Soft block size limit reached, action required by YOU

[conv3rsion]

You may have heard me say "Bitcoin is an experiment" before...  well, we're finding out right now what happens as the experiment scales up.

First:  I sent a message to the big mining pools, reminding them of the run-time options they can set to control the size of the blocks they create. I did not tell them what they should or shouldn't do, I think we need to move beyond centralized decision-making.

I did send them a pointer to this very rough back-of-the-envelope estimate on the current marginal cost of transactions:
  https://gist.github.com/gavinandresen/5044482

(if anybody wants to do a better analysis, I'd love to read it).

Second: block size is half of the equation. The other half is transaction fees and competition for getting included into blocks. All of the bitcoin clients need to do a better job of figuring out the 'right' transaction fee, and services that generate transactions will have to adjust the fees they pay (and are, already).

Finally: in my opinion, there is rough consensus that the 1MB block size limit WILL be raised. It is just a question of when and how much / how quickly.

Thank you Gavin, this is exactly what people need to hear.

[1714782811]

1714782811

[1714782811]

1714782811

[1714782811]

1714782811

[Blowfeld]

Where to begin???  Yes, I'm new to this forum.  But, I have been observing Bitcoin as an outsider for quite a while.

By and large, I agree with LukeJr.  SD is spam.  Imagine a real-world Las Vegas style casino sending a transaction to every other casino around the world (and perhaps to every gambler (and perhaps to every person with money in their pocket)) every time a customer pulls the handle on a slot machine.  The SD model is insane and completely unsustainable.

Those who equate Bitcoin with Gold are also rather extremists.  Perhaps in the old west you could buy a whiskey for some gold dust.  But I've never bought a latte with a shaving off a gold coin.

Those who demand higher fees are ignoring the reality.  As it stands now, those who operate full nodes bear the burden of the SD-like spam.  Yet, there is no fee or reward paid to run such a node.  [Back when bitcoin started, any full node who wanted compensation was occasionally rewarded by solving a block.  Now that is infeasible.]  If things aren't changed, the only people who will run full nodes are the mega-miners.  And consolidated centralized control is only a few mergers away.

As has been discussed elsewhere, the receiver of funds (not the sender) should specify the level of certainty they want in assuring their received funds are valid.

Clearly a 1 Satoshi speck of dust doesn't require the level of confirmation as a 10,000 BTC transaction.  Without a change to the protocol, there is no distinction.  But all such protocol changes I've seen discussed have flaws.  Some research effort should be extended to figure this out.  Meanwhile...

In block 224588, someone paid a 94 BTC fee.  I presume this was a mistake.  On the other hand, as I suggested in another thread, perhaps this is intended as a means of laundering tainted money into clean money.  If I had the resources to operate a mining pool, I think I would *only* include transactions with *no fee*.  [How else do you guarantee you are receiving non-stolen money in your block reward?]  [How else do you avoid the moral conundrum and substantial effort of "returning" the (obviously erroneous) transaction fee of US$4000 to the rightful party?]

In my view, the only realistic long-term solution is one that eliminates centralized control of mining pools. 

Changing the hash algorithm every 6 months would tend to thwart those (governments, large corporations, or large syndicates) who can afford to invest in ASIC technology.  Software that can adapt (almost) instantly (to a change of the constant used in Round 37 of the SHA1 algorithm, for example), puts the mining back in the hands of the regular people who operate the nodes and suffer the burden of ever-increasing block-chain bloat.  With 10,000 people deciding what they will include or exclude from their mined blocks, Bitcoin might again become a decentralized currency.

Furthermore, if it is in the *system's* best interest to pass transactions, even as their number and size increases, then the inherent reward created by generating a block should be *increasing*, not decreasing by half every four years.

[skibum]

Sounds like the beginning of the end.

[Peter Todd]

Finally: in my opinion, there is rough consensus that the 1MB block size limit WILL be raised. It is just a question of when and how much / how quickly.

