FoldingCoin

by snikeris • 11 years ago

158 88 11 years ago

lkbm2• 11 years ago

Is this basically replacing Bitcoins and their proof-of-[useless]-work mining with a new currency whose proof-of-work is useful work (IE, FAH)?

If this is viable, it would be pretty grand. As cool as Bitcoin is, it's sad to me that it just eats up energy/computation time to ground its value. Imagine if all that computation time were put to good, scientific use while still providing the same proof-of-work benefit.

wyager2• 11 years ago

Imagine if all that computation time were put to good, scientific use while still providing the same proof-of-work benefit.

Cryptographic hashes have the unique property that their difficulty is quite predictable, which is an absolute necessity for PoW based voting systems (like Bitcoin). I don't believe there are any known scientific computation problems that have the same desirable PoW properties as a cryptographic hash algorithm.

fryguy2• 11 years ago

Also, they must be a lottery, in that it's lots of "easy" problems that have a small chance. Otherwise it weakens the "50% attack" by meaning that someone with the most computing power can efficiently "hard" problem much more reliably, rather than the their percentage of the total computing power.

oafitupa3• 11 years ago

Unfortunately, it's not viable. Proof of work needs to be "wasteful" (if you can consider validating transactions and keeping the network safe wasteful). I don't understand why so many people get angry at Bitcoin's use electricity and not at say, VISA's, or any other entity that isn't curing cancer.

nctr6• 11 years ago

There is also Proof of Stake to secure the network, which does not waste electricity as much and on top of that the amount of stake you get for each vote to secure the network could also be made dependent on the amount of work you do for a science project as measured by BOINC.

The identification of which user gets what amount of "Proof of research" bonus is done via hashed BOINC email, as the email in BOINC is private and cannot be accessed by others.

See my other post about gridcoin.

rodgerd• 11 years ago

Well, the fact that Visa are able to do four orders of magnitude more transactions that Bitcoin, at a minimum, indicates they're at least a great deal more efficient.

placeybordeaux• 11 years ago

Why do you say the proof of work needs to be "wasteful". The way I see it there are three major things:

1) It's hard to compute an answer (in bitcoin a low hash)

2) It's easy to verify an answer

3) The starting parameters can not be dictated by previous winners (the hash of the block in bitcoin)

This does not dictate that it can not be useful, in fact prime coin (http://primecoin.io/) could be used as an example of a cryptocurrency that has a moderately useful proof of work.

One proof of work could be based off of protein folding, however the parameters need to be chosen pesudo randomly or an attacker could store up a chain of multiple valid proofs of work then unlease a long chain all at once. So we can not direct the work, but the work can be done in a larger fertile space.

lkbm2• 11 years ago

I'm not angry about it, but I like to minimise waste. Visa wants to, and is better if it it can, minimise its computational overhead. Bitcoin is dependent on computational overhead. Seems like a very different situation.

I'm no expert--maybe it is necessary that the computation have no benefit other than being costly--but I don't understand why. If it does, I would welcome an explanation, though I recognize it's not your responsibility to educate me on the finer details of Bitcoin. :-)

PookTwo23• 11 years ago

Heres a good analogy:

Visa and the banks are the old way of doing things, they are analog. Blockchain technology and Bitcoin are digital.

Years ago when analog signals had no longer been useful and became outdated, it became a law that all signal for television must be broadcast in digital. This was because analog required so much more energy and bandwidth then digital does.

What bitcoin offers is a replacement for all the energy and resources that are poured into money transmitters, banks, credit cards, and even recently wall street. The technology is expensive to run, but not more expensive than all of the above combined.

And to answer the question of transactions per second, the Bitcoin network is endless. The problem isnt how many transactions can you do, the problem is storage. The blockchain currently is 25 GBs of data that must be stored by many nodes. Though not everyone has to have a nod, if many do not and only big companies do, then it can become once again centralized. Even though 25 GBs is not that much, that is with only around 100-200,000 transactions a day in the BTC network. Visa does 2,000 a second alone.

