9 ms·

The elephant in the Bitcoin room is that it simply cannot scale up to high TPS rates. If any kind of meaningful adoption actually happened, the network would ke

by inoop 12y ago

The elephant in the Bitcoin room is that it simply cannot scale up to high TPS rates. If any kind of meaningful adoption actually happened, the network would keel over pretty quickly trying to globally broadcast all that data, and the blockchain would quickly balloon to unmanageable sizes. I see a lot of these articles hailing Bitcoin as the answer to some monetary injustice or other, or as the fundamental underpinning of some revolutionary new product, but people totally ignore Bitcoin's technological limitations. Instead, the blockchain is talked about as if it's some magical box that is pretty much whatever the author needs it to be. The amount of cognitive dissonance in the Bitcoin community regarding this topic is baffling. The whole point of side chains was that even core devs realized that Bitcoin can not scale. Yet side chains were hailed by the community as some sort of fantastic innovation that is 'good for bitcoin'. Maybe these people are just talking their book, I don't know, but next time you read an article about Bitcoin or blockchain tech just realize that VISA can handle 47.000 tps. Bitcoin currently is doing slightly more than 1. Inb4 'merkle trees', 'invertible bloom filters', 'something something moore's law', 'corporate shill', 'freedom hater', 'the market will sort it out', 'side chains', etc. </offtopic rant>

