News for hackers. Stuff that matters.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by jl2718 on Tuesday February 23, 2021 @09:38AM (#26237749)
This may be misunderstanding the function of PoW. Bitcoin PoW provides three things: selection, security, and incentive. The compute resource burn is an externality, although, it also ends up providing an important function, economic distribution. For PoW to work,it has to be hard to produce, easy to verify, and most importantly, attest to one and only one state of the blockchain. The last part is where this fails as PoW. The solution to a puzzle is meaningless if the same solution can be shown as the result of two different inputs. So the puzzle must be practically unique to each proposed block (infinite possibilities), and ‘fair’ to each proposal. If you can find a useful puzzle with these properties, please don’t hold back.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by roboticmind on Tuesday February 23, 2021 @10:55AM (#26238718)
The coin is proof of stake based. The rewards for BOINC come on top of staking. Each BOINC project that is rewarded has its stats looked up by a system of oracles and each node verifies that the oracles match up. The project themselves can be removed if they act up and there's a system for voting on which project to add and remove

If you have any more questions feel free to ask me or check out the Gridcoin website https://gridcoin.us

Note: I help work on the gridcoin.us website and have ~5000 Gridcoin, so I may be biased in my view of the coin :)

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by jl2718 on Tuesday February 23, 2021 @12:48PM (#26240308)
The website appears to represent it accurately, and exciting for all the reasons that BOINC is.

The Wikipedia article says:

“Gridcoin attempts to address and ease the environmental energy impact of cryptocurrency mining through its proof-of-research and proof-of-stake protocols, as compared to the proof of work system used by Bitcoin.”

I don’t see this implementation as having any relevance to cryptocurrency mining in general. Hopefully you can find a way, but the Wikipedia statement seems inaccurate.

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by roboticmind on Tuesday February 23, 2021 @03:13PM (#26242293)
I know that many (including one of the core developers) have tried to correct various things on that Wikipedia page, but my understanding is that there's an admin that keeps reverting most changes

> I don’t see this implementation as having any relevance to cryptocurrency mining

That's why a lot of people refer to running BOINC to get Gridcoin as "crunching" instead of mining to help distinguish it

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by knocte on Wednesday February 24, 2021 @01:06AM (#26247098)
> Each BOINC project that is rewarded has its stats looked up by a system of oracles and each node verifies that the oracles match up

For cryptocurrency noobs: the word "oracle" while seeming to be a buzzword that tries to provide some technical legitimacy, actually means "a node with authority", which basically goes against the whole point of cryptocurrency (decentralization).

Move on, nothing to see here. Disappointed that this reached HN frontpage (and wikipedia).

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by roboticmind on Wednesday February 24, 2021 @07:31AM (#26249187)
I'd agree that it is more centralized than I'd like, but it doesn't defeat the whole point. Gridcoin has a different purpose than most cryptocurrencies. It rewards science and is more decentralized than most coins that try similar goals [1]. I'd also note that the oracles have less direct power over the coin than you might think. See one of the comments I made about it earlier [2]

The alternative was previously flooding every project and downloading a massive gigabytes of data from the projects. It wasn't maintainable and a project left from the large amount of bandwidth it used up. As Gridcoin grew, it became a larger and larger issue. As well, some projects are now requiring tokens to access the stats (for GDPR reasons), so for some project, it became impossible to reward them without anything like this

> ...seeming to be a buzzword that tries to provide some technical legitimacy...

I'd also note that most things about Gridcoin usually doesn't use the term oracle (uses scrapers instead). I was using it here since I thought it would be a term more here were familiar with

[1] https://gridcoin.us/wiki/advantages-and-features.html

[2] https://news.ycombinator.com/item?id=26240413

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by knocte on Wednesday February 24, 2021 @12:17PM (#26252579)
Sorry I can't buy the idea. Once you have some oracle, you have more centralization than a properly decentralized currency. And then you end up with the worst of both worlds: as expensive and slow as a cryptocurrency, and as trust-dependent/weak as a centralized system (weak in the sense that there's a single point of failure and a single entity that you must trust).

Anything that can only be done with some degree of centralization should just use a PostgreSQL database instead, no point in using a blockchain.

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by roboticmind on Wednesday February 24, 2021 @04:45PM (#26256471)
> Once you have some oracle...

> ...single entity that you must trust...

It's not a single oracle. There are currently 6 that independently run and gather the same stats. The wallet compares each of those independent oracle's data to each other and makes sure it matches. There will almost certainly be more oracles in the futures

> Anything that can only be done with some degree of centralization should just use a PostgreSQL database instead, no point in using a blockchain.

You get more resiliency in running it through a blockchain. You would go from oracles having to coordinate in a tricky way to manipulate stuff to now being able to having a database where they could somewhat easily directly change balances, stop things from running, ect.

A blockchain has some degree of centralization. For instance, how the very first peers found in blockchains is fairly centralized and relies on a lot of hardcoded nodes or servers. Now you can also seek out some connection too manually if you really want to, but most* have decided not to do that and accepted the trade off. Everything has a degree of centralization. It's not an all or nothing proposition. It is where you want to move on the balance of completely centralization to decentralized

* Partly because many aren't aware of this, but it still applies to those that are aware too

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by knocte on Thursday February 25, 2021 @02:51AM (#26260686)
> It's not a single oracle.

I didn't say "single". I'm out of this convo.

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by lappa on Tuesday February 23, 2021 @11:52AM (#26239542)
In that case, why have proof of stake at all? Wouldn't it be simpler to just have the set of oracles alone facilitate the currency, since they already effectively have the power to?
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by roboticmind on Tuesday February 23, 2021 @12:57PM (#26240413)
The oracles don't have the power to completely run everything. With the current split for rewards, 25% of new coins is from just staking. As well, because new blocks are made only proof of stake, a 51% type attack would be also be much tougher for the oracles conspire and to pull off. Even if they tried to redirect 100% of research rewards (28750 GRC/day) to one address, it would take over 900 days to even reach the same amount as the address with the largest balance (~6% of current coin supply) [1].

Further, there's also more resiliency. If the oracles somehow broke or conspired and all said everyone did zero work, the network would still run. Now granted it would not be great that only previously pending research and PoS rewards would be given out, but you could still send coins, run polls, etc.

[1] https://main.gridcoinstats.eu/address/ (excluding the very largest one (the foundation) because it's multisig and can't stake. In its case it would take 1000 days to reach)

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by vmception on Tuesday February 23, 2021 @12:13PM (#26239837)
The market didnt tolerate oracle controlled coins when Gridcoin was released

The market does tolerate that now

Today you wouldnt use a new blockchain you would just issue an erc20 token with an oracle server having an admin key for distribution. You could have people merkle mine to accumulate a stake, but this is primarily to keep the “crypto/computational” branding for that audience, as the oracle aspect would still be centralized

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by Moodles on Tuesday February 23, 2021 @01:14PM (#26240613)
I once saw another example coin like this where the "useful PoW" was finding chains of prime numbers or something. However, PoW solutions were not dependent on the current blockchain sate, which rendered it insecure as someone could just stockpile solutions and then hijack the blockchain. Is it the case here too?

The closest I've seen to a useful PoW coin is one which relied in intel SGX to prove that the work done was useful and honest, but of course the central root of trust is really Intel's SGX in that case.

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by DavidHilbert on Tuesday February 23, 2021 @12:15PM (#26239864)
> The solution to a puzzle is meaningless if the same solution can be shown as the result of two different inputs.

This isn't true. There are an infinite number of inputs that result in the same solution for bitcoin (or any hash)

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by Moodles on Tuesday February 23, 2021 @01:24PM (#26240749)
Right, but that's a little unfair? SHA-256 is literally a function designed to make it impossible to find such collisions. BIONIC problems aren't designed with that in mind.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by jl2718 on Tuesday February 23, 2021 @12:27PM (#26240031)
Okay let’s see an example. ;)
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by powerbutton65 on Tuesday February 23, 2021 @12:40PM (#26240218)
Like it or not they are technically correct. SHA-256 is a one way compression and furthermore the PoW for bitcoin doesn't require a match for the entire hash.

There's loads of examples of this in the bitcoin network whenever there's been a fork and eventually orphaned blocks.

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by jl2718 on Tuesday February 23, 2021 @12:55PM (#26240385)
Those are different solutions to the puzzle. Nobody has ever published a sha-256 hash collision. It has happened once with sha-1.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by neolefty on Tuesday February 23, 2021 @01:34PM (#26240921)
The principles are the same — it's just harder to find collisions for larger digest sizes. (Although bugs in sha-1 add an interesting wrinkle to the discussion.)

In fact one could argue that encryption without requiring infinite bandwidth or computation requires finite difficulty in math puzzles.

