Once again, what about the scalability issue?

[DeathAndTaxes]

It is important that average block size not exceed what a miner on an average connection can broadcast to peers in a reasonable amount of time (3 seconds = ~0.5% orphan rate) on an average public internet connection.  Granted those are pretty vague terms.  Obviously 1MB block is below that critical level and 1GB block is obviously above it.  Is 100MB fine today? How about in 10 years? Is 5MB fine today? How about 2MB and doubling every two years? I am not saying I have the answers but that is the kind of thing we (community at large not just Gavin et all) need to think about critically before saying "yeah lets go to unlimited blocks and let the market figure it out".  I have no doubt the market will figure it out however one might not like what end state it reaches.

What makes you think the market won't take care of the average miner, such as by limiting blocksize normatively? Any claim that the market won't take care of something should be justified by specifying how you think the market is broken for that function - because the norm is for markets to work, even if we can't immediately see how.

I never said the market wouldn't take care of it, just that you might not like the outcome.

One outcome would be block sizes becomes so large that pools >51% of hashing power run private links to each other in the same datacenter in order to keep orphans out of line.  That drives other pools out of business, the major 3-4 or pools/solo-corps grow even larger and decide to optimization of the network to simply exclude the blocks of any pool/solo not in their organization.  That is one way of the market "taking care" of the problem that demand for higher tx volume exceeds the capabilities of public peer to peer links.  The market doesn't necessarily care about the advantages of decentralization, transparency, fair play, and the "solution" arrived may not be one that most bitcoiners find desirable.  It isn't a prediction, it is a risk that is all I am saying.  As long as avg block size is low relative to bandwidth capacity between miners on open public peer to peer links there is no catalyst for such a move, there is no problem for the market to "solve".

[bytemaster]

It is important that average block size not exceed what a miner on an average connection can broadcast to peers in a reasonable amount of time (3 seconds = ~0.5% orphan rate) on an average public internet connection.  Granted those are pretty vague terms.  Obviously 1MB block is below that critical level and 1GB block is obviously above it.  Is 100MB fine today? How about in 10 years? Is 5MB fine today? How about 2MB and doubling every two years? I am not saying I have the answers but that is the kind of thing we (community at large not just Gavin et all) need to think about critically before saying "yeah lets go to unlimited blocks and let the market figure it out".  I have no doubt the market will figure it out however one might not like what end state it reaches.

What makes you think the market won't take care of the average miner, such as by limiting blocksize normatively? Any claim that the market won't take care of something should be justified by specifying how you think the market is broken for that function - because the norm is for markets to work, even if we can't immediately see how.

I never said the market wouldn't take care of it, just that you might not like the outcome.

One outcome would be block sizes becomes so large that pools >51% of hashing power run private links to each other in the same datacenter in order to keep orphans out of line.  That drives other pools out of business.  That is the market "taking care of it", the market doens't necessarily care about the advantages of decentralization.  It is a risk that is all I am saying.  It isn't a prediction just pointing out that orphans are lost revenue and if public peer to peer internet links can't keep orphans low there are more centralized "solutions" to that problem.

The market certainly cares about decentralization because decentralization increases competition and profit opportunities.   Bitcoin would have no value if the market didn't value decentralization.   Mining pools already self regulate to prevent growing much over 40%.    And of course, alt-coins will force bitcoin to adapt and evolve.   Bitcoin will evolve or die and that is the market at work.

[edmundedgar]

Bandwidth is more critical than disk space for decentralization. 

Parallel chains of with a fixed limit on bandwidth per chain would be nice.

The ability to move value between chains would also be nice.

To move value between chains means a 3rd chain is required that is merge-mined with the other 2 chains.  This 3rd chain is responsible for confirming movements from one chain to the other and vice versa.   The cross-chain would have to be very low bandwidth, say 1% of bandwidth allocated for the main chains. 

With this approach you could have decentralization while still having only one crypto-currency.

