Why I am now not enthused about increasing the block size

[fuddudle]

Some people in the world don't have fast internet, or high storage. Also, i'm a user of Bitcoin Wallet on my phone, I can't imagine what it would be like to have to download 8x more data.

[CIYAM]

Ciyam.. as you are a person who was involved in the blockchain.info project, im certain that even you know that offchain tx's is a very easy thing to impliment, to send tx's instantly between users without touching the actual blocks..

I'm not sure what you mean by "involved in the blockchain.info project" (as I've never even used it as a wallet and I certainly never did any work for them) and of course offchain txs are a reasonable solution (I have not said that they aren't).

It is people that keep on insisting that Bitcoin should be used for all txs (big and very small) that I think are just not being realistic about what Bitcoin (at least in its current form) has been built to do and do well.

[Quickseller]

Apart from the slow confirmation times and the low TPS issues - for any average consumer what is the benefit of making an irreversible payment?

Benefit from an irreversible payment? Very little, however there are other benefits to the customer, primarily (current) low cost of use and the ability to send money instantly (yes, I would argue that bitcoin transactions are for all intensive purposes instant from the viewpoint of the customer, regardless of confirmation time)

Adding to that the complexity of merchants having to deal with txs that may never confirm due to minimum fees with full blocks you simply don't see anything even vaguely threatening to the likes of VISA or Mastercard.

If Child Pays for Parent (CPFP) was more widely adopted and supported by the miners as a whole then this would generally be a non-issue, and would be even less of an issue if blocks were sufficiently large such that they are not full even during peak periods.

If CPFP were more widely used then a merchant could receive a zero fee transaction from it's customer and then spend outputs from that transaction and include a sufficiently high tx fee so that both tx's get confirmed then the cost of receiving a BTC payment would essentially be absorbed by the merchant, which is the statuesque for Credit Card payments.   

Now I've thought about it more I think that those pushing for very large block sizes really are rather confused as to what they think this will achieve. If you can't reverse your payment (due to say not getting something delivered) then why would any online purchaser prefer to use Bitcoin (and talking about escrows is just silly as people don't use escrows when normally purchasing things with credit cards as they don't need to - you can generally just complain to your credit card provider and the merchant will lose out which is what most consumers prefer).

People would use Bitcoin to to pay for something online because of the lower cost. To counter the fact that payments cannot be reversed, consumers should only deal with companies that have a sufficient reputation that they can trust items will get delivered, for example if an Amazon order does not get delivered, I would think that the majority of the time Amazon can resolve this issue prior to the customer ever contacting their credit card company (yes this will lead to more centralization, and yes this will make it more difficult for newcomers to enter the market online).

Consumers can also use Bitcoin in person in similar ways that they use cash in person. If someone were to trust a mom and pop store to hand them $500 for a TV, then there is really no reason why they would not trust them with $500 worth of Bitcoin for the same TV. Such store could probably even offer a discount for paying with Bitcoin over cash because of the lower costs associated with protecting bitcoin over cash.

[Quickseller]

most things like cups of coffee in starbucks wont need confirms or reversibility.. because the coffee buyer can slap the barrista in the face with a wet fish if the customer doesnt get their coffee after payment.

Actually in the specific case of something like a cup of coffee it is the vendor who will be taking a big risk if blocks are near to full (be they 1MB or 10MB) as there are multiple threads about txs that never confirm.

I think this is exactly why we need larger blocks, so that there is no need to wait for multiple blocks for a tx to confirm, provided that a CPFP tx includes a sufficiently large tx fee.

How hard would it be to write a simple little wallet that purposely crafts txs that won't confirm (before they are dropped from the mempool)?

Not at all, however there are similar risks with accepting credit cards, however these risks can be mitigated with CPFP, and if the customer is still in the store then the manager can speak to the customer about the fact that their transaction is not going to confirm (which is not going to be the case for a customer whose "transaction" "fails" up to 180 days later when a credit card dispute is filed)

[CIYAM]

If CPFP were more widely used then a merchant could receive a zero fee transaction from it's customer and then spend outputs from that transaction and include a sufficiently high tx fee so that both tx's get confirmed then the cost of receiving a BTC payment would essentially be absorbed by the merchant, which is the statuesque for Credit Card payments.  

