My thoughts about the blocksize thing

[markm]

We are giving several fucks about actual adoption.

Actual adoption has been such that even $32 per bitcoin worth of adoption never came near to saturating the blocks, and a couple of years later as we approach a $32 per bitcoin level of adoption again now, we are still nowhere near saturating the blocks.

Thus, adoption is still nowhere near enough that we need larger block sizes yet.

When we reach a $64 per bitcoin level of adoption lets see how full our blocks are; maybe we can easily handle even a $100 per bitcoin level of adoption and still have oodles of space left in our blocks.

-MarkM-

[Akka]

Sorry, but I can't sai this often enough. Long term, Mining wont be providable.

I don't necessarily agree. It's obvious if the state of bitcoin remains like it is now, when the reward drops to zero security by mining will drop ridiculously low. There are only two ways to make it work, either increase the fees or increase the number of transactions (with fees) per block. There are currently about 0.5 btc of transaction fees per block, so if the transaction rate would increase 20-fold that would take care of things.

You probably didn't understand what I mean. Even if the reward on one Block would be 100 BTC (extreme example) thanks to fees.

As soon as you can buy plug and play mining equipment for low cost (jalapeño) is available to direct buy, people will buy this things until mining will barely be providable again.

[prezbo]

Sorry, but I can't sai this often enough. Long term, Mining wont be providable.

I don't necessarily agree. It's obvious if the state of bitcoin remains like it is now, when the reward drops to zero security by mining will drop ridiculously low. There are only two ways to make it work, either increase the fees or increase the number of transactions (with fees) per block. There are currently about 0.5 btc of transaction fees per block, so if the transaction rate would increase 20-fold that would take care of things.

You probably didn't understand what I mean. Even if the reward on one Block would be 100 BTC (extreme example) thanks to fees.

As soon as you can buy plug and play mining equipment for low cost (jalapeño) is available to direct buy, people will buy this things until mining will barely be providable again.

Ah I see, you were referencing the "if it won't be profitable for miners to include free transactions" part. The problem is, having larger blocks means validating more transactions, and that still costs cpu power/money, so in the end if it is a zero sum game, including free transactions definitely won't be worth it.

[markm]

s/providable/profitable/

It is providable just by plugging in those Jalapinos.

It won't be profitable for an individual if too many do it, but [blah blah etc etc, lots of buts].

-MarkM-

[Akka]

Ah I see, you were referencing the "if it won't be profitable for miners to include free transactions" part. The problem is, having larger blocks means validating more transactions, and that still costs cpu power/money, so in the end if it is a zero sum game, including free transactions definitely won't be worth it.

No, I only see miner provability not as an argument. For both sites.

More transaction fees mean a securer network, though.

[conv3rsion]

We are giving several fucks about actual adoption.

Actual adoption has been such that even $32 per bitcoin worth of adoption never came near to saturating the blocks, and a couple of years later as we approach a $32 per bitcoin level of adoption again now, we are still nowhere near saturating the blocks.

Thus, adoption is still nowhere near enough that we need larger block sizes yet.

When we reach a $64 per bitcoin level of adoption lets see how full our blocks are; maybe we can easily handle even a $100 per bitcoin level of adoption and still have oodles of space left in our blocks.

-MarkM-

Respectfully, I think you are conflating BTC price with adoption. Last time the price was $32 per bitcoin, the network was doing 10-12k transactions per day. It is now doing nearly 6x that number and at still not yet at that price level. We are approaching 25% of max block size. The price could drop and we could still end up saturating the blocks.

[markm]

Respectfully, I think you are conflating BTC price with adoption. Last time the price was $32 per bitcoin, the network was doing 10-12k transactions per day. It is now doing nearly 6x that number and at still not yet at that price level. We are approaching 25% of max block size. The price could drop and we could still end up saturating the blocks.

If it is not adopted by money, sucking in more penniless peasants will just invite them to lose what little money they have by putting into a system they do not have enough money to pay the costs of securing.