Strictly speaking I'm not sure you'd even need the ability to be able to move value between chains to shard Bitcoin into multiple chains for the same currency. Multiple chains, all of equal value, coins always stay on the chain where they started, your address works on all the chains. If I pay you some coins, you shouldn't care which chain I've paid you on. Your money is recorded in a permanent ledger, who cares which one?

That said, the ability to move across chains as you describe would be a nice way to guarantee that people don't get funny ideas about coins on Chain X being worth more than coins on Chain Y.

[bytemaster]

Bandwidth is more critical than disk space for decentralization. 

Parallel chains of with a fixed limit on bandwidth per chain would be nice.

The ability to move value between chains would also be nice.

To move value between chains means a 3rd chain is required that is merge-mined with the other 2 chains.  This 3rd chain is responsible for confirming movements from one chain to the other and vice versa.   The cross-chain would have to be very low bandwidth, say 1% of bandwidth allocated for the main chains. 

With this approach you could have decentralization while still having only one crypto-currency.

Strictly speaking I'm not sure you'd even need the ability to be able to move value between chains to shard Bitcoin into multiple chains for the same currency. Multiple chains, all of equal value, coins always stay on the chain where they started, your address works on all the chains. If I pay you some coins, you shouldn't care which chain I've paid you on. Your money is recorded in a permanent ledger, who cares which one?

That said, the ability to move across chains as you describe would be a nice way to guarantee that people don't get funny ideas about coins on Chain X being worth more than coins on Chain Y.

Each chain would have value based upon the 'network effect'... people already have funny ideas about Chain X and Chain Y ... aka Litecoin.  These ideas are not so funny, because if anyone could launch a new chain then you could print your own money.

[edmundedgar]

Bandwidth is more critical than disk space for decentralization. 

Parallel chains of with a fixed limit on bandwidth per chain would be nice.

The ability to move value between chains would also be nice.

To move value between chains means a 3rd chain is required that is merge-mined with the other 2 chains.  This 3rd chain is responsible for confirming movements from one chain to the other and vice versa.   The cross-chain would have to be very low bandwidth, say 1% of bandwidth allocated for the main chains. 

With this approach you could have decentralization while still having only one crypto-currency.

Strictly speaking I'm not sure you'd even need the ability to be able to move value between chains to shard Bitcoin into multiple chains for the same currency. Multiple chains, all of equal value, coins always stay on the chain where they started, your address works on all the chains. If I pay you some coins, you shouldn't care which chain I've paid you on. Your money is recorded in a permanent ledger, who cares which one?

That said, the ability to move across chains as you describe would be a nice way to guarantee that people don't get funny ideas about coins on Chain X being worth more than coins on Chain Y.

Each chain would have value based upon the 'network effect'... people already have funny ideas about Chain X and Chain Y ... aka Litecoin.  These ideas are not so funny, because if anyone could launch a new chain then you could print your own money.

Sure, but if you start with a simple technical scalabilty change, presumably by arbitrarily splitting outputs on an existing chain into two shards, it's not obvious that people would treat them as separate networks with different values, rather than different parts of a single network. For most purposes the end user wouldn't need to know anything about different shards - their client would just show them their total balance. The same addresses would work on both shards, and as a vendor when you asked a customer to pay address xyz you wouldn't know which shard their payment would be coming in on, so you'd actually have to work quite hard to set different prices for different shards and communicate the different prices to your customers. If coins on the two shards buy the same thing, the coins on the shards are worth the same.

In that situation it would be fairly mad to try to ascribe different values to different shards, but obviously it's helpful to have a bit of protection against somebody getting a mad idea...

[jubalix]

doesn't electrum solve this already?

[Come-from-Beyond]

doesn't electrum solve this already?

Is it trustless?

[tvbcof]

doesn't electrum solve this already?

Is it trustless?

I've not studied Electrum extensively, but I believe it is 'trustless' in that someone running a server cannot steal your actual Bitcoin.  But the face value in BTC is only one aspect of 'value', and other information gleaned in running a server (Electrum or SPV) can be monetized at the expense of the clients.