mrb 12y ago
Firstly, Bitcoin does not need to scale that much to be considered widely successful. Heck, Western Union[1] averages only 10 tps, which is only 10x what Bitcoin is doing today, so such a target is clearly within reach with a 2MB block size. Secondly, Bitcoin can actually scale by many orders of magnitudes, see the math and arguments there: https://en.bitcoin.it/wiki/Scalability https://en.bitcoin.it/wiki/Scalability You need solid counterarguments to this page to convince us it can't possibly scale. Thirdly, Bitcoin has done 6 years in a row what critics have always said it cannot do ("nobody would want bitcoins", "the exchange rate will never reach $1/$10/$100", "it will crash and be forgotten after the Jun 2011/Mar 2013/Nov 2013 bubble", "the US govt will make it illegal", "big companies will never accept it", etc). So you need to sit back and start thinking why is it that Bitcoin keeps growing and reaching milestone after milestone, instead of throwing one more empty statement "bitcoin can't do X" with zero reasoning behind. [1] WU does ~250 million consumer transaction/year. Edit: @inoop I misremembered and meant WU, not Paypal
- inoop 12y ago
  From your link: > PayPal, in contrast, handles around 10 million transactions per day for an average of 115 tps.
  - cgjaro 12y ago
    As the OP corrected, it's Western Union that does 10 tps. But even 100 tps would be easily within reach if the block size is bumped to 20MB, which is not that crazy.
- andrea_s 12y ago
  Isn't WU a not-so-relevant service that's mostly used by scammers? I'm not from the US, and I've always heard of it in such terms...
  - theklub 12y ago
    Pretty much, although in some areas its used more legitimately than others.
  - IkmoIkmo 12y ago
    That, and one of the biggest facilitators of remittance, a significant source of income for some of the poorest people on the planet. That may not be very relevant in SV, but they affect a lot of lives positively. At the same time, they charge exorbitant fees, and remittance fees are stuck at around 10%. I'm looking forward to the day they're as redundant as an expensive long-distance phone call is made redundant, which is possible when you have native digital money like we have digital mail or telephony. Bitcoin to me is the most compelling proposal for this to this date.
- downandout 12y ago
  > Thirdly, Bitcoin has done 6 years in a row what critics have always said it cannot do The one thing that hasn't changed a great deal is accessibility to them. When buying, you are either subject to waiting periods, or have to conduct the transaction in cash in person. No one in their right mind will take credit cards or PayPal for them. I believe this is the major hurdle to adoption of crypto by the mainstream public. Writing tools to make BTC easy to work with is the easy part. Until they can be bought at Walmart or your local check cashing store, it's going to be a fairly rough road for new adoption.
moe 12y ago
The elephant in the Bitcoin room is that it simply cannot scale up to high TPS rates. What is it about Bitcoin that makes seemingly everyone, no matter how unqualified, re-iterate the same naive knee-jerk trite at no end, at every opportunity? Can we please stop repeating the same old debates on every article with Bitcoin in the title?
- IkmoIkmo 12y ago
  Especially if it's not even true. I'm fine with repeating a debate, as long as it's not a silly one. But the core developers of bitcoin as the developers at businesses built ontop of bitcoin, as a number of independent academics simply don't agree that it can't scale. But yeah it's getting a little old. It's like someone posts a thread about some new solar invention, and the top comment is 'but the elephant in the room is that there is no global warming and it takes more energy to build a solar panel than it can ever generate' and we all sigh. This is a little less wrong than that, but the notion that bitcoin can't scale is simply false too, and and the topic itself is a lot more complex than 'it can' or 'it can't'.
- inoop 12y ago
  Respectfully, what makes you say I'm not qualified?
  - moe 12y ago
    The fact that you attack a strawman, oblivious of all prior discussion on the subject. The protocol is not set in stone. Scalability is a work in progress. https://bitcoinfoundation.org/2014/10/a-scalability-roadmap/ https://bitcoinfoundation.org/2014/10/a-scalability-roadmap/
oco101 12y ago
DPOS could do 10000 TPS. This is the blockchain with the Visa like speed.
DennisP 12y ago
Invertible bloom filters are an elegant and simple solution to one of the main limits to bitcoin scalability: the time it takes to propagate solved blocks. Once that's fixed, we can use as much bandwidth as available. The VISA average of 2000 tps could be handled with under 10 megabits/sec, available on residential connections. As home gigabit rolls out, we can go higher. Blockchain storage would still be a problem, but pruning old empty addresses will help a lot with that.
IkmoIkmo 12y ago
