistically insignificant fraction of a tiny minority of the population use, or you just want it to die on the vine.

There are a lot of optimizations which could be put in place that would allow regular users to process much higher transaction rates. Maybe we'd need to require users to have a computer purchased within the last 2 or 3 years with a broadband connection instead of a dial up modem, is that really too much to ask?

Instead of complaining about how higher transaction rates will make it harder to run a full node, why not support some of the people trying to implement optimizations to make that less of a problem?