Any entity can be put under enough pressure to either comply with government mandates (US government mandates in much of the world) or shut down and lose their infrastructure investment.  All of the PRISM participants chose the latter...it's the only logical choice.  And the only legal choice for corporations.

When the actual foundation of Bitcoin requires resources which exceed what enthusiasts can muster in order to be operated realistically it will be highly susceptible to becoming yet another cog in the state surveillance machine.  It will be either allowed to operate in order to siphon intelligence from the users or effectively shut down as a usable currency solution.  I'll bet on that...in terms of how I manage my BTC stash that is.

[DeathAndTaxes]

doesn't electrum solve this already?

Is it trustless?

I've not studied Electrum extensively, but I believe it is 'trustless' in that someone running a server cannot steal your actual Bitcoin.  But the face value in BTC is only one aspect of 'value', and other information gleaned in running a server (Electrum or SPV) can be monetized at the expense of the clients.

Any entity can be put under enough pressure to either comply with government mandates (US government mandates in much of the world) or shut down and lose their infrastructure investment.  All of the PRISM participants chose the latter...it's the only logical choice.  And the only legal choice for corporations.

When the actual foundation of Bitcoin requires resources which exceed what enthusiasts can muster in order to be operated realistically it will be highly susceptible to becoming yet another cog in the state surveillance machine.  It will be either allowed to operate in order to siphon intelligence from the users or effectively shut down as a usable currency solution.  I'll bet on that...in terms of how I manage my BTC stash that is.

Learn about bloom filters.

[tvbcof]

doesn't electrum solve this already?

Is it trustless?

I've not studied Electrum extensively, but I believe it is 'trustless' in that someone running a server cannot steal your actual Bitcoin.  But the face value in BTC is only one aspect of 'value', and other information gleaned in running a server (Electrum or SPV) can be monetized at the expense of the clients.

Any entity can be put under enough pressure to either comply with government mandates (US government mandates in much of the world) or shut down and lose their infrastructure investment.  All of the PRISM participants chose the latter...it's the only logical choice.  And the only legal choice for corporations.

When the actual foundation of Bitcoin requires resources which exceed what enthusiasts can muster in order to be operated realistically it will be highly susceptible to becoming yet another cog in the state surveillance machine.  It will be either allowed to operate in order to siphon intelligence from the users or effectively shut down as a usable currency solution.  I'll bet on that...in terms of how I manage my BTC stash that is.

Learn about bloom filters.

Are you saying that you can use them to make confusing pulls in an attempt to complicate analysis?  Good luck with that.

[DeathAndTaxes]

doesn't electrum solve this already?

Is it trustless?

I've not studied Electrum extensively, but I believe it is 'trustless' in that someone running a server cannot steal your actual Bitcoin.  But the face value in BTC is only one aspect of 'value', and other information gleaned in running a server (Electrum or SPV) can be monetized at the expense of the clients.

Any entity can be put under enough pressure to either comply with government mandates (US government mandates in much of the world) or shut down and lose their infrastructure investment.  All of the PRISM participants chose the latter...it's the only logical choice.  And the only legal choice for corporations.

When the actual foundation of Bitcoin requires resources which exceed what enthusiasts can muster in order to be operated realistically it will be highly susceptible to becoming yet another cog in the state surveillance machine.  It will be either allowed to operate in order to siphon intelligence from the users or effectively shut down as a usable currency solution.  I'll bet on that...in terms of how I manage my BTC stash that is.

Learn about bloom filters.

Are you saying that you can use them to make confusing pulls in an attempt to complicate analysis?  Good luck with that.

SPV nodes don't pull a single address or transaction.  It is more like "please send me all tx after block X which involve addresses starting with 1A".  The filter can be set as wide as one want, all the way up to everything (send me all tx from all address) as a full node would.  It is a tradeoff between privacy and bandwidth.  There is no need to even limit yourself to when you actually need tx data.  You can send a request for any subset of tx at any time to any node.    One could randomly connect to a different random full node at random times and request random bloom filters which consist of real tx the SPV is interested in and tx it will simply receive and discard.