Inb4 'merkle trees', 'invertible bloom filters', 'something something moore's law', 'the market will sort it out', 'side chains', etc. Really? Could you at least address why these things don't allow bitcoin to scale? It's pretty unproductive to criticize something, then say inb4 all the reasons why most criticism is invalid. It's like saying 'Humans are stupid because a child can't do math. inb4 comments about how adults can'. I mean why even post that? Point is, 1mb limits are arbitrary, which is where the 1 TPS comes from, 1 TPS is how much transaction data fits in a 1 megabyte block. You know this, so say it. And that 1 megabyte block can go to 10 megabytes. Why not to 1 billion megabytes? Because it's expensive to process that data (network to transfer it, storage to store it, CPU to verify signatures). And it's a PERFECTLY sensible thing to then say, that moore's law affects the ability to verify more transaction signatures. Or that Kryder's law affects the ability to store more data. Or that Nielsen's law allows networks to transfer more data. How childish is it to say 'inb4 perfectly fine reasons why my criticism is more nuanced'. And yes, the blockchain can be pruned, requiring less storage. And yes filters can be used to require less data to be transferred. And yes it is possible to run light-nodes, where you only store some transactions and not others, so that you can make a small contribution to decentralisation. But look at the current 1 MB blocks. That's 1024 kilobytes every 600 seconds, or 1.7 kb per second. Now look at an example, take Bitpay, they have made public that they process close to half a billion dollars a year. (yes, peanuts compared to VISA, don't worry I know). And they've done this by utilizing a fraction of the bitcoin blockchain. Which itself is at a fraction of its current arbitrary 1 MB block capacity. So a company that utilizes a fraction of the blockchain, which is at a fraction of full capacity, which utilizes 1.7 kb/s of network, is powering half a billion dollars of money processed per year, without any optimizations, and you're telling me that this can't scale up today, nor can it scale up up in the next 10 years despite moore, kryder, nielsen's laws, even after optimizations? It's simply not true. I won't stand here and say bitcoin can scale to 1 million VISA networks, no. I know there are limitations, and I appreciate that a distributed, redundant, decentralized system is more expensive than having to run a single centralized node. I appreciate that. But this notion that bitcoin can't scale at all to any significant TPS period is so devoid of any NUANCE. And your comment about how core devs realized bitcoin can't scale. I'll leave you with these concluding remarks by Gavin Andresen, Chief Scientist of bitcoin's core development who virtually completely disagrees with you. (in a nuanced way with caveats) [0]. Doesn't make him right and you wrong in and of itself (although he's both a very involved guy, knows what he's talking about, but isn't on the hype train. i.e. he still calls bitcoin 'an experiment'.) but at least the notion that bitcoin core devs don't think it can scale is, again, a lot more nuanced. As for the corporate shill, freedom hater etc stuff. I hear you, that kind of sentiment from some bitcoiners got old really fast. But it's not nearly as bad as a few years ago. A lot of cool sensible middle-ground folk have arrived who get as tired by a 'you're a corporate shill so your comment is invalid' remark by a crazy bitcoiner as a 'you probably think I'm a corporate shill, so you're a bitcoin nut whose comment is invalid' remark by people like you. [0] Comments by Gavin Andresen on scalability: https://bitcoinfoundation.org/2014/10/a-scalability-roadmap/ https://bitcoinfoundation.org/2014/10/a-scalability-roadmap/ There is a clear path to scaling up the network to handle several thousand transactions per second (“Visa scale”). Getting there won’t be trivial, because writing solid, secure code takes time and because getting consensus is hard. Fortunately technological progress marches on, and Nielsen’s Law of Internet Bandwidth and Moore’s Law make scaling up easier as time passes. The map gets fuzzy if we start thinking about how to scale faster than the 50%-per-increase-in-bandwidth-per-year of Nielsen’s Law. Some complicated scheme to avoid broadcasting every transaction to every node is probably possible to implement and make secure enough. But 50% per year growth is really good. According to my rough back-of-the-envelope calculations, my above-average home Internet connection and above-average home computer could easily support 5,000 transactions per second today. That works out to 400 million transactions per day. Pretty good; every person in the US could make one Bitcoin transaction per day and I’d still be able to keep up. After 12 years of bandwidth growth that becomes 56 billion transactions per day on my home network connection — enough for every single person in the world to make five or six bitcoin transactions every single day. It is hard to imagine that not being enough; according the the Boston Federal Reserve, the average US consumer makes just over two payments per day. So even if everybody in the world switched entirely from cash to Bitcoin in twenty years, broadcasting every transaction to every fully-validating node won’t be a problem