To say there is a rough consensus that the 1MB block limit will be raised at some unspecified time in the future is missing the point. The real issue is, is there a consensus that a large fraction of the transaction volume will in the future happen off-chain? Given the range of opinions between you and Mike, who expect transaction fees to stay low enough for all but microtransactions, Pieter Wuille, who if I am correct is unsure, and Jeff Garzik, and Gregory Maxwell, who are both working on designs for off-chain transaction systems, I just don't see a consensus.

After all, for the user, the limit itself isn't the issue, the issue is how they will be expected to store and spend their Bitcoins, and what kind of security Bitcoin will provide them. I haven't seen anything from you or Mike actually talking concretely about what kind of protection from authorities and others trying to control Bitcoin that you expect Bitcoin to be able to provide. For instance, in the future, do you expect that by paying a sufficiently high fee can I expect get my transaction confirmed eventually, regardless of who I am or what the transaction is for? Do you expect me to be able to vote on what transactions do get confirmed and included in blocks, in proportion to the hashing power I possess? Do you expect me to be above to participate in Bitcoin by creating transactions and mining as a full validating node anonymously?

Satoshi said back in 2010 that he intended larger block sizes to be phased in with some simple if (height > flag_day) type logic, theymos has linked to the thread before.

I think he would be really amazed at how much debate this thing has become. He never attributed much weight to it, it just didn't seem important to him. And yes, obviously, given the massive forum dramas that have resulted it'd have been nice if he had made the size limit floating from the start like he did with difficulty. However, he didn't and now we have to manage the transition.

You know, I didn't think anything at first of the Wiki page on scaling advocating simply increasing the block size as required, even to the point of requiring full nodes to have tens of thousands of dollars worth of high speed internet connections and UTXO storage hardware. But eventually I started thinking about the implications - the turning point for me was really when I realized how large miners could use large blocks as a way to force smaller miners out of business.

Satoshi didn't foresee pool mining, the dangers of bloating the UTXO set, or even that multiple clients would be used. Bitcoin was a brilliant idea, don't get me wrong, but Satoshi was a person, not some all knowing all seeing deity. It doesn't surprise me the slightest that he might not have thought through all the implications of allowing the block size to increase without bound, just like he left Bitcoin with a scripting system that turns out to not have that many actual applications.

The massive forum dramas ultimately come down to the fact that regardless of whether the block size is left as it is, or increased, Bitcoin as we know it will change; there just isn't consensus on how it will change. You come from Google, a place of massive centralized server farms controlled by one company. Google's services work pretty well - centralization can have benefits - but many of us feel that goes down a very dangerous path. It's easy to see how a world where blocks are sufficiently large that only well funded pools with highly visible high-speed internet connections can lead to government and large businesses controlling Bitcoin. Fundamentally, distributing large amounts of data in a censorship resistant way is far harder than distributing small amounts of data.

[commonancestor]

We are already bumping up against a limit. No simulation needed. We have reality instead ...

OK, wrong word. Still I hope that most pools will not immediately lift / increase their limit. So we know what will happen when we reach the hard limit, which can't be changed so easily.

So let us bump against this for a few month. And lift it only if it really damages Bitcoin.

Yes, this is the right approach. Solve the problem now soft way, rather than later hard way.

I think the solution should be increasing both tx fees and max block size. The network should not discriminate against users based on who they are (e.g. SatoshiDice). If it's possible to flood the network with cheap transactions then the fees are simply too low and should be increased. Also miners should be less dependent on generating coins and more on receiving fees. When the fee is "about right" and blocks are getting full then increase max block size. It should be such so that average Joe can still run his full node or a miner.

[lucif]

Where is the foundation during the only moment we need it?

They jerk on mtgoxlive



I don't know what sort of organization is it. Just a bunch of nerds collecting fees to join other for jerking on price.

[VinceSamios]

This thread is very interesting in that it is a place where some of the smartest people on the plant, discuss things with some of the stupidest.

I love bitcoin.

[jade087]

Increasing the block size is a temporary fix. Everyone involved should be fully aware that if you want your transaction to be processed quickly, you should include a transaction fee. Bitcoin shouldn't be the network for unlimited free transactions that millions of clients verify for you.

I don't consider SD a DDoS, but I do consider the people that use it as spammers. AFAIK SD returns winnings with an appropriate tx fee; you should be including the same in your bets.

