Bitcoin Forum
March 19, 2024, 06:48:33 AM *
News: Latest Bitcoin Core release: 26.0 [Torrent]
 
  Home Help Search Login Register More  
  Show Posts
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 [21] 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 »
401  Bitcoin / Pools / Re: [460 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool on: June 17, 2014, 12:19:19 AM

What I'm unable to figure out, and nobody gets what I'm asking, is if blocks are discovered by the pool without me finding a share during that "round", since as far as p2pool is concerned I didn't do anything. Do I just lose out in the long run, or does it somehow magically average out over the course of a month I'll get comparable payouts from a centralized pool in the long run?


Yes, it evens out in the long-run. While you may not get paid for every block found, there will be times when you get over-paid. I myself (with 10GHash/s) just went about a month with no pay-outs. This past week, I was paid 7.36311mBTC (including a donation due to the Ghash.io situation).

Also frustrating is when you find a share, but the pool does not find a block within 3 days Tongue

402  Bitcoin / Bitcoin Discussion / Re: I am selling half of my Bitcoin holdings because of Ghash on: June 16, 2014, 11:59:46 PM

Bitcoin core developers and the community need to come together to fix the Bitcoin protocol so that is no longer vulnerable to hijack by large mining pools. Until that happens, I can't fully trust Bitcoin as a value store, and neither should you.


Unfortunately, you can't work around the 51% attack.

Bitcoin's innovation was to guard against malicious updates to the block-chain by requiring a "proof-of-work". This allows the validity of the block-chain to be verified in a trustless manner. The "proof-of-work" assumes two things:
  • That people are generally good. (Bitcoin also makes is so that cheating in-system is more expensive than being honest.)
  • That competition will ensure that no miner gets a substantial market-share. (If your business relies on bitcoin, it makes sense to invest in a full node or pay somebody to do so).

Pools having greater than 30% of the hash-power breaks the second assumption. 30% is the magic number because one of the pools may be run or coerced by a malicious person: we would not know until it is too late.

"Proof-of-work" is not something to be worked around. When you participate in the BItcoin network, you have to accept that it is a valid way to form consensus. There are no known alternatives that are still decentralized. Evidently, it is not known if "proof-of-work" can stay decentralized. However, if decentralization is not possible with "proof-of-work": that means that the experiment failed, and that Bitcoins are worth approximately $0.

This has happened before. What happened was hashers left Deepbit for other pools. This time it is different: many of the hashers seem to believe it is up to the developers to code a protocol change to solve the problem. They fail to see how they are contributing to the problem.

I see only only one solution: hashers need to start avoiding pools with more than 30% market share. Pools with more than 30% market share need to start raising fees. If these solutions don't start to materialize within about a week, I will join you in selling half my holdings.

Edit: There will probably arguments that I am asking hashers to go against their own self-interest. The purpose of hashing is to secure the network. If you refuse to monitor the health of the network, it is completely reasonable to expect more variance in price.

403  Bitcoin / Bitcoin Discussion / Re: It's already 48%, No one care about Ghash? on: June 16, 2014, 11:30:37 PM
So basicly, it's better to let shit happen, than to act on it before it happens? Shit that can kill or cripple BTC?
Wouldn't it be better to prepair everyone in advance? Make some dead line, update clients months ahead etc, and than execute hard fork?

I cant see how doing it that way could be wors than attack happening and than changing protocol in rush....what can go wrong with that i wonder...

Bitcoin is an experiment in using "proof-of-work" to secure a public transaction ledger. You can not work around the 51% attack at the protocol level without killing the currency.

Bitcoin's innovation was to guard against malicious updates to the block-chain by requiring a "proof-of-work". This allows the validity of the block-chain to be verified in a trustless manner. The "proof-of-work" assumes two things:
  • That people are generally good. (Bitcoin also makes is so that cheating in-system is more expensive than being honest.)
  • That competition will ensure that no miner gets a substantial market-share. (If your business relies on bitcoin, it makes sense to invest in a full node or pay somebody to do so).

Pools having greater than 30% of the hash-power breaks the second assumption. 30% is the magic number because one of the pools may be run or coerced by a malicious person: we would not know until it is too late.