by inoop 12y ago

mrb 12y ago

Firstly, Bitcoin does not need to scale that much to be considered widely successful. Heck, Western Union[1] averages only 10 tps, which is only 10x what Bitcoin is doing today, so such a target is clearly within reach with a 2MB block size. Secondly, Bitcoin can actually scale by many orders of magnitudes, see the math and arguments there: https://en.bitcoin.it/wiki/Scalability https://en.bitcoin.it/wiki/Scalability You need solid counterarguments to this page to convince us it can't possibly scale. Thirdly, Bitcoin has done 6 years in a row what critics have always said it cannot do ("nobody would want bitcoins", "the exchange rate will never reach $1/$10/$100", "it will crash and be forgotten after the Jun 2011/Mar 2013/Nov 2013 bubble", "the US govt will make it illegal", "big companies will never accept it", etc). So you need to sit back and start thinking why is it that Bitcoin keeps growing and reaching milestone after milestone, instead of throwing one more empty statement "bitcoin can't do X" with zero reasoning behind. [1] WU does ~250 million consumer transaction/year. Edit: @inoop I misremembered and meant WU, not Paypal

inoop 12y ago
From your link: > PayPal, in contrast, handles around 10 million transactions per day for an average of 115 tps.
- cgjaro 12y ago
  As the OP corrected, it's Western Union that does 10 tps. But even 100 tps would be easily within reach if the block size is bumped to 20MB, which is not that crazy.
andrea_s 12y ago
Isn't WU a not-so-relevant service that's mostly used by scammers? I'm not from the US, and I've always heard of it in such terms...
- theklub 12y ago
  Pretty much, although in some areas its used more legitimately than others.
- IkmoIkmo 12y ago
  That, and one of the biggest facilitators of remittance, a significant source of income for some of the poorest people on the planet. That may not be very relevant in SV, but they affect a lot of lives positively. At the same time, they charge exorbitant fees, and remittance fees are stuck at around 10%. I'm looking forward to the day they're as redundant as an expensive long-distance phone call is made redundant, which is possible when you have native digital money like we have digital mail or telephony. Bitcoin to me is the most compelling proposal for this to this date.
downandout 12y ago
> Thirdly, Bitcoin has done 6 years in a row what critics have always said it cannot do The one thing that hasn't changed a great deal is accessibility to them. When buying, you are either subject to waiting periods, or have to conduct the transaction in cash in person. No one in their right mind will take credit cards or PayPal for them. I believe this is the major hurdle to adoption of crypto by the mainstream public. Writing tools to make BTC easy to work with is the easy part. Until they can be bought at Walmart or your local check cashing store, it's going to be a fairly rough road for new adoption.

moe 12y ago

The elephant in the Bitcoin room is that it simply cannot scale up to high TPS rates. What is it about Bitcoin that makes seemingly everyone, no matter how unqualified, re-iterate the same naive knee-jerk trite at no end, at every opportunity? Can we please stop repeating the same old debates on every article with Bitcoin in the title?

IkmoIkmo 12y ago
Especially if it's not even true. I'm fine with repeating a debate, as long as it's not a silly one. But the core developers of bitcoin as the developers at businesses built ontop of bitcoin, as a number of independent academics simply don't agree that it can't scale. But yeah it's getting a little old. It's like someone posts a thread about some new solar invention, and the top comment is 'but the elephant in the room is that there is no global warming and it takes more energy to build a solar panel than it can ever generate' and we all sigh. This is a little less wrong than that, but the notion that bitcoin can't scale is simply false too, and and the topic itself is a lot more complex than 'it can' or 'it can't'.
inoop 12y ago
Respectfully, what makes you say I'm not qualified?
- moe 12y ago
  The fact that you attack a strawman, oblivious of all prior discussion on the subject. The protocol is not set in stone. Scalability is a work in progress. https://bitcoinfoundation.org/2014/10/a-scalability-roadmap/ https://bitcoinfoundation.org/2014/10/a-scalability-roadmap/

oco101 12y ago

DPOS could do 10000 TPS. This is the blockchain with the Visa like speed.

DennisP 12y ago

Invertible bloom filters are an elegant and simple solution to one of the main limits to bitcoin scalability: the time it takes to propagate solved blocks. Once that's fixed, we can use as much bandwidth as available. The VISA average of 2000 tps could be handled with under 10 megabits/sec, available on residential connections. As home gigabit rolls out, we can go higher. Blockchain storage would still be a problem, but pruning old empty addresses will help a lot with that.

