Author

Topic: Gold collapsing. Bitcoin UP. - page 375. (Read 2032266 times)

legendary
Activity: 1400
Merit: 1013
May 09, 2015, 12:42:40 AM
You are violating the fundamental tenet of Satoshi's white paper which is decentralized trust, meaning we don't have to trust that people are honest.
Have you even read the Bitcoin whitepaper?

Quote
The incentive may help encourage nodes to stay honest. If a greedy attacker is able to assemble more CPU power than all the honest nodes, he would have to choose between using it to defraud people by stealing back his payments, or using it to generate new coins. He ought to find it more profitable to play by the rules, such rules that favour him with more new coins than everyone else combined, than to undermine the system and the validity of his own wealth.
legendary
Activity: 4760
Merit: 1283
May 09, 2015, 12:26:01 AM

marcus, for Bitcoin to act as digital gold it is not enough to simply have scarcity.  it's too early.  it needs to become known, disseminated to all corners of the Earth, be used on a reliable basis and w/o problems, and over many years.
...

Put another way: “It became necessary to destroy the town to save it,"

legendary
Activity: 1372
Merit: 1000
May 09, 2015, 12:24:09 AM
There is a fundamental design error. They are just farting around the edges without going to the heart-of-the-matter, which is that transactions need to be orthogonal to blocks.

Bitcoin is headed towards centralization at a few servers. It was designed that way!
Is that you MofG anyway I'm Breaking the rule to not post under the influence.

So on that point and in the words of the creator

Quote from: Satoshi
The nature of Bitcoin is such that once version 0.1 was released, the core design was set in stone for the rest of its lifetime

Quote from: Satoshi
Yes, (we will not find a solution to political problems in cryptography), but we can win a major battle in the arms race and gain a new territory of freedom for several years
legendary
Activity: 2044
Merit: 1005
May 08, 2015, 11:58:22 PM
from gmax:

"So far the mining ecosystem has become incredibly centralized over time."

i totally disagree.  the trend has been back towards decentralization since ghash and ghash has been punished accordingly by the market down to a measly 3.6%.  what i see in this graph is a normal, expected distribution gradient from large to small share.  btw, gmax has been complaining about mining centralization since at least 2012 but yet here we are.  why do we accept his arguments on this when there hasn't been one major incident of a 51% attack?:


What people seem to keep missing in this "mining is centralized" claim, is that pools are not miners. They are services with ZERO barriers to entry and exit from the mining community.

As long as there is ONE honest P2P node who would publicly flag that a pool was behaving badly (an assumption I believe will always be true), then the pools can not abuse their position.

If any corrupt pool, or collection of corrupt pools, tried to falsify the record, it would immediately become public and that pool would lose most of it's miners in a matter of hours. Miners could simply switch to another honest pool, and there will always be an honest pool to switch to.

This would completely destroy the future profit stream for an established pool. Why the heck would anyone try this, especially given that it would be a futile effort?
+100 Mining is decentralized QED.
sr. member
Activity: 420
Merit: 262
May 08, 2015, 11:22:32 PM
2.  i think that the majority of ppl in this world want to be honest and wish to live in a society that has order.  no one wants to live in chaos.  everybody loses.  in order for society to continue to progress and evolve, order, dependability, and a semblance of honesty is needed.  thus, in a system with so much potential to do good, like Bitcoin, the overwhelming desire is for participants to want to do what makes the system thrive.  to the extent that cheating, dishonesty, and colluding erodes confidence and threatens that goal, most participants will avoid those activities.

That is the same faith we put into a top-down democracy. Fact is a power vacuum sucks in those who can maximize the exploitation of the power vacuum.

You are violating the fundamental tenet of Satoshi's white paper which is decentralized trust, meaning we don't have to trust that people are honest.

You are blacksliding because there doesn't appear to be any solution the fact that pools become concentrated due to the variance cost to them not. It is pure economics.

Now what are the proposed solutions?

If there are none, we are just lying to ourselves and the decentralization is a mirage. And smooth's valuation statement follows accordingly.