Whilst this may indeed be possible it is something that is not being really used at the moment and I think it will require specialised software to do this which basically means that merchants are still going to use payment processors rather than just say a Bitcoin client (personally I think that we don't want to even have payment processors in the future).

Basically I think that the 1 MB limit has actually been a good thing in showing us all the issues that Bitcoin as a payment system has (beyond the confirmation times).

Hopefully these issues will all be resolved over time and Bitcoin can become a realistic alternative to other payment systems but I don't think this is likely to happen for at least some years (but its other great use cases such as person to person transactions are functioning very well right now).

[brg444]

Apart from the slow confirmation times and the low TPS issues - for any average consumer what is the benefit of making an irreversible payment?

Benefit from an irreversible payment? Very little, however there are other benefits to the customer, primarily (current) low cost of use and the ability to send money instantly (yes, I would argue that bitcoin transactions are for all intensive purposes instant from the viewpoint of the customer, regardless of confirmation time)

Credit cards are costless to us and send money instantly as well. It's fairly obvious Bitcoin doesn't compete and certainly does not offer a superior product on these grounds.

People would use Bitcoin to to pay for something online because of the lower cost. To counter the fact that payments cannot be reversed, consumers should only deal with companies that have a sufficient reputation that they can trust items will get delivered, for example if an Amazon order does not get delivered, I would think that the majority of the time Amazon can resolve this issue prior to the customer ever contacting their credit card company (yes this will lead to more centralization, and yes this will make it more difficult for newcomers to enter the market online).

Consumers can also use Bitcoin in person in similar ways that they use cash in person. If someone were to trust a mom and pop store to hand them $500 for a TV, then there is really no reason why they would not trust them with $500 worth of Bitcoin for the same TV. Such store could probably even offer a discount for paying with Bitcoin over cash because of the lower costs associated with protecting bitcoin over cash.

It seems you pretty much confirm CIYAM's point which is that there is little incentive for the consumer

[VeritasSapere]

I think that Bitcoin is great for using as a payment system, which is why we should not allow the blocks to get consistently full because that renders transactions unreliable. I accept cryptocurrencies at a brick and mortar store I am responsible for and if the blocksize is not increased, I will have to stop accepting Bitcoin and I will only accept altcoins like DASH instead because of this issue. Transactions are instant and I do not wait for confirmations at this store, and for most other things I use payment processors like Bitpay, I found it to be both convenient and fast. I am also ideologically motivated to use Bitcoin instead of Fiat for as many things as I can, we should not underestimate this ideological motivation that some people might have.

Using Bitcoin as a payment system does not compromise its security or its decentralization, it actually strengthens it. Since giving Bitcoin more utility gives it more value which in turn increases its security. When adoption is increased decentralization is also increased because of more people running full nodes and mining. Bitcoin is like the gold coins of ancient times, both used as a commodity and a currency.

We should not arbitrarily restrict the blocksize. We should allow the blocks to grow as big as they need to while keeping the technological limitations in mind in order to preserve the principles of decentralization and financial freedom.

[CIYAM]

We should not arbitrarily restrict the blocksize. We should allow the blocks to grow as big as they need to while keeping the technological limitations in mind in order to preserve the principles of decentralization and financial freedom.

One thing I think you need to be very careful of when you decide upon this is just how getting rid of arbitrary restrictions could actually cause severe centralisation.

Already miners in China has resorted to using an SPV approach due to the GCF (Great Chinese Firewall) slowing down all international internet traffic. So if all limits are removed then you have just removed the world's 2nd (soon to be 1st) biggest economy from even participating in mining. Is that a good idea?

And if you keep going in that direction then Africa is killed and India then Australia and eventually Europe and South America and you are pretty much just left with the US and Canada (I am guessing the region where you probably come from).

[jonald_fyookball]

We should not arbitrarily restrict the blocksize. We should allow the blocks to grow as big as they need to while keeping the technological limitations in mind in order to preserve the principles of decentralization and financial freedom.