So our current approach to encryption is fundamentally vulnerable to (vastly) more powerful adversary computers. Only things like quantum cryptography break free of that limitation, by changing the ground rules.

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by jl2718 on Tuesday February 23, 2021 @01:56PM (#26241258)
Algorithmic cryptography depends on a computation time approaching infinity for perfect security. Quantum cryptography depends on a data transmission rate approaching zero for perfect security. Either way, perfect security takes forever.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by fogof on Tuesday February 23, 2021 @02:44PM (#26241913)
One of the more famous examples of a BOINC project is Rosetta@home, which asks users to determine how proteins fold. Isn't this exactly what you stipulate? You just let the binary hash of the block encode a sequence of amino acids, and then have the output be a valid block hash if the hash of the corresponding protein fold data has enough 0 bits at the beginning.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by bowmessage on Tuesday February 23, 2021 @02:51PM (#26242011)
who validates that the protein was folded correctly? Validating that the hash of some nonce contains enough leading zeroes is very easy for the network to validate. But verifying that users are correctly computing the protein folding (rather than quickly dumping out invalid folds to test hash values) seems like a difficult problem for the network to validate.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by jmngomes on Wednesday February 24, 2021 @03:13AM (#26247781)
BOINC tasks are validated by running multiple executions on different hosts/users and comparing results; credit, and hence GRC, is only attributed to tasks with valid results: https://boinc.berkeley.edu/trac/wiki/CreditNew
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by fogof on Tuesday February 23, 2021 @03:19PM (#26242363)
This is a good point. I was assuming that computing a protein fold was something that could be done relatively quickly, but it seems like these simulations are actually much more computationally intensive than that.

This is obviously not my area of expertise, but I wonder if there is any way to make more efficient but still useful computational questions about protein folding. For example, if we try to fold just part of a protein (only 100 or so amino acids, say) and do it at a much lower fidelity of simulation, is the result still helpful?

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by dfgdghdf on Tuesday February 23, 2021 @10:13AM (#26238159)
Exactly this. The PoW scheme must commit you to a particular chain.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by Robotbeat on Tuesday February 23, 2021 @10:13AM (#26238165)
Can you just hash a new solution to the puzzle with the new proposed block?
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by proto-n on Tuesday February 23, 2021 @10:16AM (#26238210)
Solution has to depend on the proposed block, so that a solution is not reusable for the next block
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by Robotbeat on Tuesday February 23, 2021 @10:20AM (#26238254)
The solution is also announced in cleartext, so you can check against previous solutions.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by AgentME on Tuesday February 23, 2021 @01:51PM (#26241184)
The problem is that when a miner announces a block with a solution, an attacker could immediately announce their own alternative block with the same solution and there would be a chance that the network would take up their block first.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by Robotbeat on Tuesday February 23, 2021 @06:41PM (#26244598)
But if the legitimate miner announces & signs a hash of the solution well beforehand, they can prove theirs is the legitimate solution.
Re:Gridcoin: An open-source cryptocurrency that rewards volu... by AgentME on Wednesday February 24, 2021 @03:47AM (#26247982)
Then that adds a delay to each block being broadcast to the network. It also means that the longer a miner waits to broadcast the block, the more time a different miner has to finish a proof-of-work and claim the next block first. Miners are all racing each other, so they're all going to try to wait the most minimum amount.

The logic around only accepting a block if you previously saw a signed hash is vulnerable to sybil attacks. An attacker could run thousands of nodes on the p2p network, never relay anyone else's signed hashes, steal the proof-of-works from completed blocks and then broadcast signed hashes of its own blocks with the stolen pows and then broadcast those blocks. Many users might find themselves connected to only nodes run by the attacker and would get tricked into accepting the attacker's chain.

The proof-of-work not depending on the blockchain state means that a miner could queue up lots of complete proof-of-works and then mint a ton of blocks with them all at once to do a 51% attack.

Any solution to this is either going to be centralized or look like proof-of-stake, and at that point, the proof-of-work is extra and not actually load-bearing at all.

Re:Gridcoin: An open-source cryptocurrency that rewards volu... by proto-n on Wednesday February 24, 2021 @10:12AM (#26250690)
Ok, not being usable for the next block is not enough. It must not be usable for any other block. Competing chains, forks, other miners stealing your rewards, etc. Has to depend on the hash of the previous block, and also the transaction where you send the reward to yourself (-> even if someone else "steals it", you get the reward so it doesn't matter).