As for the scaling of Bitcoin, they will never get there because there are these fundamental problems, such as the design decision to charge a transaction fee and the need to propagate all the transaction data before a new block can be started.

There are fundamental design errors here. Very very fundamental.

Who is going to fix this? And pronto!

Looks like this is not so simple after all....

http://gavinandresen.ninja/utxo-uhoh

The can-kick might only be to 2 or 3MB while efficiencies in UTXO storage are worked on.

There is a fundamental design error. They are just farting around the edges without going to the heart-of-the-matter, which is that transactions need to be orthogonal to blocks.

Bitcoin is headed towards centralization at a few servers. It was designed that way!
legendary
Activity: 1764
Merit: 1002
May 08, 2015, 10:57:27 PM

Cypher, I rarely delete my posts, but I just got rid of the one quoted above, because I just read down further on gmaxwell's post:
Quote
This is frustrating; from a clean slate analysis of network
health I think my conclusion would be to _decrease_ the limit below the
current 300k/txn/day level.

This is obviously not acceptable, so instead many people--myself
included--have been working feverishly hard behind the scenes on Bitcoin
Core to increase the scalability.
ok, me too.
legendary
Activity: 1078
Merit: 1006
100 satoshis -> ISO code
May 08, 2015, 10:40:25 PM
Cypher, I rarely delete my posts, but I just got rid of the one quoted above, because I just read down further on gmaxwell's post:
Quote
This is frustrating; from a clean slate analysis of network
health I think my conclusion would be to _decrease_ the limit below the
current 300k/txn/day level.

This is obviously not acceptable, so instead many people--myself
included--have been working feverishly hard behind the scenes on Bitcoin
Core to increase the scalability.

I want to wait until I see his final word on the matter because, like Peter Todd saying the limit can be increased "eventually", he may still decide to "ack" Gavin's change.
Yep. The SC situation is a whole other complex matter which needs to be observed closely.
legendary
Activity: 1764
Merit: 1002
May 08, 2015, 10:31:18 PM
My bold emphasis.  If even 100MB might be "less devastating" to decentralization in a few years, then the capacity of the network must be improving by a fraction of that each year. Why not limit to 3,5,10 or even 20MB right now, when blocks of these larger sizes are still years off, but improving capacity rises to meet the limit? Something has seriously changed in his thinking, so that the idea that of hitting the 1MB is now fine. This is new and dangerous ground.

personally, i think this is what $21M in investment money on your back will do to one's thinking.  there has to be enormous pressure on him and i'm sure those investors are constantly talking to him demanding progress reports.  many of them probably read this forum and maybe this thread to monitor opposition to Blockstream.  they then go report back to gmax.  i definitely have a hard time following his writing and logic sometimes when it comes to things non-cryptographic.  he may be confused.  maybe this, in reality, is where the pressure is coming from that marcus refers to.

i think even if you disagree with me on my supposition that SC's are a competitor to Bitcoin, i think it's pretty easy to see that gmax sees Bitcoin as a competitor to SC's.
sr. member
Activity: 420
Merit: 262
May 08, 2015, 09:37:44 PM
Quote
Maybe the reason is that Bitcoin only grows to the point where its network state provides adequate confidence and security. So you are looking at causality in the wrong direction. Maybe it would be much, much bigger if it were actually decentralized and it were in fact not possible for "the big three" to attack it and users/investors didn't have to rely on "why aren't they doing it".


is "not possible" achievable?  as JR asked, what distribution would make you happy?  of course, what makes you happy maybe won't satisfy marcus.

Maybe in a metaphysical sense it isn't but if it requires a non-trivial number of actors (e.g >3), who don't know each other, to collude, that becomes practically impossible.

What has the largest number of pools needed to achieve a majority ever been since pool mining got entrenched? 4 or 5?

This of course ignores selfish-mining type attacks where a strict majority might not even be needed.




here's what Ed Felten of Princeton thinks of selfish mining:

https://freedom-to-tinker.com/blog/felten/bitcoin-isnt-so-broken-after-all/

Yes that's why I wrote a majority might not be needed. I'm aware it is controversial.You can ignore that point though and it is still the case that security is much weaker with a small number of large pools. Are people going to be doing billion dollar-value transactions when a few sketchy pool operators (or their employees, or a hacker exploiting their pool) could steal it? No. Value can only grow as far as confidence permits and concentration undermines (or at least limits) confidence.