IkmoIkmo 12y ago

Inb4 'merkle trees', 'invertible bloom filters', 'something something moore's law', 'the market will sort it out', 'side chains', etc. Really? Could you at least address why these things don't allow bitcoin to scale? It's pretty unproductive to criticize something, then say inb4 all the reasons why most criticism is invalid. It's like saying 'Humans are stupid because a child can't do math. inb4 comments about how adults can'. I mean why even post that? Point is, 1mb limits are arbitrary, which is where the 1 TPS comes from, 1 TPS is how much transaction data fits in a 1 megabyte block. You know this, so say it. And that 1 megabyte block can go to 10 megabytes. Why not to 1 billion megabytes? Because it's expensive to process that data (network to transfer it, storage to store it, CPU to verify signatures). And it's a PERFECTLY sensible thing to then say, that moore's law affects the ability to verify more transaction signatures. Or that Kryder's law affects the ability to store more data. Or that Nielsen's law allows networks to transfer more data. How childish is it to say 'inb4 perfectly fine reasons why my criticism is more nuanced'. And yes, the blockchain can be pruned, requiring less storage. And yes filters can be used to require less data to be transferred. And yes it is possible to run light-nodes, where you only store some transactions and not others, so that you can make a small contribution to decentralisation. But look at the current 1 MB blocks. That's 1024 kilobytes every 600 seconds, or 1.7 kb per second. Now look at an example, take Bitpay, they have made public that they process close to half a billion dollars a year. (yes, peanuts compared to VISA, don't worry I know). And they've done this by utilizing a fraction of the bitcoin blockchain. Which itself is at a fraction of its current arbitrary 1 MB block capacity. So a company that utilizes a fraction of the blockchain, which is at a fraction of full capacity, which utilizes 1.7 kb/s of network, is powering half a billion dollars of money processed per year, without any optimizations, and you're telling me that this can't scale up today, nor can it scale up up in the next 10 years despite moore, kryder, nielsen's laws, even after optimizations? It's simply not true. I won't stand here and say bitcoin can scale to 1 million VISA networks, no. I know there are limitations, and I appreciate that a distributed, redundant, decentralized system is more expensive than having to run a single centralized node. I appreciate that. But this notion that bitcoin can't scale at all to any significant TPS period is so devoid of any NUANCE. And your comment about how core devs realized bitcoin can't scale. I'll leave you with these concluding remarks by Gavin Andresen, Chief Scientist of bitcoin's core development who virtually completely disagrees with you. (in a nuanced way with caveats) [0]. Doesn't make him right and you wrong in and of itself (although he's both a very involved guy, knows what he's talking about, but isn't on the hype train. i.e. he still calls bitcoin 'an experiment'.) but at least the notion that bitcoin core devs don't think it can scale is, again, a lot more nuanced. As for the corporate shill, freedom hater etc stuff. I hear you, that kind of sentiment from some bitcoiners got old really fast. But it's not nearly as bad as a few years ago. A lot of cool sensible middle-ground folk have arrived who get as tired by a 'you're a corporate shill so your comment is invalid' remark by a crazy bitcoiner as a 'you probably think I'm a corporate shill, so you're a bitcoin nut whose comment is invalid' remark by people like you. [0] Comments by Gavin Andresen on scalability: https://bitcoinfoundation.org/2014/10/a-scalability-roadmap/ https://bitcoinfoundation.org/2014/10/a-scalability-roadmap/ There is a clear path to scaling up the network to handle several thousand transactions per second (“Visa scale”). Getting there won’t be trivial, because writing solid, secure code takes time and because getting consensus is hard. Fortunately technological progress marches on, and Nielsen’s Law of Internet Bandwidth and Moore’s Law make scaling up easier as time passes. The map gets fuzzy if we start thinking about how to scale faster than the 50%-per-increase-in-bandwidth-per-year of Nielsen’s Law. Some complicated scheme to avoid broadcasting every transaction to every node is probably possible to implement and make secure enough. But 50% per year growth is really good. According to my rough back-of-the-envelope calculations, my above-average home Internet connection and above-average home computer could easily support 5,000 transactions per second today. That works out to 400 million transactions per day. Pretty good; every person in the US could make one Bitcoin transaction per day and I’d still be able to keep up. After 12 years of bandwidth growth that becomes 56 billion transactions per day on my home network connection — enough for every single person in the world to make five or six bitcoin transactions every single day. It is hard to imagine that not being enough; according the the Boston Federal Reserve, the average US consumer makes just over two payments per day. So even if everybody in the world switched entirely from cash to Bitcoin in twenty years, broadcasting every transaction to every fully-validating node won’t be a problem