We need the people paying $15+ per wire to wire half their paycheque home to their family abroad much more than we need people who might as a novelty buy a coffee with it some day but don't even think about the fee they pay with their current electronic payment method (credit cards and such) when buying coffee currently with fiat because heck not only is bending over to pick up a penny a waste of calories but a handful of pennies in fees to buy a coffee is also not worth one's time to even think or worry about.

Our big fee-savings advantage is on the big-ticket transfers, not on puny sums.

We want the guy paying gosh knows what to pay for a shipload of automobiles to be delivered to his autopark to inspect and repair to send out to dealers across the province, heck we don't even want the guy selling individual grams of weed as much as the guy ordering another container of it to supply to the guys out on the street selling grams. Etc.

-MarkM-

[Draino]

Why is this even a thing?
So we're going to solve a deficit issue by jumping right to inflation?

Great way to show off your principle, Bitcoin.

Full retard mode engage

[conv3rsion]

Respectfully, I think you are conflating BTC price with adoption. Last time the price was $32 per bitcoin, the network was doing 10-12k transactions per day. It is now doing nearly 6x that number and at still not yet at that price level. We are approaching 25% of max block size. The price could drop and we could still end up saturating the blocks.

If it is not adopted by money, sucking in more penniless peasants will just invite them to lose what little money they have by putting into a system they do not have enough money to pay the costs of securing.

We need the people paying $15+ per wire to wire half their paycheque home to their family abroad much more than we need people who might as a novelty buy a coffee with it some day but don't even think about the fee they pay with their current electronic payment method (credit cards and such) when buying coffee currently with fiat because heck not only is bending over to pick up a penny a waste of calories but a handful of pennies in fees to buy a coffee is also not worth one's time to even think or worry about.

Our big fee-savings advantage is on the big-ticket transfers, not on puny sums.

-MarkM-

I disagree. More usage is more trust. More trust is more value in a crypto-currency.

But to your point about big fee-savings advantage, fewer transactions = higher cost per transaction. Higher cost per transaction = less advantage over existing methods, which equals less value at all. It also means, fewer use cases. Why, on earth, do we want fewer use cases?

[conv3rsion]

Why is this even a thing?
So we're going to solve a deficit issue by jumping right to inflation?

Great way to show off your principle, Bitcoin.

Full retard mode engage

In no way will raising the number of transactions per block  (not the number of blocks, the number of coins, or the reward per block) increase inflation.

[twolifeinexile]

Why is this even a thing?
So we're going to solve a deficit issue by jumping right to inflation?

Great way to show off your principle, Bitcoin.

Full retard mode engage

Nothing here involves any inflation of deficit of sorts...where did your idea come from?

[Draino]

isn't this about facing a deficit of transaction numbers vs. block size?  and hacking a solution that involves making the rules arbitrarily larger?

i must be confused.

[conv3rsion]

isn't this about facing a deficit of transaction numbers vs. block size?  and hacking a solution that involves making the rules arbitrarily larger?

i must be confused.

It has nothing to do with the money supply (inflation). Its about not being able to perform enough transactions to keep up with usage. The current hard limit of 1MB block size coupled with the fixed number of blocks means about 7 transactions per second. To put that into perspective about 220 million people could do 1 transaction per year. That's less than 2/3 of the United States  (ignoring the rest of the world). ONE PER YEAR!

There are people here who seriously think this is acceptable. 220 million transactions per year for the whole world, total. Those people have a very very different vision of Bitcoin then the one that I do.

[Draino]

It has nothing to do with the money supply (inflation).

that's not the type of inflation i was talking about..

parallels, metaphor, subtlety

that sort of thing.

[markm]

isn't this about facing a deficit of transaction numbers vs. block size?  and hacking a solution that involves making the rules arbitrarily larger?

i must be confused.

The opposite. Really.

We face a deficit of paying transactions as compared to our ever-since-we-started huge oversupply of block size.