And so what is your proposed solution to the inherent concentration of pools due to variance cost of smaller pools?

And that doesn't apply only to billion txs but also to for example trusting to hold significant crypto-currency over gold.

Remember I had pointed out that in theory P2Pool is vulnerable to share withholding attack even with a fix (as all pools are in Bitcoin because the fix has not been made to Bitcoin, but a centralized pool could more viably using profiling to defeat share withholding) because those submitting shares can see the solution and withhold if they have the winning block hash. The only economic reason to do this is to discourage P2Pools, because the withheld share is lost profit for the holder of the withheld share. However this is asymmetric cost, because the holder has profited most of the time on all the others who don't withhold. Thus P2Pool can be defeated by those with vested interested in large pools.

-AnonyMint
legendary
Activity: 1153
Merit: 1000
May 08, 2015, 09:24:39 PM
The discussion is not necessarily about performing an attack, it is about incentives and concentration of voting power. If the big pools want lower fees and/or a higher soft limit, because they are at a competitive advantage in providing the infrastructure required for it, then they get it, regardless of what the rest want. Likewise it has always been the prerogative of pools which transactions to include, so if they want to block some (as eligius does, I understand) then they will do that too. Neither of these rise to the level of an attack that causes the pool to be viewed as hostile and shunned by the community, but it gives disproportionate influence to a few actors.

Increasing blocksize increases the # of transactions allowed, which in turn increases the usage of bitcoin and the # of fee paying transactions, and this is how miners and pools will increase their fees. Increased fees lead to more monetary incentive to mine, which increases the hash rate and also the # of independent miners. This increases the bitcoin security mechanism.

gmaxwell on the other hand just suggested that we should lower the blocksize below 1MB in order to increase fee pressure.
http://sourceforge.net/p/bitcoin/mailman/message/34090559/
Quote
This is frustrating; from a clean slate analysis of network health I think my conclusion would be to _decrease_ the limit below the current 300k/txn/day level.

This is simply absurd. His suggestion of limiting bitcoin to less than 1tps, as a means to increase the value of the bitcoin ecosystem is simply wrong.

To put it simply, a bitcoin that most people can not use directly and where blocks only contain 1000 transaction each, will generate a much smaller amount of economic value in fees than a bitcoin that is widely adopted by people and where blocks contain billions of transactions each.
legendary
Activity: 1764
Merit: 1002
May 08, 2015, 09:20:44 PM
Quote
Maybe the reason is that Bitcoin only grows to the point where its network state provides adequate confidence and security. So you are looking at causality in the wrong direction. Maybe it would be much, much bigger if it were actually decentralized and it were in fact not possible for "the big three" to attack it and users/investors didn't have to rely on "why aren't they doing it".


is "not possible" achievable?  as JR asked, what distribution would make you happy?  of course, what makes you happy maybe won't satisfy marcus.

Maybe in a metaphysical sense it isn't but if it requires a non-trivial number of actors (e.g >3), who don't know each other, to collude, that becomes practically impossible.

What has the largest number of pools needed to achieve a majority ever been since pool mining got entrenched? 4 or 5?

This of course ignores selfish-mining type attacks where a strict majority might not even be needed.




here's what Ed Felten of Princeton thinks of selfish mining:

https://freedom-to-tinker.com/blog/felten/bitcoin-isnt-so-broken-after-all/

Yes that's why I wrote a majority might not be needed. I'm aware it is controversial.You can ignore that point though and it is still the case that security is much weaker with a small number of large pools. Are people going to be doing billion dollar-value transactions when a few sketchy pool operators (or their employees, or a hacker exploiting their pool) could steal it? No. Value can only grow as far as confidence permits and concentration undermines (or at least limits) confidence.