That said I wouldn't be surprised if the block size will be increased eventually. It shouldn't be taken lightly. It should be conservative. It should not be algorithmic. But increasing the block size isn't going to solve the problem of a busy network which people are trying to use for free (c.f. building more roads invariably increases traffic).

Edit: Remember that everybody needs to have the ability to run their own complete node with the complete blockchain. This is crucial and fundamental to the security of the network.

[SgtSpike]

This thread is very interesting in that it is a place where some of the smartest people on the plant, discuss things with some of the stupidest.

I love bitcoin.

Indeed... fascinating to watch the interaction.

[Anon136]

it only means that we will inevitably need to embrace multiple blockchains. They could be clones of bitcoin or they could be competing cryptocurrencies like litecoin. With a little luck cryptocurrency echanges will be decentralized censorship resistant and ubiquitous in the future. If this is the case than this will make the whole problem a non problem. I dont think this means is that bitcoin will be worth nothing some day, what i do think this means is that bitcoin as we know it has a soft cap on its value. Probably this cap is much higher than 43 dollars or w/e bitcoin is valued at right now.

Multiple blockchains sounds like a mess, considering people are putting their faith into these currencies and a lot of their value comes from scarcity, i'm not sure there is going to be a future with a lot of alt-chains with similar prices. I imagine 1 chain/system will emerge and it will be the one that really solves the block size issue.

Also the only reason Bitcoins are worth $1 - $100 or even more is because many people think that Bitcoin has no transaction cap at all. People think this is the future and the future does not cap out at 7 transactions a second.

also this totally possible. supposing it can really be solved than that cryptocurrency will probably surpass bitcoin and be the final one to rule them all. Still when ever that currency arrives if it does arrive bitcoin will be the way to get into and out of it early on, so ill be ready with my bitcoins. And if it doesnt arrive and this problem really cant be "solved", than my proposal will be good enough to get the job done.

[mintymark]

Increasing the block size is a temporary fix. Everyone involved should be fully aware that if you want your transaction to be processed quickly, you should include a transaction fee. Bitcoin shouldn't be the network for unlimited free transactions that millions of clients verify for you.

The probalem is not about speed, as I understand it. Time was, back in the day when no fee meant that transactions could take a while to confirm, - I once had one take 24 hours, but confirm it did.

The problem is this: transactions with insufficient fee can easily end up effectively lost to sender and receiver, gone, poof.  A skilled operative may be able to retrieve them in some situations, but most people no. People have already lost money in this way.

This is not a characteristic anyone wants Bitcoin to have. Whatever change we make, it should result in a range of transactions speeds I think, so that no transaction fee or small fee can result in very slow transactions, but never loss, and a high fee guarantees inclusion in the next block, as it pretty much does now.

[chmod755]

Note this isn't really a problem if miners are responsible and filter out the SatoshiDice flooding.
My git repository contains a "block_dice" branch to do just that.

I didn't read the code, but I hope it prevents blockchain spam in general and not just satoshiDice (1dice... addresses). Does it?

[K1773R]

Note this isn't really a problem if miners are responsible and filter out the SatoshiDice flooding.
My git repository contains a "block_dice" branch to do just that.

I didn't read the code, but I hope it prevents blockchain spam in general and not just satoshiDice (1dice... addresses). Does it?

only SD, or fake SD adresses

[Blowfeld]

Changing the hash algorithm every 6 months would tend to thwart those (governments, large corporations, or large syndicates) who can afford to invest in ASIC technology.  Software that can adapt (almost) instantly (to a change of the constant used in Round 37 of the SHA1 algorithm, for example), puts the mining back in the hands of the regular people who operate the nodes...

You would have been a load of fun after the invention of the Gutenberg printing press:

Changing the fontsize every few days would tend to thwart those who can afford to invest in Gutenberg printing technology.  Scribes who can adapt (almost) instantly (to a change of the lettering size, for example), puts publishing back in the hands of the regular people who operate the quills...

Do you disagree with the goal of reducing the ability of a large entity from controlling Bitcoin.  Or do you disagree with my (admittedly novel) approach?