There have never been anywhere near enough paying transactions to fill our blocks, thus some of us are maybe doubting we can afford to increase the size of our blocks, because the capital cost of equipment to accommodate larger blocks is not being paid for because blocks have always been far far from full.

Our blocks are too large compared to the number of paying transactions in them, and people are claiming that no one is going to provide us enough paying transactions per block to come anywhere close to filling our blocks unless we increase the size of our blocks, that is, start leaving them far more empty, far less well supplied with paying transactions, than they have always so far been.

It seems to amount to "no one is going to pay you for a cup of water while you only have a lake full available, you need to have ten lakes full first then maybe people might consider buying a cup".

-MarkM-

[conv3rsion]

isn't this about facing a deficit of transaction numbers vs. block size?  and hacking a solution that involves making the rules arbitrarily larger?

i must be confused.

The opposite. Really.

We face a deficit of paying transactions as compared to our ever-since-we-started huge oversupply of block size.

There have never been anywhere near enough paying transactions to fill our blocks, thus some of us are maybe doubting we can afford to increase the size of our blocks, because the capital cost of equipment to accommodate larger blocks is not being paid for because blocks have always been far far from full.

Our blocks are too large compared to the number of paying transactions in them, and people are claiming that no one is going to provide us enough paying transactions per block to come anywhere close to filling our blocks unless we increase the size of our blocks, that is, start leaving them far more empty, far less well supplied with paying transactions, than they have always so far been.

It seems to amount to "no one is going to pay you for a cup of water while you only have a lake full available, you need to have ten lakes full first then maybe people might consider buying a cup".

-MarkM-

Ahh we get to the crux of it. You want to be paid more to mine blocks, period. You are hoping that with a hard block size that each block you mine will be worth more and require less work to mine and store. You don't care if fewer people use Bitcoin, or if Bitcoin is less useful to the world. And you ignore that as a niche with higher transaction fees more closely resembling those of its competitors, Bitcoin will remain less valuable as either a technology or a payment system.

I'm sorry if mining isn't currently as profitable as you want it to be. The network is already the worlds largest distributed computing platform and very secure. I care more about the average person using the technology easily and cheaply, and the industry that will ultimately enable (impossible at this point to even imagine), then making mining more profitable.

[tvbcof]

...
Our big fee-savings advantage is on the big-ticket transfers, not on puny sums.

-MarkM-

I disagree. More usage is more trust. More trust is more value in a crypto-currency.

More use of Bitcoin means significantly less trust to me.  The crypto-currency world is rife with participants who have difficulty running a reliable forum (and I like it just fine this way.)  The suggestion that the class of ordinary citizenry who currently move Bitcoin are going to be able to run a reliable economic solution at scale (with load-balanced processing clusters, high quality provisioned bandwidth, etc, etc) is absurd to me.  Throw in the potential to need to deal with coordinated state level attacks and the trust I would have in the solution is next to nill.

The alternative is that if the infrastructure required to operate the Bitcoin network shifts to well capitalized corporate entities.  My 'trust' in this solution is nearly nill as well.  And it is much less palatable from a philosophical level besides.

But to your point about big fee-savings advantage, fewer transactions = higher cost per transaction. Higher cost per transaction = less advantage over existing methods, which equals less value at all. It also means, fewer use cases. Why, on earth, do we want fewer use cases?

Almost certainly wrong.  The transaction cost at current scale is so close to zero that it is not worth thinking about.  The cost of performing a transaction is not related to the BTC size of the transaction.  Even at the current block size the Bitcoin solution could handle a fair part of the worlds need for high value transactions (I suspect.)  It's probably about perfect for balancing between other crypto-currency systems...accident???

The dream that my skittles purchase, your penny-ante Satoshi Dice bet, and someone's goat cheese sale in outer Mongolia coexist in a common block chain is nice and cool and all that, and I like it, but it is something I would trade for a robust solution which looks vaguely like what we see today.