One thing I think you need to be very careful of when you decide upon this is just how getting rid of arbitrary restrictions could actually cause severe centralisation.

Already miners in China has resorted to using an SPV approach due to the GCF (Great Chinese Firewall) slowing down all international internet traffic. So if all limits are removed then you have just removed the world's 2nd (soon to be 1st) biggest economy from even participating in mining. Is that a good idea?

And if you keep going in that direction then Africa is killed and India and pretty much you are just left with the US (I am guessing where you probably come from).

It can argued that offchain solutions create more centralization than geographical distribution of high speed internet.

I don't think its an issue.  South Korea has faster internet than the US.

[CIYAM]

I don't think its an issue.  South Korea has faster internet than the US.

You have read that Thailand is looking to introduce its own version of the GCF and that other countries are starting to do the same (can't blame them as the NSA has been spying on everyone in the world so the rest of the world is now wanting to slow access down to the US and its allies)?

South Korea being still a big supporter of the US is unlikely to change its approach anytime soon but some of its neighbours are less likely to follow its lead IMO.

[jonald_fyookball]

I don't think its an issue.  South Korea has faster internet than the US.

You have read that Thailand is looking to introduce its own version of the GCF and that other countries are starting to do the same (can't blame them as the NSA has been spying on everyone in the world so the rest of the world is now wanting to slow access down to the US and its allies)?

South Korea being still a big supporter of the US is unlikely to change its approach anytime soon but some of its neighbours are less likely to follow its lead IMO.

I hear what you're saying, but if nodes are concentrated in the US, then those nodes are all going to be closer to one another.
I think limiting the blocksize to try to level the playing field with the rest of the world doesn't feel like its the right solution. 
Kind of like using a hammer to burst a pimple.

[CIYAM]

I think limiting the blocksize to try to level the playing field with the rest of the world doesn't feel like its the right solution.  

Well - I'm not sure what is the best solution but a US only solution would actually spell the end of my interest in Bitcoin for sure.

Bitcoin's best feature is that you can send txs to anyone, anywhere in the world - I'd hope that should remain its main focus (as nothing else does this as easily or as cheaply).

[Klestin]

If you can't reverse your payment (due to say not getting something delivered) then why would any online purchaser prefer to use Bitcoin (and talking about escrows is just silly as people don't use escrows when normally purchasing things with credit cards as they don't need to - you can generally just complain to your credit card provider and the merchant will lose out which is what most consumers prefer).

Bitcoin is fully capable of reversible transactions, with far more flexibility than existing methods such as credit cards, at vastly reduced cost, within a competitive open market.  See any of the many many talks on this subject.

[CIYAM]

Bitcoin is fully capable of reversible transactions, with far more flexibility than existing methods such as credit cards, at vastly reduced cost, within a competitive open market.  See any of the many many talks on this subject.

Huh?

I think you are not understanding how Bitcoin works (unless you think that reversible equals playing with tx malleability or low fees).

[jonald_fyookball]

I think limiting the blocksize to try to level the playing field with the rest of the world doesn't feel like its the right solution.  

Well - I'm not sure what is the best solution but a US only solution would actually spell the end of my interest in Bitcoin for sure.

Bitcoin's best feature is that you can send txs to anyone, anywhere in the world - I'd hope that should remain its main focus (as nothing else does this as easily or as cheaply).

And you still would be able to do that, even if mining nodes became geographically centralized. 

Also, keep in mind internet speed is just one factor.  Mining may become more centralized in
the place with the cheapest electrcity, or in the place that produces the cheapest ASICS.

[CIYAM]

Also, keep in mind internet speed is just one factor.  Mining may become more centralized in
the place with the cheapest electrcity, or in the place that produces the cheapest ASICS.

The cheapest ASIC equipment all comes from China - it also has very cheap electricity (which is why more than 50% of the mining power is in China now).

Iceland would have the cheapest electricity of all though (but it would need to import the ASIC from China which I understand has been happening).

Certainly I don't want to see Bitcoin being centralised to any one country so I do hope we'll see mining being performed throughout the world.

