• Forum has been upgraded, all links, images, etc are as they were. Please see Official Announcements for more information

some crap my brain did

camosoul

Well-known member
So, I had some free time inside my head, and I wandered into what the future may bring....

Masternodes really should be in charge of the network. But Proof of Work does still matter. Masternodes and IX have made the payment tracking not only instantaneous, but ansynchronous from the block discovery... It only makes sense to make Proof of Work, Blockchain creation, subservient to the Proof of Service... So then, we could go back to CPU mining. We wouldn't need to secure the network with a hash race. Hash validity could even have a cap integrated. A software based PLL so that hash is constant per node. We no longer have to settle for the rule of the wild. MNs ARE the Miners. Creating a blockchain becomes utilitarian and nothing more. We need a ledger to dump the locked TX memory pool into. We do need to mint at a predictable rate. But it is now asynchronous to the formation of TXes, and thus, takes a back seat, so to speak. It's no longer the central role, even though it does still matter. This also means there is no longer a need to split payments between miners and MNs, because MNs are the only miners. DGW seems to be overkill there, as the only time hashrate changes is when more MNs come or go. But overkill in securing the chain never hurt, right? May as well leave DGW. Cap on the hashrate so that VPSes don't peg the needle... We don't need crazy hashrate. Only a consensus of much more distributed Masternodes can mint a block, so diff only need be high enough to assure it's real, not beat out anyone else since they're not allowed in at all without PoService... Also, no more 51% problem. No more pools consolidating the hash "exit points." Vastly more distributed.

Anyway, I just wanted to think it out loud. I could be all wet, but it seems like the natural evolution...
 
I don't think that would work because the whole point of hashing is to get a difficult number that can't be reproduced in time before another block happens. And we need that hash number, according to the presentation, to choose the quorums. So they still need to be secure.

We also know every transaction will be an IX transaction, so that means that the MN quorums make the decisions. If that's so, then the miners are indeed subservient to the MNs.

I think the balance of power (or at least, the splitting of the work) is what makes our system so secure, and probably all there is to being 51% attack resistant. By making the MN network process all transactions as IX, means miners can't process any transactions, they just put it into the blockchain.
 
Remember, if there is a conflict the transaction reverts from Instant to normal.
 
I don't think that would work because the whole point of hashing is to get a difficult number that can't be reproduced in time before another block happens. And we need that hash number, according to the presentation, to choose the quorums. So they still need to be secure.
Yes, but because the only way to participate in the first place, is through Proof of Service with hashrate caps built in, it's plenty secure. No one can outrun it. If your hash rate is above the cap, you're automatically rejected. If you don't have 1000 DASH, you're automatically rejected. If your PoService score isn't < 6, you're automatically rejected. There's a qualifier for being part of the mining network now. Well, at least how I'm picturing it. Since it's not merely a factor of brute force anymore, you don't have to use brute force for security.
We also know every transaction will be an IX transaction, so that means that the MN quorums make the decisions. If that's so, then the miners are indeed subservient to the MNs.
And that's where I drew my conclusions. If all TXes are asynchronous and locked in via a separate system, there's no need to be so obsessed with the mining anymore. Brokencoin and it's clones depend on brute force, always having more hashpower than a would-be attacker, but anyone can participate. Since we're putting the PoService in charge, not just anyone can participate. So, that's actually better security. It not only discourages pooling consolidation, I mean, damn, the word pool means consolidate... It makes pooling impossible if a hashrate cap is included. Yes, the number needs to be secure. But, we no longer have to depend on that flawed system to get it, and can actually improve upon it drastically by relegating that system to what it's actually good for instead of creating an unnecessary dependency with 51% potential...
I think the balance of power (or at least, the splitting of the work) is what makes our system so secure, and probably all there is to being 51% attack resistant. By making the MN network process all transactions as IX, means miners can't process any transactions, they just put it into the blockchain.
Exactly! Which is why hashrate doesn't have to be such a big deal anymore. It's just a verifiable storage medium which is what we're finally realizing the blockchain is good for. Mining still matters as a way of securely generating that storage system, but we have more than one way to verify that, it isn't brute-force dependent, and can't be pooled, so every MN mining is an independent solo miner getting paid 100% of the reward because there are no other miners...

At least that's my vision...
 
Min and max hashes per block would make sense, might make it difficult for anyone running a MN on a rasbery-pi but it'd stop the network getting lazy and I'd guess the barrier to entry will tick down at some stage so the hashrate could go down but the difficulty stay the same.
 
Yah, I don't think an instantX was ever subverted. If all of the transactions are IX in the future, then I would think a messed up IX would simply go to another quorum?
 
Min and max hashes per block would make sense, might make it difficult for anyone running a MN on a rasbery-pi but it'd stop the network getting lazy and I'd guess the barrier to entry will tick down at some stage so the hashrate could go down but the difficulty stay the same.
I'm less worried about Pis, and more concerned about pissing-off VPS hosts... But with a limited access hash system, the cap could be low enough that an 8088 could handle it.

It's all tinfoil hattery speculation. I have no idea what Evan is doing with Evolution. This is just some goofy thought I had... One could use a hash of the payment vote history and skip the PoW altogether, now that there's a non-centralized deterministic method to it...

It'd be neat if Ugjin could comment on how far up my ass my head has gone. Nothing at all to do with DASH Evolution, just whether or not the idea would work, make sense, etc...
 
It just seemed to hit the nail on the head when I read through it. I'd thought of the masternodes managing a mining pool so the only way to mine was through them but it wouldn't be needed and it'd probably only serve any useful purpose as a stepping stone to say "don't even bother trying" and for some added confidence inherited from the Bitcoin way of doing things.

The bit that has me pretty intrigued now is the DAPI, somewhere it was mentioned that a wallet won't even be needed and that ties in with some other stuff I was thinking about, that instead of using a private key to access entries in a ledger you'll be using a key to access an interface to the current state of the ledger and that interface can do all sorts of seriously smart stuff, it's an entity on the network rather than an app for interacting with the ledger and the masternode network is the environment it exists on. I'd chatted with someone on BCT (CIYAM) at one stage about something along those lines, transactions that could exist across blockchains and a kind of VM or lib for them to interact with different clients but I'd never thought beyond that :)
 
Yes, but because the only way to participate in the first place, is through Proof of Service with hashrate caps built in, it's plenty secure. No one can outrun it. If your hash rate is above the cap, you're automatically rejected. If you don't have 1000 DASH, you're automatically rejected. If your PoService score isn't < 6, you're automatically rejected. There's a qualifier for being part of the mining network now. Well, at least how I'm picturing it. Since it's not merely a factor of brute force anymore, you don't have to use brute force for security.
And that's where I drew my conclusions. If all TXes are asynchronous and locked in via a separate system, there's no need to be so obsessed with the mining anymore. Brokencoin and it's clones depend on brute force, always having more hashpower than a would-be attacker, but anyone can participate. Since we're putting the PoService in charge, not just anyone can participate. So, that's actually better security. It not only discourages pooling consolidation, I mean, damn, the word pool means consolidate... It makes pooling impossible if a hashrate cap is included. Yes, the number needs to be secure. But, we no longer have to depend on that flawed system to get it, and can actually improve upon it drastically by relegating that system to what it's actually good for instead of creating an unnecessary dependency with 51% potential...

Exactly! Which is why hashrate doesn't have to be such a big deal anymore. It's just a verifiable storage medium which is what we're finally realizing the blockchain is good for. Mining still matters as a way of securely generating that storage system, but we have more than one way to verify that, it isn't brute-force dependent, and can't be pooled, so every MN mining is an independent solo miner getting paid 100% of the reward because there are no other miners...

At least that's my vision...
Not sure if I really understand how it works. But, I have a question. How can you enforce that hash rate cap? It is open source and people can modify the source code and fake a lower hash rate to do evil things behind. Right?
 
I do like the "everyone can participate" part. Instead of having so many pools that may suddenly have more than 51%, everyone should be able to mine on MN network. But, Evan said that he has solved the 51% problem. Let's wait and see what his solution is first. When will we hear from him on the technical details of DASH Evolution?
 
I'm less worried about Pis, and more concerned about pissing-off VPS hosts... But with a limited access hash system, the cap could be low enough that an 8088 could handle it.

It's all tinfoil hattery speculation. I have no idea what Evan is doing with Evolution. This is just some goofy thought I had... One could use a hash of the payment vote history and skip the PoW altogether, now that there's a non-centralized deterministic method to it...

It'd be neat if Ugjin could comment on how far up my ass my head has gone. Nothing at all to do with DASH Evolution, just whether or not the idea would work, make sense, etc...
One could possibly run 32 MNs with one single 1Gbps home connection with an Avoton 8-core or better CPU at the current moment. Paying VPS hosts is unnecessary.
 
There's no incentive to, if you can share the burden of mining between masternodes relatively equally then you can also share the rewards equally so there'd be no point trying to claim a greater share of the hashrate, be that through the MNs themselves securing the ledger or acting as a mining pool for external miners.