2 things:

1.  i think ppl under-appreciate the extent to which all participants in Bitcoin, including miners, volunteer and want to be part of a system that has the potential to make themselves extraordinary profits if it works as intended that being in an open, honest manner.  there's a lot at stake in constructing a new financial system and those profits can only be made if it works properly as advertised in that open and honest manner that ordinary ppl can depend on.  this is what will result in the trust needed so that the vast majority of humanity can buy into such a reliable system.

2.  i think that the majority of ppl in this world want to be honest and wish to live in a society that has order.  no one wants to live in chaos.  everybody loses.  in order for society to continue to progress and evolve, order, dependability, and a semblance of honesty is needed.  thus, in a system with so much potential to do good, like Bitcoin, the overwhelming desire is for participants to want to do what makes the system thrive.  to the extent that cheating, dishonesty, and colluding erodes confidence and threatens that goal, most participants will avoid those activities.
legendary
Activity: 1764
Merit: 1002
May 08, 2015, 09:06:03 PM
Maybe in a metaphysical sense it isn't but if it requires a non-trivial number of actors (e.g >3), who don't know each other, to collude, that becomes practically impossible.

so to apply what Felten says in his paper, would you participate honestly in a coalition to cheat honest miners when you don't know your partners all of whom may be as dishonest as you?

Quote

What has the largest number of pools needed to achieve a majority ever been since pool mining got entrenched? 4 or 5?

i don't remember honestly.
legendary
Activity: 2968
Merit: 1198
May 08, 2015, 09:02:52 PM
Quote
Maybe the reason is that Bitcoin only grows to the point where its network state provides adequate confidence and security. So you are looking at causality in the wrong direction. Maybe it would be much, much bigger if it were actually decentralized and it were in fact not possible for "the big three" to attack it and users/investors didn't have to rely on "why aren't they doing it".


is "not possible" achievable?  as JR asked, what distribution would make you happy?  of course, what makes you happy maybe won't satisfy marcus.

Maybe in a metaphysical sense it isn't but if it requires a non-trivial number of actors (e.g >3), who don't know each other, to collude, that becomes practically impossible.

What has the largest number of pools needed to achieve a majority ever been since pool mining got entrenched? 4 or 5?

This of course ignores selfish-mining type attacks where a strict majority might not even be needed.




here's what Ed Felten of Princeton thinks of selfish mining:

https://freedom-to-tinker.com/blog/felten/bitcoin-isnt-so-broken-after-all/

Yes that's why I wrote a majority might not be needed. I'm aware it is controversial.You can ignore that point though and it is still the case that security is much weaker with a small number of large pools. Are people going to be doing billion dollar-value transactions when a few sketchy pool operators (or their employees, or a hacker exploiting their pool) could steal it? No. Value can only grow as far as confidence permits and concentration undermines (or at least limits) confidence.
legendary
Activity: 3920
Merit: 2349
Eadem mutata resurgo
May 08, 2015, 08:59:27 PM
Quote
maybe the anti-Gavin crowd is just whining about something they don't fully understand.

yeah, sipa and jgarzik are just a bunch of whiners who don't fully understand bitcoin ... as well as you, I'm supposing you were implying? recall we were spoon-feeding you UTXO's knowledge for free just last week? That is a very good reason to oppose increasing the maximum block size. - Gavin Andresen

You don't have to be such a full time abrasive jerk to make your point, the pointless, incessant snark, insinuations and character assassination is nauseating not to mention it adds nothing to understanding.

It was the same with the sidechains 'debate' on here ... you didn't want to listen to counterpoints but just blasted the volume of FUD for your chosen position on some kind of 1 man PR propaganda campaign. I don't know about net benefits surrounding sidechains, they seem interesting and there is no implementation in the wild, but I do know you did absolutely nothing to advance understanding, yours and anybody reading your emotive blather ... I just scrolled on most of your SC stuff, 1 level above ignore tbh. God gave you 2 ears and 1 mouth ...
legendary
Activity: 1764
Merit: 1002
May 08, 2015, 08:55:58 PM
Quote
Maybe the reason is that Bitcoin only grows to the point where its network state provides adequate confidence and security. So you are looking at causality in the wrong direction. Maybe it would be much, much bigger if it were actually decentralized and it were in fact not possible for "the big three" to attack it and users/investors didn't have to rely on "why aren't they doing it".


is "not possible" achievable?  as JR asked, what distribution would make you happy?  of course, what makes you happy maybe won't satisfy marcus.