[tvbcof]

Learn about bloom filters.

Are you saying that you can use them to make confusing pulls in an attempt to complicate analysis?  Good luck with that.

SPV nodes don't pull a single address or transaction.  It is more like "please send me all tx after block X which involve addresses starting with 1A".  The filter can be set as wide as one want, all the way up to everything (send me all tx from all address) as a full node would.  It is a tradeoff between privacy and bandwidth.  There is no need to even limit yourself to when you actually need tx data.  You can send a request for any subset of tx at any time to any node. 

One could randomly connect to a different random full node at random times and request random bloom filters which consist of real tx the SPV is interested in and tx it will simply receive and discard.

Yup.  That's what I thought you were implying.

Note that when all server operators are induced to provide their meta-data to the NSA, your scheme of connection to random servers will be, if anything, a marker to induce more strident analysis.  The data will be completely centralized.

The addresses of interest to you will almost certainly be able to be extracted via statistical analysis as long as you have a bias.  The best you can do is to apply the same bias to a pool of unrelated addresses.  Then you shoot yourself in the foot because you get tagged if any of those addresses happen to be under scrutiny.

Lastly, if somehow you can devise an effective framework to, as a user, subvert a near total 'free' use of Bitcoin and popularize the technique, the plug is pulled on the solution in total which, being 'supernodes' relegated to operation in datacenters, it fairly straightforward to do (ask Kim Dotcom about this works.)

[DeathAndTaxes]

Learn about bloom filters.

Are you saying that you can use them to make confusing pulls in an attempt to complicate analysis?  Good luck with that.

SPV nodes don't pull a single address or transaction.  It is more like "please send me all tx after block X which involve addresses starting with 1A".  The filter can be set as wide as one want, all the way up to everything (send me all tx from all address) as a full node would.  It is a tradeoff between privacy and bandwidth.  There is no need to even limit yourself to when you actually need tx data.  You can send a request for any subset of tx at any time to any node.  

One could randomly connect to a different random full node at random times and request random bloom filters which consist of real tx the SPV is interested in and tx it will simply receive and discard.

Yup.  That's what I thought you were implying.

Note that when all server operators are induced to provide their data to the NSA, your scheme of connection to random servers will be, if anything, a marker to induce more strident analysis.  The data will be completely centralized.

The addresses of interest to you will almost certainly be able to be extracted via statistical analysis as long as you have a bias.  The best you can do is to apply the same bias to a pool of unrelated addresses.  Then you shoot yourself in the foot because you get tagged if any of those addresses happen to be under scrutiny.

Lastly, if somehow you can devise an effective framework to, as a user, subvert a near total 'free' use of Bitcoin and popularize the technique, the plug is pulled on the solution in total which, being 'supernodes' relegated to operation in datacenters, it fairly straightforward to do (ask Kim Dotcom about this works.)

Yeah nonsense.  Some node running in Russia for example is going to be forced to give their data to the NSA?  I showed above how tx volume up to 100 tps is possible on existing hardware (run of the mill VPS).  Moore's law is still occuring, 1,000 tps+ in a decade is certainly possible on an even modest server with decent connectivity.  

Now is it likely we will reach a point where most users can't run a full node on a non-dedicated computer on a residential connection?  Probably given enough growth (at least 2-3 magnitudes).  However it is a logical fallacy to jump from that to "there will only be a handful of nodes controlled by the NSA".  That is just a figment of your imagination.  

We aren't talking about facebook scale dedicated datacenters.  A colocated server with 100 Mbps connectivity is more than sufficient.  Nothing crazy or exotic just decent amount of computing power, storage, and memory.  The idea that there can't be tens of thousands of nodes in hundreds of countries under a scenario like that is highly implausible.  If your paranoid then build a SPV client which only connects to different random nodes in US unfriendly countries and operates over TOR.

