croutonrant #237 - the happy happy dream of decentralised mining!

Waffle is at 43% last 100 blocks...
https://chainz.cryptoid.info/drk/#!extraction

EDIT: 45%
ateJynz.png
 
VPS diversity is slightly less urgent of an issue I think. Subsetting greatly increases the cost of attack to MN functionality, you would have to wipe out a much greater percentage of them.

It's still a weakness, but how it gets addressed depends on how Evan goes about the IP obfuscation thing/multipath routing thing. If MNs were resolvable via something like DDNS then the fixed IP requirement would vanish, it would be more feasible to run them from any reasonably reliable domestic connection, which would aid distribution - and maybe numbers too. I can get a business connection with multiple IPs on fibre for not much more than I'm paying for my VPS instances, and have my MNs running on far better hardware, but I haven't got around to actually doing it yet.
 
If sites like chainz and drk.mn can show the extraction percentage of blocks found in the past 100 to 1000 blocks, there is a way of identifying the pools hashing. If that's the case, a simple check involving masternodes and/or every client could check current pool identifier's hash against network hash and if the result is >25-30%, orphan blocks just like it's already done for those pools that aren't complying with masternode percentage payouts. Seems straight forward. Then you cut the bullshit trying to convince people to mine on x pool or go p2pool. Might also convince these large pools to throttle back hash power to other X11 coins.

Further, if the argument is well the masternode network is centralized on a few VPS providers, then open the check to any and all clients on the network. Every client is capable of getting the network hash from the getmininginfo command. Pass around the pool identifiers and on majority, orphan those above X threshold. Then again, the statistics of blocks found are in hindsight, not current network hash. Could the same metric be used?

I'm by no means suggesting mining through masternodes. I like the idea of having the two entities separate, especially from an attack standpoint.
 
Last edited by a moderator:
thelonecrouton said:
Really baffles me that so few people seem to care about this. It basically destroys the whole value proposition of a cryptocurrency.

I hope Evan care spare it some thought.

I'm learning C++ so I can figure out the low level mechanics. Will be up to speed by early 2017. :confused:
Click to expand...
You need "Learn <insert anything here> in 21 days" book :grin:
 
oblox said:
If sites like chainz and drk.mn can show the extraction percentage of blocks found in the past 100 to 1000 blocks, there is a way of identifying the pools hashing. If that's the case, a simple check involving masternodes and/or every client could check current pool identifier's hash against network hash and if the result is >25-30%, orphan blocks just like it's already done for those pools that aren't complying with masternode percentage payouts.
Click to expand...

The sites you mentioned _know_ the addresses of the pools and compute the percentages from that. If a pool (or exchange) changes the address they are lost, so you can't hard-code the addresses.

But one thing you certainly could do: the Masternodes could log the block-finders and if some address approaches 40% over a 1000 (or whatever) block period it could reject those blocks.

But, this will open a BIG can of worms and it would be VERY important to communicate this properly.
 
crowning said:
The sites you mentioned _know_ the addresses of the pools and compute the percentages from that. If a pool (or exchange) changes the address they are lost, so you can't hard-code the addresses.

But one thing you certainly could do: the Masternodes could log the block-finders and if some address approaches 40% over a 1000 (or whatever) block period it could reject those blocks.

But, this will open a BIG can of worms and it would be VERY important to communicate this properly.
Click to expand...
Then you almost need some sort of registration system network wide that says hey, this hash belongs to this group. But then again, to your point, it would be no different than changing the mining address. Bah.
 
crowning said:
The sites you mentioned _know_ the addresses of the pools and compute the percentages from that. If a pool (or exchange) changes the address they are lost, so you can't hard-code the addresses.

But one thing you certainly could do: the Masternodes could log the block-finders and if some address approaches 40% over a 1000 (or whatever) block period it could reject those blocks.

But, this will open a BIG can of worms and it would be VERY important to communicate this properly.
Click to expand...
What will prevent pool from using 1000 addresses?
 
vertoe said:
Well people mining at Waffle probably don't know what Darkcoin is. All they want is most out of their miners. Can't blame them really.
Click to expand...
I was referring to 'us' - I'm aware that Wafflepool miners don't give a damn about Darkcoin. I don't care about them either, except that they and every big pool jeapardises the security of my DRK.

They are worthless parasites and if the problem they pose can be dealt with I will be a happier investor. :smile:
 
That "Unknown" pool is going to grow much larger once VTC switches algo, I'm going to contact the owner of it now to see if he can spread the hash when that happens.
 
https://bitcointalk.org/index.php?topic=715435.msg8754610#msg8754610

This is as far as I have read the Spreadcoin thread on BCT so far, but it seems this guy has come pretty close to making solo-mining de-rigeur. Pardon my French.

If Darkcoin could be made truly decentralised it would have vastly greater appeal to the wider world. Big money and their risk analysts are going to laugh at having to trust a handful of pools/pool ops, and walk away.

whitepaper: http://spreadcoin.net/files/SpreadCoin-WhitePaper.pdf
 
Last edited by a moderator:
thelonecrouton said:
https://bitcointalk.org/index.php?topic=715435.msg8754610#msg8754610

This is as far as I have read the Spreadcoin thread on BCT so far, but it seems this guy has come pretty close to making solo-mining de-rigeur. Pardon my French.

If Darkcoin could be made truly decentralised it would have vastly greater appeal to the wider world. Big money and their risk analysts are going to laugh at having to trust a handful of pools/pool ops, and walk away.

whitepaper: http://spreadcoin.net/files/SpreadCoin-WhitePaper.pdf
Click to expand...

This is quite interesting approach and it corresponds with one fernando gave link to

fernando said:
Read this about turning mining into a 2 phase process:

http://hackingdistributed.com/2014/06/18/how-to-disincentivize-large-bitcoin-mining-pools/

For the second phase you need to use your private key, so pools can't rely on other people's miners. They need their own hardware for that and that limits their growth. The second phase can require less hash power than the first so they only need a fraction of the total hash power under their direct control. That difference in difficulty could start small and increase with time to give time to people to adapt.
Click to expand...

However this "spreading" smells (and actually described by its developer) like "stealing" because a "cheating" miner still sends shares but keeps full reward for himself.

PS. While the developer of this coin actually has some programming skills he should be more accurate with Copyright things.
The coin is not a fork of Bitcoin but Darkcoin actually.:rolleyes:
https://github.com/spreadcoin-project/spreadcoin/commit/057007c2741e7c771489a714599dd20d17db48a1
https://github.com/spreadcoin-proje...n-descriptors/linux/gitian-linux-darkcoin.yml
https://github.com/spreadcoin-project/spreadcoin/commit/d9e169eb51230858bfd5059be02f6582fa20b704
etc.

PPS. I guess we could forgive him if he joins us though :grin:
 
You must log in or register to reply here.
Back
Top