Maybe in a metaphysical sense it isn't but if it requires a non-trivial number of actors (e.g >3), who don't know each other, to collude, that becomes practically impossible.

What has the largest number of pools needed to achieve a majority ever been since pool mining got entrenched? 4 or 5?

This of course ignores selfish-mining type attacks where a strict majority might not even be needed.




here's what Ed Felten of Princeton thinks of selfish mining:

https://freedom-to-tinker.com/blog/felten/bitcoin-isnt-so-broken-after-all/
legendary
Activity: 2968
Merit: 1198
May 08, 2015, 08:30:56 PM
Quote
Maybe the reason is that Bitcoin only grows to the point where its network state provides adequate confidence and security. So you are looking at causality in the wrong direction. Maybe it would be much, much bigger if it were actually decentralized and it were in fact not possible for "the big three" to attack it and users/investors didn't have to rely on "why aren't they doing it".


is "not possible" achievable?  as JR asked, what distribution would make you happy?  of course, what makes you happy maybe won't satisfy marcus.

Maybe in a metaphysical sense it isn't but if it requires a non-trivial number of actors (e.g >3), who don't know each other, to collude, that becomes practically impossible.

What has the largest number of pools needed to achieve a majority ever been since pool mining got entrenched? 4 or 5?

This of course ignores selfish-mining type attacks where a strict majority might not even be needed.

legendary
Activity: 1764
Merit: 1002
May 08, 2015, 08:04:34 PM
from gmax:

"So far the mining ecosystem has become incredibly centralized over time."

i totally disagree.

I guess it is in the eye of the beholder. I still see 3 pools with >50%. The other tiny little slices don't matter at all as long as the top 3 agree and if even two of them agree, the tiny slices still hardly matter at all.



all it takes is for the owners of the 3 pools to be corrupt and btc is screwed. that isnt decentralization... thats asking for trouble.

then tell me, why hasn't it happened already?  there are supposedly billions to be made doing this.

Maybe the reason is that Bitcoin only grows to the point where its network state provides adequate confidence and security. So you are looking at causality in the wrong direction. Maybe it would be much, much bigger if it were actually decentralized and it were in fact not possible for "the big three" to attack it and users/investors didn't have to rely on "why aren't they doing it".


is "not possible" achievable?  as JR asked, what distribution would make you happy?  of course, what makes you happy maybe won't satisfy marcus.

fact of the matter is, Bitcoin has done something in a mere 6 yrs that is unprecedented.  no one thought we'd be where we are today.  look at all the banks, gvts, and CB's scrambling to compensate.

sure what you say is possible but otoh, maybe the anti-Gavin crowd is just whining about something they don't fully understand.
legendary
Activity: 1764
Merit: 1002
May 08, 2015, 07:57:04 PM
from gmax:

"So far the mining ecosystem has become incredibly centralized over time."

i totally disagree.  the trend has been back towards decentralization since ghash and ghash has been punished accordingly by the market down to a measly 3.6%.  what i see in this graph is a normal, expected distribution gradient from large to small share.  btw, gmax has been complaining about mining centralization since at least 2012 but yet here we are.  why do we accept his arguments on this when there hasn't been one major incident of a 51% attack?:


What people seem to keep missing in this "mining is centralized" claim, is that pools are not miners. They are services with ZERO barriers to entry and exit from the mining community.

As long as there is ONE honest P2P node who would publicly flag that a pool was behaving badly (an assumption I believe will always be true), then the pools can not abuse their position.

If any corrupt pool, or collection of corrupt pools, tried to falsify the record, it would immediately become public and that pool would lose most of it's miners in a matter of hours. Miners could simply switch to another honest pool, and there will always be an honest pool to switch to.