[tvbcof]

Yeah nonsense.  Some node running in Russia for example is going to be forced to give their data to the NSA?  I showed above how tx volume up to 100 tps is possible on existing hardware (run of the mill VPS).  Moore's law is still occuring, 1,000 tps+ in a decade is certainly possible on an even modest server with decent connectivity.  

Now is it likely we will reach a point where most users can't run a full node on a non-dedicated computer on a residential connection?  Probably given enough growth (at least 2-3 magnitudes).  However it is a logical fallacy to jump from that to "there will only be a handful of nodes controlled by the NSA".  That is just a figment of your imagination.  

We aren't talking about facebook scale dedicated datacenters.  A colocated server with 100 Mbps connectivity is more than sufficient.  Nothing crazy or exotic just decent amount of computing power, storage, and memory.  The idea that there can't be tens of thousands of nodes in hundreds of countries under a scenario like that is highly implausible.  If your paranoid then build a SPV client which only connects to different random nodes in US unfriendly countries and operates over TOR.

You are going to put your trust in the likes of Putin and the Chinese Central Party bearing in mind that an alternate monetary system threatens the control of their economies as much as it does the West (or would if it goes anywhere?)  OK, you do that.  For my part, I suspect that monitoring the Bitcoin network will be one of the areas where there is genuine cooperation as the benefits to the respective governments are huge.

Megauploads was just some racks of servers here and there with names like Carpathia on the cages.  Carpathia (and I suppose others) actually owned the gear, the likes of Equinix owned the floor-space, and the likes of Sprint owned the fiber and switches.  It took all of 5 minutes to halt the system (which was, IIRC, about 2% or 4% of the public Internet at one time.)  Pressure on anyone of the three providers would be sufficient to put Megauploads out of business for a period of time, and the legal justification was pretty weak compared to the (very real) national security implications of running a competitive alternate monetary solution.

As for TOR, all the government would have to do to severely limit it would be to stop funding it for Christsake.  Given the framework of backbone taps I'm even more convinced that it is nothing much more than a honeypot even when I try to convince myself otherwise.

[DeathAndTaxes]

Yeah nonsense.  Some node running in Russia for example is going to be forced to give their data to the NSA?  I showed above how tx volume up to 100 tps is possible on existing hardware (run of the mill VPS).  Moore's law is still occuring, 1,000 tps+ in a decade is certainly possible on an even modest server with decent connectivity. 

Now is it likely we will reach a point where most users can't run a full node on a non-dedicated computer on a residential connection?  Probably given enough growth (at least 2-3 magnitudes).  However it is a logical fallacy to jump from that to "there will only be a handful of nodes controlled by the NSA".  That is just a figment of your imagination. 

We aren't talking about facebook scale dedicated datacenters.  A colocated server with 100 Mbps connectivity is more than sufficient.  Nothing crazy or exotic just decent amount of computing power, storage, and memory.  The idea that there can't be tens of thousands of nodes in hundreds of countries under a scenario like that is highly implausible.  If your paranoid then build a SPV client which only connects to different random nodes in US unfriendly countries and operates over TOR.

You are going to put your trust in the likes of Putin and the Chinese Central Party bearing in mind that an alternate monetary system threatens the control of their economies as much as it does the West (or would if it goes anywhere?)  OK, you do that.  For my part, I suspect that monitoring the Bitcoin network will be one of the areas where there is genuine cooperation as the benefits to the respective governments are huge.

Megauploads was just some racks of servers here and there with names like Carpathia on the cages.  Carpathia (and I suppose others) actually owned the gear, the likes of Equinix owned the floor-space, and the likes of Sprint owned the fiber and switches.  It took all of 5 minutes to halt the system (which was, IIRC, about 2% or 4% of the public Internet at one time.)  Pressure on anyone of the three providers would be sufficient to put Megauploads out of business for a period of time, and the legal justification was pretty weak compared to the (very real) national security implications of running a competitive alternate monetary solution.

