For those who want to increase the block size limit.

[Piper67]

Remember: Bitcoin is a proof of concept and an experiment not a plan to take over the world.

Wrong. 

I guess we agree to disagree then

I can hear the same conversation in my head from the mid-80's "this Internet shit is just a proof of concept experiment, it's not meant to change the world"! 

[ElectricMucus]

to you as well:

Wouldn't it be better to wait with our plans till the original objective is successful? I gladly discuss world domination plans with you once we have hit the hard limit consistently and have a balance between initial block reward and fees in terms of mining income.

[ArticMine]

I am still waiting for the calculation.

Currently there's a maximum of ~7 transactions per second.

If 0.0001% of the total population (http://en.wikipedia.org/wiki/World_population --> ~7 billion, 0.0001% of that are 7000 persons) use Bitcoin, they can do max. 604800 transactions alltogether --> 86.4 transactions per person per day.

ArcticMine might be off by ~2 orders of magnitude (0.01% --> 700k users --> 0.864 transactions per day max.) but still even with "only" less than a million active users, they have to wait more than a day to be able to do 2(!) transactions!


About transaction markets:
I gave the example elsewhere as well, TOTAL transaction fees will not be different with small and unlimited block sizes (miners need to pay for their electricity etc.).
Let's say 2 BTC/block are required in fees for a miner.
There's a cost (blocks take time to transmit to other miners and the longer it takes the more you risk an orphan block) associated with block size, so if a miner can choose between 200000000 transactions with 1 satoshi fee and 2 transactions with 1 BTC fee, he's better off with the 2 transactions than the 200000000. The equlibrium might be higher with unlimited block sizes than with some limits, but rest assured that miners will find ways to determine how to force fees and make sure they are paid.

My estimate was conservative but realistic. Visa for example has a burst capability equal to 5x its average transaction rate. https://en.bitcoin.it/wiki/Scalability. Furthermore many people perform more than one financial transaction a day. Ten for example is a realistic number in many cases especially during busy times of the year. Applying a factor of of 50 to compensate for the burst requirement and number of transactions puts my estimate in the right order of magnitude. Even if one can argue that 70000 users rather than 7000 users can be accommodated it is very hard to not argue that this change is desperately needed. Also such a hard fork needs considerable time in order the needed consensus to be obtained in the Bitcoin community so this cannot be left to the last minute.

[ArticMine]

to you as well:

Wouldn't it be better to wait with our plans till the original objective is successful? I gladly discuss world domination plans with you once we have hit the hard limit consistently and have a balance between initial block reward and fees in terms of mining income.

Waiting until the last minute is not a good idea since it would make a sane implementation of this next to impossible without seriously disrupting the Bitcoin network. Gavin's proposed approach needs close to two years.

A hard fork won't happen unless the vast super-majority of miners support it.

E.g. from my "how to handle upgrades" gist https://gist.github.com/gavinandresen/2355445

Example: increasing MAX_BLOCK_SIZE (a 'hard' blockchain split change)

Increasing the maximum block size beyond the current 1MB per block (perhaps changing it to a floating limit based on a multiple of the median size of the last few hundred blocks) is a likely future change to accomodate more transactions per block. A new maximum block size rule might be rolled out by:

New software creates blocks with a new block.version
Allow greater-than-MAX_BLOCK_SIZE blocks if their version is the new block.version or greater and 100% of the last 1000 blocks are new blocks. (51% of the last 100 blocks if on testnet)
100% of the last 1000 blocks is a straw-man; the actual criteria would probably be different (maybe something like block.timestamp is after 1-Jan-2015 and 99% of the last 2000 blocks are new-version), since this change means the first valid greater-than-MAX_BLOCK_SIZE-block immediately kicks anybody running old software off the main block chain.

[ElectricMucus]

I would at least wait till free transactions are sweated out. Then we can see what solution we actually need and not before.

[Littleshop]

Remember: Bitcoin is a proof of concept and an experiment not a plan to take over the world.

Wrong.  

I guess we agree to disagree then

Wouldn't it be better to wait with our plans till the original objective is successful? I gladly discuss world domination plans with you once we have hit the hard limit consistently and have a balance between initial block reward and fees in terms of mining income.

Well maybe, maybe not.  The original plan seems to be successful to me.  The world domination was a joke, but the world changing part is not.  

Transaction volumes are growing as we speak.  Soon there could be regular backlogs of NON FREE transactions.  Increasing the allowable volume is not really dangerous to mining income.  If 1/4 of the miners allow unlimited free transactions then free transactions could wait 40 minutes or more (and have uncertainty) while paid transactions move to the front of the line.  Right now with a block reward of 25 BTC, the miners are being paid.  When the block reward drops to 12.5 then I am sure even more miners will be unwilling to do free transactions.  

At some point miners will need to voluntarily decide to move to pools that offer less or no free transactions.  This is much more important to mining income then the maximum block size.  This, along with a larger blocksize then current, both seem to be a part of Satoshi's original plan.  

[Piper67]

One of the beauties of Bitcoin is the self-regulatory nature of the protocol. As BTC become more valuable, more people want to mine, and in doing so they make the process more difficult... It has a certain symmetry, elegance, which I find very appealing.

Is there a way to apply the same notion to the block size limit? In other words, is it possible to make it so that those who benefit from the increased block size MUST pay a price by the block size increasing?

Could the block size not be made self-regulating in a similar way to the difficulty vs hash power vs price/adoption concept?

[ElectricMucus]

The problem is that there is no real market as long as there are free transactions. And you are crashing an existing transaction market every time you make them reliable again.