So if the blockchain was to transact 2,000 a second, you are looking at PB hard drives in a year or so. That is the issue to be concerned with. here is a great article for reading on possibly solutions / problems https://en.bitcoin.it/wiki/Scalability

great_kraken• 11 years ago

I recently was reading up on folding again. I thought, wouldn't it be great if people could manage to combine protein folding for a good cause with bitcoin mining. Cool idea.

lbotos• 11 years ago

Has anyone done the math to know if this is profitable in terms of power spent to mine vs mined coin value?

askmike• 11 years ago

This will be dominated by the price, which has been going down in the last 2 weeks [1].

[1]https://www.poloniex.com/exchange#btc_fldc

wyager2• 11 years ago

This is a terrible idea for several reasons, the chief of which is the fact that the difficulty of protein folding simulations is not easily predicted. One consequence of this is that if someone were to develop a more efficient algorithm for protein folding, they could easily dominate the mining market and pull off a 51% attack. You could argue that this might have a positive societal impact, but it would also probably destroy the foldingcoin network.

The reason Bitcoin (and most/all other good proof-of-work systems) use repeated application of a cryptographic hash is that it's A) very predictable how long finding a solution will take (on average) and B) unlikely that someone will destroy the proof-of-work system by inventing a more efficient method of calculating the hash.

PookTwo23• 11 years ago

Where there is profit there is competition. As a miner years ago, i was able to produce a lot of bitcoins with my GPU setup. When the ASICs came out i tried to compete, but simply did not have the capital. But other people did. And it wasnt just one person, it was a few big companies that did this.

So no longer is the individual able to mine BTC, its now a profession. So if this happened to FLDC, that wouldnt really be a bad thing, because if it is to happen, then that would mean FLDC has gained enough value for someone to pour money into being the top folder. So then the economy of FLDC would be where the general public has the interest and the Folders would simply be a profession.

mrb• 11 years ago

A decentralized digital currency where the proof-of-work has a second utility having a great commercial value (such as folding proteins) would be quite frankly revolutionary and would have a good chance at displacing Bitcoin.

But FoldingCoin is not that currency, because there is a huge downside in their design: they had to completely give up decentralization, which is the primary innovation and differentiator that block chain-based currencies like Bitcoin bring. Instead, FoldingCoin relies on the Stanford University Folding@home network, on the stats reported by this network, to distribute units of the currency proportionately to the amount of work. This means the Folding@home network could alter the distribution of coins at will, if it is attacked by hackers (likely), or if it operates maliciously (less likely). This means FoldingCoin would be unable to continue operating the day a single entity (the university) decides to stop running the Folding@home network. And so on. All the inconvenients and risks of centralization exist with FoldingCoin.

Trust me, I really wish it would be possible to design a decentralized digital currency built on a protein-folding proof-of-work, but so far nobody has found a way to do this in a decentralized way.

Edit: they openly acknowledge the fact FoldingCoin is not decentralized, see http://foldingcoin.net/fldc-vs-alts/ : "With Counterparty there is no current way of doing decentralized asset creation and issuance"

Edit #2: in theory a decentralized currency based on a protein-folding proof-of-work can exist and would work this way: compute a hash of a block of pending transactions and of the previous block hash (like Bitcoin). Use the resulting hash as a seed to deterministically generate a set of protein-folding problems. Miners try to solve them. Once a sufficient amount of these problems is solved (depending on a "difficulty factor"), broadcast the solutions and the transactions on the network, hence creating a new hash for the next set of protein folding problems. I don't know very much about protein folding, but as I understand the open problems are: (1) how to deterministically generate a set of useful protein-folding problems according to rules that should stay immutable for decades (you can't ask the participants to have to frequently update their mining software because consensus on these rules is hard to change), and (2) how to reduce the size of the solutions data published to the network every block without overflowing it (it is my understanding that protein folding would generate way more than a few hundred kilobytes every 10min which is the current average Bitcoin block size).

PookTwo23• 11 years ago

Hello, this is Robert with FoldingCoin :) I love that this discussion is here and I am replying to these comments right now.