How would you run 32 MNs from a single IP? I think that's being worked on but for now multiple VPSs is the only simple way to manage it.

EDIT: Something else that comes to me with all these brain-farts in the air is Dash is addressing weaknesses that have only appeared in Bitcoin over time and with adoption and usage but that's not necessaries all the weaknesses that will appear over time, etc.

One issue I had to go and find for myself in the Bitcoin blocksize trollfest was issues that increasing the capacity would cause with network propagation times and recently there was talk of a 40% attack that worked by causing a divide in the network, lots of pool power either side but there's no crosstalk so a slightly bigger pool could, in theory, take over either side and the combined power cause a fork.

Network issues like that are just some things that could show up with time and they're probably already addressed but there are bound to be others.

EDIT: Just jokingly proposed a budget for beer but thinking about it, that's a weakness, something that benefits voters directly without benefiting the project. Idk, a constitution for what can be proposed springs to mind but that's messy, maybe a multi-stage implementation, locked categories of proposal and open proposals within them?

EDIT: Time skew correction might show up attempts to divide parts of the network. Someone having the majority of votes just came up on BCT and a regional component might both defend against that and have other uses as the network grows, local and global projects for ex.

YETANOTHEREDIT: Sorry for spamming this post, notepad.
 
Last edited by a moderator:
But, Evan said that he has solved the 51% problem.
I'm just sniffing my own farts on this speculation... Wondering how it could be done, following logical trails to where they lead. Nothing I say should be considered the path of the project... I'm just seeing if I can divine it with logic and what I know, which is limited but insightful, kinda.
When will we hear from him on the technical details of DASH Evolution?
Some gig in Mexico around Dec 4th is when some tidbits may be released.

Then a little more in Miami Jan 26th?

3 months sitting on the edge of my seat.

How long can a man wait for the world to grow up?
 
There's no incentive to, if you can share the burden of mining between masternodes relatively equally then you can also share the rewards equally so there'd be no point trying to claim a greater share of the hashrate, be that through the MNs themselves securing the ledger or acting as a mining pool for external miners.

How would you run 32 MNs from a single IP? I think that's being worked on but for now multiple VPSs is the only simple way to manage it.

EDIT: Something else that comes to me with all these brain-farts in the air is Dash is addressing weaknesses that have only appeared in Bitcoin over time and with adoption and usage but that's not necessaries all the weaknesses that will appear over time, etc.

One issue I had to go and find for myself in the Bitcoin blocksize trollfest was issues that increasing the capacity would cause with network propagation times and recently there was talk of a 40% attack that worked by causing a divide in the network, lots of pool power either side but there's no crosstalk so a slightly bigger pool could, in theory, take over either side and the combined power cause a fork.

Network issues like that are just some things that could show up with time and they're probably already addressed but there are bound to be others.

EDIT: Just jokingly proposed a budget for beer but thinking about it, that's a weakness, something that benefits voters directly without benefiting the project. Idk, a constitution for what can be proposed springs to mind but that's messy, maybe a multi-stage implementation, locked categories of proposal and open proposals within them?

EDIT: Time skew correction might show up attempts to divide parts of the network. Someone having the majority of votes just came up on BCT and a regional component might both defend against that and have other uses as the network grows, local and global projects for ex.

YETANOTHEREDIT: Sorry for spamming this post, notepad.
Crypto, as we have know it up to now, is just too overly simplified. It contains some very creative paradoxical rulesets, but it's too heavy handed, and that can be gamed.

Peer to Peer is excellent, but on a volunteer system, it's too easy to fragment, delay, etc... Exactly like TOR and why it sucks. MNs are incentivised. That's why the sideways approach by the suggestion of that old iris concept came to mind...the idea what thought of for a reason. It was just a bit out of place. Ultragenius Evan saw that coming. I'm pretty sure the sparkle of DASH Evolution was already in his eye this past February.
 
I understand that if all participate for a fair distribution of the block reward you would think there'd be no reason to game the system. But if you could double spend, if you could create coins, simply by being the one that found the block, then the ease of making a more powerful miner than everyone else is using would still be a problem. With everyone under control, except those that want to cheat, it'd be even easier to double, triple etc... the hash rate and cause problems deep down in the past of the blockchain. And the longest chain wins.

I don't see how that could be stopped. And with the current system of miners and MNs, I like the checks and balance between them. It's healthy. Just my opinion. Maybe I don't understand something again ;P
 
Back
Top