This would completely destroy the future profit stream for an established pool. Why the heck would anyone try this, especially given that it would be a futile effort?

yep, i was a long time miner and it is incredibly easy to switch back and forth btwn pools.  i would sometimes do it multiple times a day.  it's simply a matter of pointing your machine to a specific ip address.  software, like cgminer, allows auto switching btwn pools if one goes down or payout terms changed.  pool operators have 0 control over you.  in fact, they do everything they can to bend over and court your hashing power.  they all run their own IRC channels where you can go immediately if somethings up and there always manned by staff.  the scrutiny i'm sure can be overwhelming for them. 
legendary
Activity: 1764
Merit: 1002
May 08, 2015, 07:51:51 PM
from gmax:

"So far the mining ecosystem has become incredibly centralized over time."

i totally disagree.

I guess it is in the eye of the beholder. I still see 3 pools (assuming "unknown is a mostly some big pool, if not that changes things") with >50%. The other tiny little slices don't matter at all as long as the top 3 agree and if even two of them agree, the tiny slices still hardly matter at all.



well, i don't see them EVER agreeing to perform an attack.  they would destroy confidence in the network and all their equipment value goes poof.  cashing out in USD after an attack is problematic as well as the exchanges are watching for any funny business and Bitstamp is an example of how account funds can be frozen (Jed McCaleb).

this graph is probably never going to change much even with huge Bitcoin success.  there will always be bigger pools than others.

furthermore, pools consist mainly of individual participants who will defect if their pool operator is abusing their hashing power.  ghash has been decimated by defection.

The discussion is not necessarily about performing an attack, it is about incentives and concentration of voting power. If the big pools want lower fees and/or a higher soft limit, because they are at a competitive advantage in providing the infrastructure required for it, then they get it, regardless of what the rest want. Likewise it has always been the prerogative of pools which transactions to include, so if they want to block some (as eligius does, I understand) then they will do that too. Neither of these rise to the level of an attack that causes the pool to be viewed as hostile and shunned by the community, but it gives disproportionate influence to a few actors.



have we seen evidence of them doing this yet? i'm not aware of them throwing their weight around like that.  and that's even when alot of the mining is in China.  perhaps their is more decentalization going on than we think with mining spread out round the world that it is not possible for them to collude. 

as far as which tx's to include in a block, that is being worked on with things like getblocktemplate.
legendary
Activity: 1764
Merit: 1002
May 08, 2015, 07:44:44 PM
These guys have worked themselves to the point of mental exhaustion in some cases to keep this thing scaling up to handle the transaction growth rate thus far and have made huge advances. They would all simply agree (ya know, a consensus) to raise the limit if it was such an obvious 'fix', it is not and there are no obvious fixes, that is just wishful thinking by the unthinking majority. Raising the limit is the last ditch 'suck it and see' hail mary pass when everything else has been optimised as much as possible ...

If the limit gets raised substantially above the technological improvements growth rate I'll be pulling the vast majority of my investment out because it is not going to be operating like we thought it would, i.e. it won't be a clearing and settlement digital gold network but a paypal2.0, fun internet googlesque reversable, traceable payments network

You are doing the same thing you are accusing others of doing. There is no consensus or even an easy answer, whether Bitcoin should be a "clearing and settlement digital gold network" or a "payments network" (I don't think too many want it to be reversible and traceable but you are injecting that as a form of commentary). If there were an easy answer, it would already be accepted as consensus, but there isn't and it isn't.


Implicitly,

decentralised => clearing and settlement digital gold network

centralised => reversable, traceable payments network

(without thinking that is not well-known by now).

b/c an African kid can't currently touch, admire, coddle, weigh, polish, covet, put it in his pocket, or sell a Bitcoin, he will never value it as gold. with a 1MB cap, he won't even ever get a chance to know about Bitcoin, let alone what it can do.  and even if he does get the chance, since he can't do all those physical things i just described with Bitcoin, what will confirmation delays, insufficient fees, double spending losses, waiting times at the store all do for his confidence in Bitcoin acting as a gold substitute as a result of the 1MB cap?
Jump to: