 Show Posts
|
|
Pages: « 1 [2] 3 »
|
It sounds like you have done a lot of research and experimentation in trying to understand the challenges of decentralized systems and how to improve their scalability and consensus mechanisms. Using other decentralized systems in the universe such as genetics, societal norms, and colonial organisms as reference points is an interesting approach. Your idea of using transactions as a mechanism for defining resource reallocation in a consensus mechanism sounds intriguing, and it's good to hear that you have done local simulations and talked to experienced people in the field.
It's understandable that you want to minimize community interactions and reliance until you have a solid solution, and that you want to be careful about how and when you release it to the public. It's important to be mindful of potential risks such as illegitimate dupes that could cloud market trust in a new consensus method. It may also be a good idea to consider the market demand for a better long-term consensus solution before launching it, as it can help you plan accordingly.
It's great that you are thinking about all these aspects carefully before launching your new consensus method, and I wish you all the best in your endeavors.
P.s.The mycelium network reminds me a lot of star trek
Thanks I appreciate the support, I'll try to update this topic when I make it public. Do you know what it's used for in star trek? |
|
|
|
ok so the OP is not actually trying to describe or learn PoW
nor wanting to learn how bitcoin uses PoW
but wants to instead make a whole different system where PoW means something completely different.. as thats how i am reading his posts now
where he wants to create a new system without first understanding how the current system works
..
so here goes another attempt at explaining the current system
PoW is not about "selecting proposals" or selecting a president who controls the network..
PoW is just doing a complicated method of using electric and computational time to create a strong ID that fixes the contents of a block. whereby it would take someone else more electric to go back and change that block and afix a new block ID to the changed data.. and then try to do the same for the next newest block to overtake the leading blockheight to become the valid blocks usernodes read and follow.
which to even be capable of doing this and consistantly win every block requires alot more electric spent than the entire networks worth of mining hardware
PoW has nothing to do with changing rules. its about securing the DATA inside a block with a strong and expensive ID for the blocks contents of transactions
and that is it in a nutshell
if a malicious pool owner with such massive amount of hardware to achieve repeated winning blocks. all he can ultimately do is just select (be picky) about what transactions to include or not
separately from PoW
changing the rules(proposals) requires having a majority of user nodes, or more importantly the nodes used by businesses and serives using your changed rules code.. to THEN accept blocks that may contain a new format.
meaning you need to convince user nodes to adopt your code first.. BEFORE a new block format would be accepted
thus 2 separate activities
as for who wins the block creation. its not about nodes randomly accepting something(being picky).
its about nodes receiving something random
its the mining pool that manages to solve the puzzle with enough difficulty faster than its opponent, and if the block meets the rules. then that is accepted as the winner
because there is a bit of 'randomness' in the blockID creation. its not always the pool with the most hashpower that wins every time.
dumbing it down by (whatever the term is for several factors above Vigintillion)
one pools attempt at making a block might be searching for the number 987654(xVigintillion)
while another pool is searching for the number 1234(xVigintillion) where the end result needs to when converted to hex have a prefix of a certain amount of zeros
their next attempts would be a different number where a different pool has to find a lower number meaning it gets to win even though other pools might have more hashpower their number this time they need to find is higher thus needing more time to find it
Im afraid you may be too focused on specifics of how bitcoins POW function operates to understand the relationship i'm explaining. You also seem to be misinterpreting the things I have said but if thats how you read it that may be on me speaking too abstractly. Believe it or not I have written my own POW functions so I understand how they work and calling me dumb because you don't understand my analogy is far from constructive. I understand you feel you fully understand how bitcoin works but how POW consensus operates on a macro level is where the analogy works. Once again, it is not about data processing/selection specifics as much as it is about representing the system from a singular network perspective. To summarize in simple terms and try to move away from this analogy. A decentralized network must be represented my some, all, or none of its members in consensus. Bitcoin POW falls in the category of being represent by 'some' of the network with those that solve the POW function being the only ones capable of making a public proposal. Government is also decentralized representation through 'some' of the population. Sure the selection methods might be different (solving POW functions and elections) but what i'm trying to explain is it is much more difficult for a network to adapt to network conditions if proposals are only made from a single network perspective. The way gossip protocols work not everyone in the network will have the same transactions at the same time. This means that not everyone will be building the same blocks. If the only proposal thats evaluated is the one that first solved the POW function that means only one network perspective is checked for each block(assuming two nodes don't solve the function at the same time, in which case longest chain is selected). While this method meets the minimum requirements for block security it makes it extremely difficult for the network to optimize itself as it doesn't have anything to compare with only one potential proposal. The main issue is proportional representation(some of system/network) rather than complete representation (entire network/system). |
|
|
|
|
I guess a better question to ask is could bitcoin ever replace CBDC's? (bitcoin outcompetes CBDC's)
|
|
|
|
I haven't got it yet, though: is this like private stuff? Academic work? If you have an idea, about a new consensus mechanism, which you haven't discussed yet, probably talking about it is wise choice.
I spent a bunch of time building a game on ethereum awhile back and when I discovered all the issues ethereum was having with consensus I became afraid to launch it and instead got real into trying to understand why decentralized systems always seem to always fail at scale. I ended up feeling like I was hitting a wall as I only had so many blockchain networks to reference because they all boiled to some iteration of POW or POS. Thats when I started looking at different observable decentralized systems in the universe because if you look at something like genetics as software its been debugging and optimizing itself for way longer that anything we've built. Things like societal norms, government, fungus, particle based physics, colonial organisms, etc. there were just so many reference options. I started playing around with different locally simulated p2p networks that I based off some of these systems to try and understand how they work. Thats when I realized that the transactions in an economic ecosystem advance state of an economy similar to the way environmental nutrients is used in fungal mycelium to advance its next physical state(mycelium uses the nutrients to grow in a way that optimizes nutrients access in the next state thus creating a cycle of continuous environment based optimization). Using this concept as a new base mechanism for fuelling consensus I started working on a consensus method that uses transactions as the mechanism for defining resource reallocation. At this stage I am fairly confident that the general consensus works as I have done multiple local simulations and talked to a few people I know who are fairly experienced in the field. Regarding why I haven't publicly discussed it, during my time working on the ethereum project I dealt with so many scammers and dishonest people that it made me want to minimize community interactions and reliance until I felt I had a real solution. Im also still feeling out how I want to release it as its very important to me that its launched properly distributed and decentralized. It seems that many of the potential risks with bitcoin go back to the way it was launched and initially distributed. I'm mostly just testing the market to see what demand is like for a better long term consensus solution so I can plan accordingly. If I make too much public too early people can start making illegitimate dupes which I feel may cloud market trust when it comes to trusting a new consensus method. and thanks ill check out matrix |
|
|
|
To summarize:
POW:
Selects proposal based on what node is making it.(selects potential proposal based on node who first solved POW)
Ideal:
Selects proposal based on best proposal for the current conditions
(Selects node based on best potential proposal)
It's not any first, it has to meet the difficulty too. And this is important because it also regulates the speed of blocks being mined. "First" is relatively easy to quantify. "Best" on the other hand... who and how to decide what's best, in an unbiased way, plus to make sure it's not cheated and also making sure the blocks are mined at the correct speed? (The who part is meant to be a trap, since the only correct "who" is the nodes.) "Best" can be quantified by basic quantifiable metrics like transaction speed, value spent, number of transactions, who spent transactions, where in the network they are spent from etc. just depends what the network is designed/optimized for Regarding First solution and block difficulty they are separate things. Block difficulty is constantly changing so the system approaches on 10 min block times. This means blocks will get easier or harder to solve based on how long the last block took to mine. When the network is operating smoothly nodes who propagate a solution to the POW function first are usually accepted as that makes that solution the longest chain. This usually means the first solution is accepted unless malicious/illegitimate activity is detected. |
|
|
|
|
Okay sounds good I will post the white-paper in next couple weeks once I have further tested the reference client. In the meantime for discussing more technical details I would rather not do it in public forums with the amount of fraudulence currently on the market. In the past discord was great for private communities but no one trusts invite links these days so is there a more trusted alternative?
|
|
|
|
|
I feel I understand how bitcoin works as I have rebuilt it multiple times testing various small changes, although I can understand why it was unclear what I meant. When I say bitcoin is like the government, I meant it in a very general sense with nodes acting as the citizens within the population. I am not referring to transactions acting as votes but rather nodes accepting/confirming the longest chain as their vote.
Citizens(nodes) do their jobs everyday to contribute to the economy which is the ongoing macro system(make and distribute transactions). Presidents(nodes) are randomly selected to group transactions for the next block through attempts at solving the POW function for that block. Just like a president a citizen(node) is selected for a brief time period to make potential decisions for the networks next economic state. The difference being instead of 4 years of decisions it is building 1 block/proposal worth of decisions.
When it comes to the analogy, I am referring less to the specifics of the voting/election aspect and more to how it is unlikely for a system to be represented accurately from a single point or origin. A president is selected and then treated as a temporary origin point for decisions. The population can always rebel(accept another chain), however, the problem more so lies in the fact that the “president “ only has access to a single market perspective for forming proposals.