"Proof-of-work" is not something to be worked around. When you participate in the BItcoin network, you have to accept that it is a valid way to form consensus. There are no known alternatives that are still decentralized. Evidently, it is not known if "proof-of-work" can stay decentralized. However, if decentralization is not possible with "proof-of-work": that means that the experiment failed, and that Bitcoins are worth approximately $0.
404  Bitcoin / Bitcoin Discussion / Re: Is Bitcoin not decentralized anymore? on: June 16, 2014, 07:20:55 PM

This is a hard fork that is required.

What happens next time we get 51% ?
Move your miners again?


Yes! that is exactly what hashers (distinct from miners) should do. My advice is simple: if a pool has more than 30% of the network hash-power, point your hashers at a different one.

If you get tired of switching pools, you can join P2Pool (or something like it): it is a decentralized pool.

If a hard-fork happens, I will have to re-evaluate my interest in Bitcoin. A pool (Deepbit) controlled more than half the hash-power before. No hard-forks were used to correct that: people simply moved away from the pool. I fail to understand why that is not a viable option every time a pool gets too large.
405  Bitcoin / Bitcoin Discussion / Re: Is Bitcoin not decentralized anymore? on: June 16, 2014, 06:36:29 PM

No mate, my argument is that I am not the one responsible for this.
The dev team could have prevented it with a hard fork yet still choose to ignore it.

Why should I lose on my investment when the ones controlling the entire project ain't doing nothing about it?

A hard-fork is a last resort that very well may kill the project.

If the dev team controls the entire project, then something is very wrong. Gavin Anderson has had a policy of being very conservative with protocol changes. This allows the development of alternate implementations. Currently the vast majority of the miners use the reference implementation, possibly with minor tweaks. That does not always have to be the case though.



406  Bitcoin / Bitcoin Discussion / Re: Fake Bitcoin on: June 16, 2014, 05:42:02 PM
One thing that convinced me that Bitcoin was here to stay was learning that Gold is easier to counterfeit.

One thing that concerns me is the Concentration of hash-power.

Having more than half the hash-power allows you to re-write history to the last check-point of about January 7th. They would not be able to make fake Bitcoins; but they could make it so that the Bitcoins you acquired in the last few months never existed. The farther you want to go back, the more time/hash-power you would need: all the "proof-of-work" in the blocks to be re-written would have to be re-done.

TL:DR Fake Bitcoins are not possible. In a degenerate network (with a bad actor controlling the majority of the hash-power), the real ones are not worth the paper they are printed on.
407  Bitcoin / Bitcoin Discussion / Re: Is Bitcoin not decentralized anymore? on: June 16, 2014, 07:05:25 AM
Philip and Yakamoto
Your problem is not with me, it's with the way the protocol works right now.

So pls don't take it on me.
It is NOT my fault that the ones responsible failed to make sure that the protocol is secure.

My intention is not to single you out. As you pointed out earlier, you are only a small contribution to Ghash.io's hash-power. I am not upset with you specifically, but miners hashers1 as a group that keep pointing their hash-power at pools with greater than 30% market-share.

Quote
That will not change the fact that it IS NOT GHash's fault NOR MINE, it's the ones who are responsible for making sure the network is secure.
Apparently it's not.

EDIT: That being said, BTC is centralized because we DEPEND on them.
Maybe it it my selective reading again, but you don't explicitly say who is responsible. The implication seems to be that the developers (specifically Satoshi) screwed up with a flawed protocol.

Bitcoin's innovation was to guard against malicious updates to the block-chain by requiring a "proof-of-work". This allows the validity of the block-chain to be verified in a trustless manner. The "proof-of-work" assumes two things:
  • That people are generally good. (Bitcoin also makes is so that cheating in-system is more expensive than being honest.)
  • That competition will ensure that no miner gets a substantial market-share. (If your business relies on bitcoin, it makes sense to invest in a full node or pay somebody to do so).

Pools having greater than 30% of the hash-power breaks the second assumption. 30% is the magic number because one of the pools may be run or coerced by a malicious person: we would not know until it is too late.

"Proof-of-work" is not something to be worked around. When you participate in the BItcoin network, you have to accept that it is a valid way to form consensus. There are no known alternatives that are still decentralized. Evidently, it is not known if "proof-of-work" can stay decentralized. However, if decentralization is not possible with "proof-of-work": that means that the experiment failed, and that Bitcoins are worth approximately $0.