As for TOR, all the government would have to do to severely limit it would be to stop funding it for Christsake.  Given the framework of backbone taps I'm even more convinced that it is nothing much more than a honeypot even when I try to convince myself otherwise.

Your right it is doomed.  Uninstall the client now before you are assassinated.

[tvbcof]

Your right it is doomed.  Uninstall the client now before you are assassinated.

I think there is a pretty good chance that it is good for at least one more native pop where I can capitalize big time.  If not, so be it.

Without the benefit of real P2P, and with all the hassles of keeping keys secure, latency, etc, I cannot see Bitcoin being a competitive exchange currency in the long run.  Most people don't care about this shit and it is human nature to put faith in 'more powerful' entities than one senses in themselves.  So Bitcoin will be competing with more fully centralized entities which can easily eliminate the hassles associated with Bitcoin's early P2P efforts.  And with the same cryptographic strengths (and probably more) that Bitcoin has.

Even in the pool of early Bitcoin adopters and developers who have a generally higher grasp of technology and a generally less confidence in central governments, it is still rare to find people who care greatly about real P2P and pro-active threat mitigation.  At this point I've no real faith in the ecosystem to produce a solution I can have real confidence in.  There is a potential for Bitcoin to be a subsidized carrier for development of a robust solution which meets the threats of our day (and tomorrow), but I'm more interested in cutting up the logs in my field into lumber and beams than working on that at the moment and am headed out the door right now to work on that problem.  So, I'll talk at ya later.

[Come-from-Beyond]

If you are a casual user unable to keep the client online why not just use a SPV client.

I thought there wasn't any. Which one would u recommend?

https://multibit.org/

It is linked to and recommended from bitcoin.org

Thx, I'll learn about it to make sure I'm not supposed to trust it to be able to use it.

Multibit can't be used without trusting to a 3rd party. The blockchain size crossed 9 GB mark and keeps growing...

[TippingPoint]

Any entity can be put under enough pressure to either comply with government mandates (US government mandates in much of the world) or shut down and lose their infrastructure investment.  All of the PRISM participants chose the latter...it's the only logical choice.  And the only legal choice for corporations.

This is worth remembering ^ ^

[tvbcof]

Any entity can be put under enough pressure to either comply with government mandates (US government mandates in much of the world) or shut down and lose their infrastructure investment.  All of the PRISM participants chose the latter...it's the only logical choice.  And the only legal choice for corporations.

This is worth remembering ^ ^

 - snip, supposed NSA slide about PRISM - 

I bolded the part which I feel is especially important.  Most for-profit corporations under US law are legally compelled to maximize shareholder profit.  This is wholly incompatible with bucking the state's desire for the organization to participate in surveillance and incurring the expense of non-compliance.

In the corporations I've worked in, I've known a minority of people who are deeply disgusted by the police state apparatus being constructed,  and a majority to are ambivalent and/or willfully ignorant about it.  I've never met anyone who was in favor of it (though I mostly worked in engineering.)  In the end, it does not matter.  Upper management who draw plan the direction of the corporation's trajectory comply with the directive of maximizing shareholder profit.  If they fail, they are pushed out of management positions.

To me this is classic 'merger of state and corporate power' which Mussolini used when he preferred the term 'corporatism' to 'facsism' in seeking a characterization of his system of government in Italy.  It is also why even back in the 1700's there was considerable concern about the utility and dangers of 'corporations'.  It is also worth noting, however, that the definitions of 'corporation' has changed as society and business has evolved, but the basic structure of cooperation between parties in a corporation remains.  And more importantly, the concerns about the priorities of the corporate and how much harm/good they do for a society at large associated with these priorities, remains an interesting question.

[TippingPoint]

Any entity can be put under enough pressure to either comply with government mandates (US government mandates in much of the world) or shut down and lose their infrastructure investment.  All of the PRISM participants chose the latter...it's the only logical choice.  And the only legal choice for corporations.

This is worth remembering ^ ^

 - snip, supposed NSA slide about PRISM - 

Do you doubt its authenticity?