There is no way of decentralizing with the Counterparty platform. I encourage everyone to read all of the great features that you get by being on Counterparty in the same link provided by @mrb. We feel also including the downsides is crucial to have a better understanding of what we are trying to accomplish.

Though we have measures in place to deal with the centralization issue we currently have:

1. We are incorporated as a Non profit in Indiana and we report to the government. We are 100% transparent about everything that we do. If we are to have malicious intentions, then we would be held accountable against the government.

2. We are setting up a board of directors to control the undistributed FLDC that we currently hold. This probably will not become a reality until the end of the year, but we are working on it. The first step was to become an official non profit found here https://secure.in.gov/sos/online_corps/name_search_results.a...

runeks3• 11 years ago

Why are you creating a digital currency and distributing this to Folding@Home participants? What's the point of this? Why would I want FoldingCoins?

Why not collect donations (in usable currency), and give that out to Folding@Home workers?

haliax• 11 years ago

Is it even possible to have a decentralized currency where the mining problem isn't directly related to the maintenance of the ledger? Won't people just be incentivized to mine rather than preventing double spends, and the like?

PookTwo23• 11 years ago

The counterparty protocol is built inside of the Bitcoin network. All assets (including FLDC) are built on this platform. It houses the assets inside of the Bitcoin network by using the 140 bytes of unused data in every BTC transaction. So everytime you send or receive FLDC, you will notice it shows up in the blockchain as a 0.00006 BTC transaction. This actually represents the Counterparty asset. http://blockscan.com/ is a site that decrypts this information and displays it as the assets, blockchain.info will only see it as a BTC transaction.

So this prevents the double spend, because Counterparty is secured by the BTC miners, so the hashing power of the BTC miners is the same as the Counterparty assets. Only an attack on the Bitcoin network would affect FLDC

bbcbasic• 11 years ago

I don't thinks so.

The proof of work needs to be tied to the block. It also needs to be very quick to verify but hard to compute (like getting a hash under a certain value). With scientific problems, you may need to solve a problem, then ever node on the network solves the same problem to verify what you did - which would waste a hell of a lot of resources!

darkmighty• 11 years ago

What if they kept a decentralized part and sort of "plug in" the proof of work part? They could have a fallback proof of work method s.t. if the outsourced proof of work fails they can keep the network running and call a decentralized voting to replace the source.

nctr6• 11 years ago

There is already such a currency, it is called gridcoin.

The network consensus is reached by Proof of Stake instead of Proof of work, so no computational cycles are "wasted" on proof of work. But how much stake you get with is also determined by the amount of research you contribute to the BOINC group gridcoin relative to the total research done in this group. See further details there:

http://wiki.gridcoin.us/Proof-of-Research

So there is two ways to mine gridcoin: with the normal proof of stake, which means you get an interest rate whenever you use your votes to secure the network and an additional "proof of research" bonus that you get with the stakes and that depends on your relative contribution to the gridcoin-BOINC team: http://boincstats.com/en/stats/-1/team/detail/118094994/over...

There are certain projects whitelisted, folding@home is one of them but there are many more.

What makes this even more interesting is the idea of commercial BOINC projects. As soon as you have a system where the inflation/money printing process depends on the computation power given to BOINC, BOINC projects to forecast stocks, AI, general machine learning tasks.. could also be created. Then this would be the first currency that has internal value, because the money in the inflation process does not come out of thin air but is based on computing power that is either used for science or for computing projects that deliever value.

The forum where the development is discussed is here: https://cryptocointalk.com/forum/464-gridcoin-grc/

TeMPOraL2• 11 years ago

BOINC projects to forecast stocks (...) could also be created.

That particular project would, IMO, defeat the whole purpose of gridcoin. The whole idea is to have computation do something actually useful, and if we'd be directing it into stocks (especially gridcoin stocks), it would be no different than bitcoin - i.e. wasting increasing amount of electricity just to support the very process of wasting it.

vanzard3• 11 years ago

Gridcoin's proof-of-research is not decentralized: it relies on the BOINC project servers. So it shares all the same centralization flaws as FoldingCoin which relies on the Folding@Home servers.

Gridcoin is a little different because if you remove the BOINC project servers, it is not centralized anymore, and only relies on proof-of-stake. But this would make it no different that the many proof-of-stake altcoins that already exist.

haakon• 11 years ago

Primecoin (http://primecoin.io/) has been out for a long time. It's a decentralised cryptocurrency which uses prime number chains as its proof-of-work. I don't think it revolutionised anything, but some mathematicians were a little excited when it was released.

jcoffland4• 11 years ago

Note that prime coin computes an archane form of primes that is of questionable value to anyone.

runeks3• 11 years ago

A decentralized digital currency where the proof-of-work has a second utility having a great commercial value (such as folding proteins) would be quite frankly revolutionary and would have a good chance at displacing Bitcoin.

Why?

Let's say someone were to solve the protein folding 20 years from now. It would no longer be a challenge. This would mean the crypto-currency that used the once-hard problem of protein folding would no longer work, as the mining problem would no longer be difficult.

I think a crypto-currency is much better off doing something useless, as the worst thing that can happen to a crypto-currency is mining no longer being difficult. We want mining to solve the most difficult problem there is: breaking hash functions. Not solving something that has a greater chance of being solved, because it would mean the end of the crypto-currency.

We need something that we think and hope will stay difficult for as long as possible. Hopefully, protein folding will not stay difficult, because there are immense advantages to it being easy.

venaoy• 11 years ago

Protein folding can't be "solved". For a small protein of 100 amino acids, there are 2^432 (20^100) possible combinations (http://xray.bmc.uu.se/~lars/biowww/Proteinfolds.html). Bitcoin's SHA256 has only 2^256 possible values. So in a practical sense protein folding is a harder problem that Bitcoin mining.

lappa8• 11 years ago

The reason SHA256 was chosen is that it is a hash function. Hash functions are meant to be easy to calculate but difficult or near impossible to reverse and near impossible to optimize.

MD2 was designed by cryptographers and has tons of research regarding it's security. It is now broken.

Protein folding basically has no research regarding it's security as a hash function, because it isn't designed as a hash function.

Protein folding almost certainly has a large optimization problem, an attacker likely could find ways to increase his mining power by orders of magnitude. If the currency became valueable, it is inevitable that someone would make solving this problem constant time.

Now, is making protein folding more a efficient a good thing? Absolutely! But the only way it can be made more efficient is by making this currency useless.

Essentially, the only way this currency can be useful to DNA folding is if the currency becomes useless for securing money.

Edit: It appears that foldingcoin is on top of the bitcoin blockchain so miners aren't even there for consensus. They are simply creating a speculative asset that has an increasing monetary supply that equates to donations to protein folding. The miners are completely unnecessary for security.

jonas21• 11 years ago

Protein folding almost certainly has large optimization problem, an attacker likely could find ways to increase his mining power by orders of magnitude.

As you mentioned, this is a good thing. And as far as the effects on the currency go, this doesn't seem any different from Bitcoin miners moving from CPUs to GPUs to ASICs.

If the currency became valueable, it is inevitable that someone would make solving this problem constant time.

Protein folding / structure prediction is NP-complete, so seems unlikely.

lappa8• 11 years ago

As you mentioned, this is a good thing. And as far as the effects on the currency go, this doesn't seem any different from Bitcoin miners moving from CPUs to GPUs to ASICs.

The optimization "problem" is a problem because it can wipe out the currency. The transition to better hardware was gradual, while a software optimization could literally make one persons CPU more efficient than the rest of the world combined.

Protein folding / structure prediction is NP-complete, so seems unlikely.

While it may be difficult to actually fold the protein, folding the protein such that it breaks one of the assumptions I listed may be trivial.

If we could simply use NP-hard proven problems as hash functions, we would.

dnautics• 11 years ago

"Protein folding / structure prediction is NP-complete, so seems unlikely."

Are you sure about that? I am not sure that a protein structure can be checked in polynomial time, given there is no algorithm to check a protein structure besides manual labor.

rcpt• 11 years ago

ASICs.