One of the beauties of Bitcoin is the self-regulatory nature of the protocol. As BTC become more valuable, more people want to mine, and in doing so they make the process more difficult... It has a certain symmetry, elegance, which I find very appealing.

Is there a way to apply the same notion to the block size limit? In other words, is it possible to make it so that those who benefit from the increased block size MUST pay a price by the block size increasing?

Could the block size not be made self-regulating in a similar way to the difficulty vs hash power vs price/adoption concept?

Yes
I would consider it an acceptable compromise. But that's the reason we have to wait till we have an actual transaction market to know how to do it so it doesn't do more harm than good.
Wanna have it before that? Well that's what alt chains are for.

[bg002h]

Miners will want to increase the block size eventually...false short supply of transaction space in a block might temporarily increase the fee paid per transaction...but a block chain that can't support many transactions won't be very useful...so people won't use it...whereas a high block limit would increase the total number of fee payers (and probably decrease the mega fees paid by a very small number of people).

[ArticMine]

Miners will want to increase the block size eventually...false short supply of transaction space in a block might temporarily increase the fee paid per transaction...but a block chain that can't support many transactions won't be very useful...so people won't use it...whereas a high block limit would increase the total number of fee payers (and probably decrease the mega fees paid by a very small number of people).

Exactly. The problem with the current block limit is that increasing the price does not increase the supply once the limit is reached. So a proper market cannot develop. Increasing the block size increases the miners effective cost so the market can properly set a price in terms of transaction fees; however if the block size cannot be increased no matter how much in fees is offered then this market and consequently Bitcoin will fail. One only needs to take a look at the current rate of Bitcoin adoption to recognize than when this hits it will be quite sudden and there will not be the time to implement the change over an extended period of time as Gavin has suggested.

[notig]

Miners will want to increase the block size eventually...false short supply of transaction space in a block might temporarily increase the fee paid per transaction...but a block chain that can't support many transactions won't be very useful...so people won't use it...whereas a high block limit would increase the total number of fee payers (and probably decrease the mega fees paid by a very small number of people).

Exactly. The problem with the current block limit is that increasing the price does not increase the supply once the limit is reached. So a proper market cannot develop. Increasing the block size increases the miners effective cost so the market can properly set a price in terms of transaction fees; however if the block size cannot be increased no matter how much in fees is offered then this market and consequently Bitcoin will fail. One only needs to take a look at the current rate of Bitcoin adoption to recognize than when this hits it will be quite sudden and there will not be the time to implement the change over an extended period of time as Gavin has suggested.

+1

[Zangelbert Bingledack]

^ +1

[ElectricMucus]

You just took my argument and reversed it. This has absolutely no basis in reality.

As long as transaction space is abundant there is no transaction market period. Markets spawn from scarcity not abundance. Scarcity does not prevent the formation of a transaction market since you cannot hoard transactions, they are there to be used, but abundance does.

The worst that can happen if the hard limit is reached is that transactions become much more expensive, but it won't reach astronomical levels since bitcoin transactions stand in natural competition with other systems.
This is a good thing since it would make bitcoin transactions matter and then only then we can collect the necessary data to come up with the right solution.

[notig]

You just took my argument and reversed it. This has absolutely no basis in reality.

As long as transaction space is abundant there is no transaction market period. Markets spawn from scarcity not abundance. Scarcity does not prevent the formation of a transaction market since you cannot hoard transactions, they are there to be used, but abundance does.

The worst that can happen if the hard limit is reached is that transactions become much more expensive, but it won't reach astronomical levels since bitcoin transactions stand in natural competition with other systems.
This is a good thing since it would make bitcoin transactions matter and then only then we can collect the necessary data to come up with the right solution.

Ah exactly as I thought. Someone who basically wants bitcoin to fail at normal day to day activities so that "natural competition" like litecoin... can succeed. The worst that can happen is not that bitcoin transactions become more expensive (that's actually guaranteed if you assume that bitcoin grows in popularity) . But of course what you're saying is that transactions might not get more expensive because the "natural competition" will start being cheaper to transact with than bitcoin so people will use that instead.

If you do not increase the hard limit of 1MB you do more than just produce scarcity. You create a bottleneck and you force a large portion of people off the chain entirely to use "natural competition" . Like lite coin. Or Clearing houses . Which are centralized, unaccountable, and weak links.

[markm]

Exactly. The problem with the current block limit is that increasing the price does not increase the supply once the limit is reached.

We don't know that. I actually suspect that as long as both the hard coded transaction fees and the hardcoded block size limit are not changed the price of bitcoins themselves will in fact increase the block size at some exchange rate.

A reducio ad absurdum of that is, suppose tonight bitcoins skyrocketed to $10,000 per coin. For one coin one could not only budget a year of home fibre but also pick up a nice 8-core, 32-MB-RAM, 2TB-disk workstation/server machine and, at least if one were of the or a "me generation", then happily join the "moar size, moar size, moar size" brigade with an undertone of "I'm okay Jack, sucks to be you" directed at no third world resident in particular.

Sucks to be the guy who bought in at $9999, of course; had he bought in back when it was only two digit values of dollars though maybe he'd see it the moar moar moar brigade's way too.

Sucks to be the miner who sold their coins instead of holding them that magic night too, I guess.

But it is a market. Offer us enough dollars per bitcoin on the bitcoin exchanges and sure, we'll double the max block size. No telling at what exchange rate though of course, that's for the moar moar moar brigade to discover by making offers for coins on the exchanges...

-MarkM-