[brg444]

Also, keep in mind internet speed is just one factor.  Mining may become more centralized in
the place with the cheapest electrcity, or in the place that produces the cheapest ASICS.

The cheapest ASIC equipment all comes from China - it also has very cheap electricity (which is why more than 50% of the mining power is in China now).

Iceland would have the cheapest electricity of all though (but it would need to import the ASIC from China which I understand has been happening).

Certainly I don't want to see Bitcoin being centralised to any one country so I do hope we'll see mining being performed throughout the world.

What's funny is that in a scenario where block size are increased and cause propagation problems it's the rest of the network that's gonna be shit out of luck, not the chineses.

[VeritasSapere]

Also, keep in mind internet speed is just one factor.  Mining may become more centralized in
the place with the cheapest electrcity, or in the place that produces the cheapest ASICS.

The cheapest ASIC equipment all comes from China - it also has very cheap electricity (which is why more than 50% of the mining power is in China now).

Iceland would have the cheapest electricity of all though (but it would need to import the ASIC from China which I understand has been happening).

Certainly I don't want to see Bitcoin being centralised to any one country so I do hope we'll see mining being performed throughout the world.

One thing I think you need to be very careful of when you decide upon this is just how getting rid of arbitrary restrictions could actually cause severe centralisation.

Already miners in China has resorted to using an SPV approach due to the GCF (Great Chinese Firewall) slowing down all international internet traffic. So if all limits are removed then you have just removed the world's 2nd (soon to be 1st) biggest economy from even participating in mining. Is that a good idea?

And if you keep going in that direction then Africa is killed and India then Australia and eventually Europe and South America and you are pretty much just left with the US and Canada (I am guessing the region where you probably come from).

Bitmain does have competition from Spondoolies, Bitfury and KNC who all are all not based in Asia. I think that the biggest threat to mining centralization today comes from the centralization of manufacturing, which is a very different issue. Bitcoin is free and open, mining can be carried out anywhere in the world anonymously regardless of block size. Since the pools can be located in high bandwidth locations with good block propagation. These pools do not even need to be located in the same country as the miners since the data that the miners send to the pools is extremely small and is also not effected by blocksize. This is why other countries and locations will not be excluded from any of the critical aspects of Bitcoin since mining today is done using pools, and full nodes that are not used for mining can still function well in these countries even with much larger blocks. This is also why I think that increasing the block size will not lead to increased centralization compared to keeping the one megabyte restriction in place. Since keeping this limit in place will introduce centralization in the form of increased reliance on third parties.

[CIYAM]

Personally I am beginning to wonder whether the rules about coin age and dust should be re-evaluated as what has mostly been seen when the blocks are near to full is loads of tiny txs (as this type of attack can still occur even with larger blocks).

[spartacusrex]

This is also why I think that increasing the block size will not lead to increased centralization compared to keeping the one megabyte restriction in place. Since keeping this limit in place will introduce centralization in the form of increased reliance on third parties.

This.

..

I use Bitcoin... a lot.

Yes, I keep some stashed away, but most of the time I spend them. On things that I can ONLY buy with Bitcoins.

No one ever seems to mention that THIS is Bitcoin's Killer Feature.  

And before anyone gets judgemental, please don't. Freedom is all that matters to me. And if you have a problem with Freedom, we're from different planets, and speak different languages. So I won't understand your incoherent babbling.

Will I be able to spend my coins WHEREEVER I want if I have to use a centralised 3rd Party ? No Chance.

As far as I am concerned, keeping the 1MB limit (or to be more precise, restricting the TX throughput to tiny levels) is the DEATH of the Bitcoin I have known and loved. Simple as that...  

As for mining centralisation, I hate to break it to you, but it's already happening. So stop saying that increasing the limit will cause it. Unless we have different definitions of cause and effect.

..

On a cheer-ier note - There is a solution! There's always a solution. We just haven't figured it out yet.  

Hmm.. Maybe we can please BOTH sides..

{Spartacus throws an idea into the fighting pit}  : (Only include the TXID in Blocks ? Then you can fit 1,000,000 / 32 = 31,250 TXs per block.. roughly.. Although these txs would still need to be transmitted across the whole network..)