[conv3rsion]

More use of Bitcoin means significantly less trust to me.  The crypto-currency world is rife with participants who have difficulty running a reliable forum (and I like it just fine this way.)  The suggestion that the class of ordinary citizenry who currently move Bitcoin are going to be able to run a reliable economic solution at scale (with load-balanced processing clusters, high quality provisioned bandwidth, etc, etc) is absurd to me.  Throw in the potential to need to deal with coordinated state level attacks and the trust I would have in the solution is next to nill.

The alternative is that if the infrastructure required to operate the Bitcoin network shifts to well capitalized corporate entities.  My 'trust' in this solution is nearly nill as well.  And it is much less palatable from a philosophical level besides.

You like it fine that people using the currency have a hard time with it? I don't even understand what you are saying. You don't want it to be run by "ordinary citizens" or "mega corps", you want it run by a select few with specialized skills. That's the definition of centralized.

But to your point about big fee-savings advantage, fewer transactions = higher cost per transaction. Higher cost per transaction = less advantage over existing methods, which equals less value at all. It also means, fewer use cases. Why, on earth, do we want fewer use cases?

Almost certainly wrong.  The transaction cost at current scale is so close to zero that it is not worth thinking about.  The cost of performing a transaction is not related to the BTC size of the transaction.  Even at the current block size the Bitcoin solution could handle a fair part of the worlds need for high value transactions (I suspect.)  It's probably about perfect for balancing between other crypto-currency systems...accident???

The dream that my skittles purchase, your penny-ante Satoshi Dice bet, and someone's goat cheese sale in outer Mongolia coexist in a common block chain is nice and cool and all that, and I like it, but it is something I would trade for a robust solution which looks vaguely like what we see today.

Its not wrong, its very simple math. A hard limit of 7 transactions per second means that once we hit our transaction bandwidth (already 25% there!), the price of individual transactions will go up. Free transactions will disappear and paid transactions will enter into an increasingly expensive auction.  If you think this is wrong, explain why.

I said nothing about the BTC size of the transaction. Also total value from transactions may end up going down because even with higher individual transaction prices, so many previously possible cheaper transactions will become impossible to perform, and total value (all fees paid on all transactions) may decrease by removing the ability for those other transactions to exist at all.

And with the individual transaction price going up, many use cases (micro transactions, retail transactions) will disappear. Huge numbers of individuals will not be able to participate as users at all (except perhaps to validate transactions that they themselves cannot take part in, but at least it works a 56.6k modem.... give me a fucking break). Centralization of ACTUAL USE will increase and total usage stops growing by the network limiting to 7 transactions per second. The remaining advantage over expensive wire transfers will be diminished as the price difference decreases. And ultimately the value of Bitcoin as a payment network and currency of the people will disappear.

The dream that my penny-ante dice bet, your skittles purchase, and someone's cheese sale in mongolia coexists in a common block chain IS THE REALITY TODAY. It needs to be the reality tomorrow as well. Anything less than that is total failure.

[tvbcof]

More use of Bitcoin means significantly less trust to me.  The crypto-currency world is rife with participants who have difficulty running a reliable forum (and I like it just fine this way.)  The suggestion that the class of ordinary citizenry who currently move Bitcoin are going to be able to run a reliable economic solution at scale (with load-balanced processing clusters, high quality provisioned bandwidth, etc, etc) is absurd to me.  Throw in the potential to need to deal with coordinated state level attacks and the trust I would have in the solution is next to nill.

The alternative is that if the infrastructure required to operate the Bitcoin network shifts to well capitalized corporate entities.  My 'trust' in this solution is nearly nill as well.  And it is much less palatable from a philosophical level besides.

You like it fine that people using the currency have a hard time with it? I don't even understand what you are saying. You don't want it to be run by "ordinary citizens" or "mega corps", you want it run by a select few with specialized skills. That's the definition of centralized.

I could have said it more clearly.