I am explaining this, not to take you down, but because I believe that this situation can still be remedied with miner hasher education. With the recent DDOS, the hash-rate at Ghash.io dropped substantially; while the hash-rate at P2Pool went up substantially. That means that miners are using P2Pool as a fail-over. I would prefer it if more miners would use P2Pool (or something like it) voluntarily.



1. Hashers are not miners, and Bitcoin network doesn't need them.
That said, I was a hasher myself last month. I normally use P2Pool, but was distraught to learn that the share-chain only applies to Bitcoin. That is, if you are merge-mining some alt-coins, the miner that finds a block does not share the reward. I mined with Eligius to reduce my variance in finding namecoins.
PS: I currently mine at a loss (10Ghash/s). Because I read fine-print and am running a full node, my biggest expense is commercial Internet access. I understand that not everybody has the time/money to do that.
408  Bitcoin / Bitcoin Discussion / Re: Is Bitcoin not decentralized anymore? on: June 15, 2014, 09:48:31 PM
I have heard about selective hearing but never heard of selective reading before....
Read what I wrote.
I emphasized it for you.
Have a good one.

There was a "large miner" on reddit who claimed to be sticking with ghash.io as well. It is not just you.

I actually do have selective reading. Learned this when my mom told me about a want-ad in the classified section of the newspaper. I missed it because it was ¼ page, instead of postage-stamp size. Over the years I had started reading fine-print first, and ignoring 72pt fonts.
409  Bitcoin / Bitcoin Discussion / Re: Is Bitcoin not decentralized anymore? on: June 15, 2014, 09:34:19 PM

Historically, only 2 viable solutions exist:
1) An authority to regulate (% in this case)
2) Privatization

http://en.wikipedia.org/wiki/Tragedy_of_the_commons#Modern_solutions

Pick your poison..... but be quick about it.

You forgot:
3) The community is small enough for ostracism to work (less than about 150 people).
410  Bitcoin / Bitcoin Discussion / Re: Is Bitcoin not decentralized anymore? on: June 15, 2014, 09:30:33 PM
WTF man! You think they have over 51% and you still point miners at them... Where is your logic?!?

Think about it, if they have more than 51% what is 720Gh going to do if I remove it from them?
720Gh means jack shit for hashing power. Apart from that I would like to at least get my investment back (which I didn't yet).
I also have more than that on other pools - I don't like to keep all my eggs in one basket Wink


Raindrops become a torrent.

... Recently, my brother reminded me that I claimed a 51% attack could not happen because people would desperately add hash-power to the network (even at a loss) to stop it.

People seem to be saying: "my hash-power does not make a difference, so I might as well benefit from the lower variance." The unseen risk is that concentration on only one pool may dramatically increase the variance in price.


411  Bitcoin / Bitcoin Discussion / Re: Why Bitcoin Experts Aren’t Afraid of a 51% Attack on: June 15, 2014, 07:53:17 PM
That article uses quotes from 2009. Recently, my brother reminded me that I claimed a 51% attack could not happen because people would desperately add hash-power to the network (even at a loss) to stop it.

With large miners claiming to be on Ghash.io for reduced variance, this is clearly not happening. Though I just checked: Antminer S1s are currently out of stock.
412  Bitcoin / Pools / Re: [460 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool on: June 15, 2014, 07:37:30 PM
Does anyone know of a way to capture the password input a miner uses? I'm wondering if it's possible for miners to differentiate workers this way.

I think you should assume that any password sent to a P2Pool node will be known to the node operator.

And, no, passwords are not a good way to differentiate users:
It’s Me, and Here’s My Proof: Why Identity and Authentication Must Remain Distinct
Quote
Published: February 14, 2006
By Steve Riley
Senior Security Strategist
Security Technology Unit
Microsoft Corporation
413  Bitcoin / Bitcoin Discussion / Re: PSA: Add a Full Node for just $19/year! on: June 15, 2014, 06:56:43 PM
Does anyone know if this pricing is locked in year over year? If so I'm going to have to set up a ton of these!