D. E. Shaw already has a machine for it http://www.dfi.uchile.cl/~jsgonzal/aton.pdf

letitgo12345• 11 years ago

Some formulations of protein folding are NP complete. That doesn't mean those models always correspond to reality. There's no evidence that nature is solving an NP hard problem when folding proteins (See the paper NP complete problems and physical reality by Scott Aaronson)

imaginenore• 11 years ago

Actually folding is sort-of similar to a one-way function. It's difficult to compute, but easy to verify. The verification is the calculation of the energy.

You give the same protein sequence to multiple miners, and the one with the minimum energy after time X wins the money. This will force the miners to find lowest energy solutions.

lappa8• 11 years ago

Being difficult to compute and easy to verify doesn't meet the full criteria for being a secure hashing function.

coderzach• 11 years ago

Who distributes the protein sequences?

jcoffland4• 11 years ago

It's not easy to verify at all. Free energy calculations are very expensive.

PookTwo23• 11 years ago

An attacker could alter the amount of points that they receive. It has happened in the past with Folding@home, but they have put measures in place to prevent this from happening again. Although it isnt a guarantee that someone will not find a work around, we would become instantly aware and a patch could be put in place to correct the problem. https://folding.stanford.edu/home/faq/faq-points/#ntoc3

Each core is digitally signed for security reasons and authenticity of a real WU http://folding.stanford.edu/home/faq/#ntoc45

PookTwo23• 11 years ago

Real quick, this forum does not allow me to post a lot of responses, so I encourage everyone to email me rross@foldingcoin.net or join in the discussion here https://bitcointalk.org/index.php?topic=781352.0 but i will still answer as often as this forum allows me to per day :)

The miners are there for a reason. Looking at the Counterparty protocol you will see that since the CP assets live inside the individual BTC transactions, that unless the miners verify the transaction to be accepted, then the assets will work in the same way that BTC does and not go through.

lappa8• 11 years ago

You didn't really address why you think the miners are there for a reason, you just asserted it.

Realistically, what you have done here is made a centralized currency in which you issue new coins for yourself then we trust you donate the coins to protein folding.

We can already donate to protein folding with Bitcoin so I don't see much use in making an altcoin that does it in a less useful way.

mintplant• 11 years ago

Can the result of one of these "Work Units" be easily checked for correctness? Or does someone have to redo the work to confirm?

PookTwo23• 11 years ago

if you are a certified medical researcher you can request to review the WUs. Many might not produce real results, but in the words of thomas edison:

I have not failed. I've just found 10,000 ways that won't work.

web007• 11 years ago

Sounds like http://primecoin.io/about.php but more widely useful. Good luck @PookTwo!

Forbo• 11 years ago

For a similar concept, there's also Gridcoin which uses the Berkeley Open Infrastructure for Network Computing as it's proof of work.

http://gridcoin.us/

stegosaurus• 11 years ago

Proof-of-work having secondary commercial utility is troublesome because it can result in a situation where the secondary utility becomes more valuable than the former, in which case the currency is liable to collapse.

I tend to think that proof-of-work must be inherently useless in order for it to function for this reason - otherwise, it's not 'work'.

PookTwo23• 11 years ago

I disagree. I think that Counterparty adds value to the Bitcoin platform. I envision BTC almost being like gold, not something that will ultimately be used everyday. I think a token built inside of Counterparty will be that everyday used token.

Its kind of like the internet, the original value of the internet was to transfer files from point A to point B. Then came: email, the browser, instant messaging, online shopping, social media, streaming, ext.....

This adds additional value to the Bitcoin blockchain thus making it more desirable to have.

But i do agree that the miners should work on useless algorithms, that is the best way to keep it just and fair, but building applications inside of that is just fantastic

jabgrabdthrow• 11 years ago

"Useful proof-of-work" is still less desirable for the coin holders than proof-of-stake, so capital will move out of this coin and into a POS coin

nosuchthing• 11 years ago

Seems harder to solve algorithmically, rather than as a single private company with clients who rent hardware / CPU / GPU time for hire.

grcnation• 11 years ago

Really decentralized cryptographically secured rewards for advancing research: Gridcoin (intro at http://uscore.net)

FoldingCoin

Related Stories