I favor a situation where the network is perfectly stable running on people's hardware who can barely afford a low-end car and can barely have the technical skills needed to keep a forum operational on the Internet.  That is what we have today and I don't want to lose it.

But to your point about big fee-savings advantage, fewer transactions = higher cost per transaction. Higher cost per transaction = less advantage over existing methods, which equals less value at all. It also means, fewer use cases. Why, on earth, do we want fewer use cases?

Almost certainly wrong.  The transaction cost at current scale is so close to zero that it is not worth thinking about.  The cost of performing a transaction is not related to the BTC size of the transaction.  Even at the current block size the Bitcoin solution could handle a fair part of the worlds need for high value transactions (I suspect.)  It's probably about perfect for balancing between other crypto-currency systems...accident???

The dream that my skittles purchase, your penny-ante Satoshi Dice bet, and someone's goat cheese sale in outer Mongolia coexist in a common block chain is nice and cool and all that, and I like it, but it is something I would trade for a robust solution which looks vaguely like what we see today.

Its not wrong, its very simple math. A hard limit of 7 transactions per second means that once we hit our transaction bandwidth (already 25% there!), the price of individual transactions will go up. Free transactions will disappear and paid transactions will enter into an increasingly expensive auction.  If you think this is wrong, explain why.

I think you are wrong for the following reasons.  The actual cost on a per-transaction basis is so low currently that the fees required to  operate profitably are not all that high.  Severe gouging by miners would destroy the system and/or draw out competition who wish to see the system survive.  In fact it might be the case that some people will fund mining simply to preserve the system for their own use rather than trying to make a huge profit on it.

But as the developers say, it is kind of hard to know exactly what will happen.  I'm all for finding out empirically.

I said nothing about the BTC size of the transaction. Also total value from transactions may end up going down because even with higher individual transaction prices, so many previously possible cheaper transactions will become impossible to perform, and total value (all fees paid on all transactions) may decrease by removing the ability for those other transactions to exist at all.

And with the individual transaction price going up, many use cases (micro transactions, retail transactions) will disappear. Huge numbers of individuals will not be able to participate as users at all (except perhaps to validate transactions that they themselves cannot take part in, but at least it works a 56.6k modem.... give me a fucking break). Centralization of ACTUAL USE will increase and total usage stops growing by the network limiting to 7 transactions per second. The remaining advantage over expensive wire transfers will be diminished as the price difference decreases. And ultimately the value of Bitcoin as a payment network and currency of the people will disappear.

The dream that my penny-ante dice bet, your skittles purchase, and someone's cheese sale in mongolia coexists in a common block chain IS THE REALITY TODAY. It needs to be the reality tomorrow as well. Anything less than that is total failure.

Going forward you will be able to live your dream and I will be able to live mine.  Which of us, if either, finds ourselves on the Bitcoin block-chain will be an interesting thing to find out.

As far as I am concerned Satoshi's main work is done;  peer-2-peer crypto-currencies are proven to be a workable concept and nobody can deny this.  I'm personally attached to Bitcoin because I have a lot of respect for certain parts of the community and separately because I have a significant 'speculative bet' on it.  But I'm not glued to Bitcoin specifically, and I will dump it if it turns sour.  Probably I am not alone.

[mpfrank]

Sorry, but I can't sai this often enough. Long term, Mining wont be providable.

I don't necessarily agree. It's obvious if the state of bitcoin remains like it is now, when the reward drops to zero security by mining will drop ridiculously low. There are only two ways to make it work (keeping the same security), either increase the fees or increase the number of transactions (with fees) per block. There are currently about 0.5 btc of transaction fees per block, so if the transaction rate would increase 40-fold that would take care of things.

I don't see any problem to begin with.  Just like any other business, the amount of competition in the mining business automatically increases to the point where it is just barely profitable enough to keep the most efficient operators in business.  It reaches an equilibrium, yes, but at that equilibrium point, the work still gets done and the surviving miners still make a living.