Cheers,
Adam

There is some risk concentrating them all in the same data-center.

I have idle long-term plans to set up a Bitcoin node in Iceland, which I believe to be relatively Bitcoin friendly. Would be valuable if Bitcoin is ever banned at home.


Wouldn't the same risk be present if the servers that host the nodes are leased by the same provider?

The person I was responding to was talking about spinning up a bunch of identical nodes in the same data-center. Are you asking if 1 server is just as risky as a bunch? I think my point is you don't get the same diversity benefits with all of them in one place.

My "idle long-term plans" are essentially about keeping the network going if the G8 decide to crack down on it. My current ISP prohibits illegal use in their TOS. So if hosting a node becomes illegal, I would have to shut it down: or move it to a more friendly jurisdiction.
414  Bitcoin / Bitcoin Discussion / Re: Bitcoin Mining Pool Ghash.io DDos-ed in Response to threat of 51% attack? on: June 15, 2014, 06:32:30 PM
If Ghash.io had all of their hash-power in-house, a DDOS would not  be an effective way to reduce their hash-rate. They would only need a few modem connections (or 1 VDSL/cable connection) to keep participating in the network: even if their public links are saturated.

The reason their hash-rate went down appears to be that the miners using them as a pool were unable to connect. Unless their miners are going to start using tor or I2P, their links for such mining must always have a public IP address.

In conclusion, DDOS will not discourage well-funded adversaries from attacking the network. DDOS attacks are really only effective against entities with public-facing architecture.
415  Bitcoin / Bitcoin Discussion / Re: Could bitcoin save the music industry? on: June 13, 2014, 10:16:31 PM
Could you do this using a protocol in a similar way that bitcoin does?
No.
416  Economy / Economics / Re: List of Bitcoin Hostile (and friendly) Banks on: June 13, 2014, 10:07:40 PM
Lloyds Bank in the UK appears to be bitcoin hostile.

One of the commenters mentioned that Metrobank is also Bitcoin (and Adult industry) hostile.

5flags, who also had their Lloyds accounts closed, recommended HSBC.

Guido recommended Nationwide, but it is not clear that person disclosed Bitcoin-related activity. Guido advises against RBS and  Barclays over general scandal, rather than specific Bitcoin hostility
417  Bitcoin / Legal / Re: Lloyds Bank Account Forcibly Closed on: June 13, 2014, 09:57:12 PM
So, I really would like to know which banks are not funny about Bitcoin, preferably without spending a week finding out in each case.

Does anyone have advice as to which Uk bank to use?

There is the List of Bitcoin Hostile (and friendly) Banks

unfortunately, it looks neglected and has no UK entries yet.
418  Bitcoin / Bitcoin Discussion / Re: Could bitcoin save the music industry? on: June 13, 2014, 09:28:16 PM
Thanks dwma, I will look up the Bitshares Music project. So does anybody know of an alt-coin project that attempts to facilitate the transfer of ownership of encrypted or hidden data?

Encryption is not copy protection.

TL;DR: Your are trusting the viewer to not retain the decryption key after viewing the work.
419  Bitcoin / Bitcoin Discussion / Re: Could bitcoin save the music industry? on: June 13, 2014, 09:02:37 PM
Because this would have positive impact on the music industry as well as any other industry that are having issues with piracy.  

This is becoming a FAQ.

  • The Bitcoin blockchain uses "proof-of-work" to prevent counterfeiting.
  • The entertainment industry has been pushing to conflate unauthorized copying with counterfeiting physical goods.
  • People then confuse the Block-chain with "copy protection": when in fact, it has the opposite purpose.

People are encouraged to keep a copy of the block-chain. Most people don't bother because it is inconvenient or expensive. The only closely guarded secret in bitcoin is the private keys for spending coins. They are guarded secrets because they have no "copy protection".

Music with effective copy protection is worthless anyway: it is a means to share information. If it can't be heard, what is the point?
420  Bitcoin / Bitcoin Discussion / Re: Is Bitcoin about to be 51%ed??? on: June 10, 2014, 08:00:10 PM
Ghash.io did turn off 'some' hashes to lower their network %.

How would you verify that? an attacker keeps their hashing secret until they release their private copy of the chain.
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 [21] 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 »
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!