[eleuthria]

Pushed an update to all of BTC Guild's stratum nodes this morning.  The nodes are now running essentially double default values:

500kb max size.  50kb set aside for low-fee/high-priority.  50kb minimum size so other transactions will get in if there aren't that many waiting for confirmations.  Already seeing a reasonable reduction in unconfirmed transactions as more blocks get pushed out with 2x the defaults.

[Daily Anarchist]

I can't wrap my head around why so many people are hating on Satoshi Dice.

What if there were some African business helping to educate or feed starving kids in Africa and it had a business model that produced a huge quantity of transactions, would you still be against it?

I really think these anti-Satoshi Dice people are elitists who think that because gambling is stupid in their minds that nobody should be able to.

As if the Bitcoin Protocol gives a shit about what the transactions are used for.

And Luke-Jr says SD is breaking the rules. Really? I was under the impression that any transactions that broke the rules were immediately rejected by the bitcoin protocol. I mean, it looks like they're playing by the rules to me.

Seriously, if Bitcoin can't handle Satoshi Dice just throw in the towel right now, because I have some news for you. In the future, when Bitcoin is the GLOBAL MONETARY STANDARD, 95% of all transactions are likely to be gambling, sex, and drug related.

Those seem to be things that people value highly in life. Just because your puritan asses can't handle it doesn't mean we can't.

I, for one, LOVE that Satoshi Dice is flooding the blockchain.

[SgtSpike]

I can't wrap my head around why so many people are hating on Satoshi Dice.

What if there were some African business helping to educate or feed starving kids in Africa and it had a business model that produced a huge quantity of transactions, would you still be against it?

I really think these anti-Satoshi Dice people are elitists who think that because gambling is stupid in their minds that nobody should be able to.

As if the Bitcoin Protocol gives a shit about what the transactions are used for.

And Luke-Jr says SD is breaking the rules. Really? I was under the impression that any transactions that broke the rules were immediately rejected by the bitcoin protocol. I mean, it looks like they're playing by the rules to me.

Seriously, if Bitcoin can't handle Satoshi Dice just throw in the towel right now, because I have some news for you. In the future, when Bitcoin is the GLOBAL MONETARY STANDARD, 95% of all transactions are likely to be gambling, sex, and drug related.

Those seem to be things that people value highly in life. Just because your puritan asses can't handle it doesn't mean we can't.

I, for one, LOVE that Satoshi Dice is flooding the blockchain.

A big thumbs-up to this post!

[eleuthria]

And don't forget: Satoshi Dice has probably paid more in transaction fees than the entire network combined throughout history (not counting mistakes and/or testing transactions).

Dice payouts include fees, as do most of the bets that go there in the first place.

[Gabit]

I'm sorry to say this, but this debate is moronic. Mining needs to be profitable to groups of people. It is a natural evolution. Everything, and I mean EVERYTHING has gone so, let it be bumping oil, or mining gold. When competition gets hard, it is natural that people group and act together for mutual goal. This means that everyone cuts of fat and aims for the best result available TOGETHER (read: company). Let it be in this case, network speed or processing power, or in gold mining business, the best machinery and the chemicals. The competition it self brings stabilization. And the real value of BTC, is not that some one holds it, It is that it cannot be generated from nothing. One can only spend it once. Centralization of Hashpower is not issue at all, it only drives more competition = security for the network.

If we cannot handle SD, we can't handle shit. 68 trillion is the world GDP, there will be SD^SD of "dust", when BTC makes it. If it cannot handle this, this wasn't suppose to handle anything.

The most efficient wins.

[johnyj]

This topic always as hot as fire  

The scalability strategy should be distributed, not centralized, to centralize all the transaction in a single network is the sure step towards centralization

Bitcoin network is not designed to be a trading platform, high-frequency trading should be handled by brokers with dedicated data center, outside of block chain. If VISA network were used to handle stock trading, they will crash immediately (Nasdaq can do 1 million tx/s while VISA barely can 10k tx/s)

And bitcoin should not become a duplication of existing services like paypal or VISA , let them handle the daily transctions, and let bitcoin handle the more important low frequency transaction, like salary/saving and pension