This means each time a proposal is made it is only coming from the perspective of that position in the network thus limiting the systems ability to build the most efficient block for adapting to market conditions. Focusing on network communication you would be able to instead decide on the best potential block for the network rather than the block of the first person that solves the POW.
In this analogy you could take it even farther and say that mining pools are similar to political parties how the network uses mining pools to group candidates and simplify the selection process but I will admit it's a bit of a stretch.
To summarize:
POW:
Selects proposal based on what node is making it.(selects potential proposal based on node who first solved POW)
Ideal:
Selects proposal based on best proposal for the current conditions
(Selects node based on best potential proposal)
Transactions = participation in the economy
Proposals =transitions system to next economic/market state
Nodes = citizens
President = node/citizen who solves POW first that term(block)
Votes = which proposal/chain a node is willing to confirm/using
|
|
|
|
|
where would you recommend posting this paper?
|
|
|
|
|
if you use the government's attempt at decentralized consensus as an analogy for bitcoin. Proof of Work is like randomly selecting presidents and not revealing who they are until they are done with their term (pretend the president is actually the one who makes decisions). Sure you may prevent murder but if you select randomly it is rare you will select the best candidate for the current market condition, as you are limiting access to all the potential information. This makes it difficult for the network to optimize itself into the future as it is constrained to a single network perspective for each decision(proposal). Seeing as blockchain is software I feel like it should be built to scale alongside computational innovation rather than being restrained by it.(block size wars, time server constraints, pow function)
If you based a network in communication could it possibly be the best way for all relevant data to be compared and the system optimized for the next state. In an ideal world wouldn't everyone be able to properly communicate their perspectives and reach common ground? The only way to have all the relevant information is communication, and we can program computers to communicate clearly and uniformly through protocols. Why is network communication being minimized rather than optimized amongst consensus methods?
|
|
|
|
Firstly why not move this thread to beginners and help.
Then thread you want to create is it on this Forum? Or you want to create a website were all regarding decentralized consensus-ness will be discussed.
I want to create a small community of people that can be referenced for potential security risks associated with alternative consensus methods. I am not attached to what medium this is done through more so just looking for people that may be interested. Ideally something formatted similar to discord allows for easy casual chatting but I am not sure how the community feels about it at the moment so currently I am open to anything. |
|
|
|
|
I want to create a community or thread/forum for discussing new decentralized consensus methods. I was wondering if anyone would be interested?
|
|
|
|
|
I agree POW > POS by far I just think that POW may also be limiting the potential of decentralized currency in its own way.
It seemed promising a first but with all the required amendments is it really still the best solution out there or is it time to rethink the base principles of consensus.
|
|
|
|
|
Only two bitcoin pools were responsible for mining more than half of Bitcoin's blocks in the second half of December. Doesn't that prove POW is vulnerable as well in the sense that a few mining pools make up a super large portion of the network?
|
|
|
|
|
What do you think it would take for people to be open to a Consensus method that works off a new base mechanism(not based in POW or POS)?
|
|
|
|
|
each node is only responsible for the load of its neighbouring nodes so as the network scales, proposal processing load remains consistent on a per node basis. The time it takes for a proposal to propagate and be confirmed will be longer with larger networks, however, increasing the number of connections each node is required to have allows speeds to remain low.
Imagine a diagram that shows how force is distributed through the particles of an object. The forces travels through neighbouring particles and force transmission rate will increase based on the number of neighbouring particles. Ie. force travels faster through a solid than a liquid
|
|
|
|
|
if there was a new type of consensus what would be the best way to test it or notify the network though?
|
|
|
|
|
after doing dev work with ethereum and bitcoin it seems that both consensus options are blatantly fairly flawed. POW has obvious scaling issues and POS seems like a bandaid solution with wealthy people automatically being allocated more currency through an infinitely compounding cycle. Is the crypto/blockchain community not open to new type of consensus? It seems like POW and POS just keep getting updated rather than addressing the real issue and switching base mechanisms. Do others just not see issues with POW or POS?
|
|
|
|
|
If a new consensus method came along that was undeniably more efficient would bitcoin be able to come together enough to agree and make the change across the entire network?
|
|
|
|
|
in the midst of working a new type of decentralized consensus modelled after the way mycelium processes nutrients. It treats the network like a living system with transactions acting as network nutrients. Im looking for people who are able to point out potential security risks I may be missing before I make the network live.
|
|
|
|
|
With the current state of market trust what will incentivize the average person to use bitcoin over a CBDC?
Since banks can control supply they can use inflation and other mechanisms to encourage spending. Bitcoin on the other hand requires additional layers to handle even the current network traffic and lacks incentive for day to day transaction use.
|
|
|
|
|
