Topic: [ANN] the ScienceCoin Project - Initial Planning Phase. (Read 4991 times)

Bumping. Being in the ANN section this will probably be on the first page for about 30 seconds

This is interesting: http://www.wired.com/2015/06/mits-bitcoin-inspired-enigma-lets-computers-mine-encrypted-data/  I have basically 0% time to work on this now, but with this new development there is huge potential now with ScienceCoin.

Not much. I've been working on a few smaller projects, which once complete, will help me add some dev funding to
ScienceCoin. Obviously this can't be a fork of anything currently available, so I really need a team to help build this, or it's going to take many more months before a beta version can be released.

Great Idea. Wait to see what will happen!!

any news here?

freequant, how does Folding@home and the similar scientific distributed computing projects handle your points of objection?
As far as I know the proof of work bit is done by consensus, sending the same work to a small number (say 5) of miners/processors and seeing if they all come with the same result. For larger number of duplicates, some percentage of outliers can be discarded first.

In any case, whether this particular project succeeds or not, there are more immediate ways of helping science. The aforementioned Folding@Home (especially if you have nVidia cards) is one.

Helping to preserve the biodiversity of the planet we call Earth, is another. The dedicated team of researchers and volunteers from Durrell Wildlife Conservation Trust work tirelessly both in the Jersey zoo and in the field (e.g. Madagascar) to find the ways to preserve endangered species and to raise awareness among the local population so as to preserve the endangered specie's habitats. This work requires funds. A large portion is raised through the zoo itself. Another large portion comes from donations. There every little helps, and here this community can be contributing to save species from extinction.
I have established a good contact with the Trust, being a life-time member and donor myself, and started a donation fund-raising drive.

You can donate to 1DodoExzsNPvVRXFrgkKw6E259VjfUW8Kh and help another potential species of "Dodos" not to become extinct.

First CureCoin, now ScienceCoin: I like the idea of USEFUL altcoins based on something else that creators' GREED.
Good luck, I'll be rady to mine.

At this point, it looks like it will be more logical to give out ScienceCoins as a reward for work performed. So, 100% power will go toward the network, rather than 3/4. As for ensuring the integrity of each coin and transaction and preventing 51% attacks, there is certainly some work to be done. Something along the lines of ppcoin might be viable. 

As for confidentiality: This is the route we're hoping to go.

Not really worried about it and don't know what is Homomorphic_encryption. Some of sort of encryption used by homosexuals?
Could be quite a problem.
The real problem. In the time it requires to check the solution the miners could be running other useful computations. The beauty of Bitcoin is that the required time to check the solution is tiny compared to the time it takes to generate it.

Please tell us how you are going about addressing these problems.

Gonna add 2 of my own:

How are you going to  set the value of a work unit against a bitcoin? Maybe 1BTC will go to some absurd value against USD.

51% attacks are a documented phenomenon against chains. Worst case attacks are also possible against certain computational engines. How do you ensure such an attack is not carried out on your network without forcing the client to reveal their work?

freequant - all good thoughts that need to be addressed.  Lots of folks had the same attitude about Bitcoin when it was first proposed. There are those with vision and those without, stating that something is impossible and doing nothing is certainly the easy road.

ScienceCoin will require a LOT of effort to be successful, and I'm guessing the same for Curecoin, though I can't speak for cygnusxi as cygnusxi != Joe_Bauers, we just appear to have the same type of vision for a better future 

Here is why it isn't possible (so at least you can make a more credible ScienceCoin and CureCoin scams next time):
- confidentiality: you can't guarantee the confidentiality of the data that is being processed unless it is encrypted, but you can't run computations on encrypted data unless they are encrypted using an homomorphic encryption scheme and the operations you do on the data are valid operations within this scheme. At the moment, the best known homomorphic encryption supports two operations: addition and multiplication. Computing only with addition and multiplication isn't turing complete. So with current state-of-the-art encryption, you can't run general computing on a distributed network without leaking at the same time all the data you are working with, your code, and your purpose.
- quantification of calculus: if you want to use some operations for mining, you need to make sure that this operation is going to cost in average the same amount of work for every miner to do, OR you need to set the reward proportionnaly to the a-posteriori cost of the computation. In either case, it means that you (more exactly: the network) have to be able to determine how long the problem you give to miners is going to run and therefore how much it's going to cost. That looks like it's easy, but unfortunately for you, this problem known as the halting problem is one of the hardest problem in the whole realm of mathematics. Deciding when a random program is going to stop will need to run the program to find out, which means that to verify that the reward the miner granted herself is fair, the *whole* network will need to do the complete computation.
- asymmetry between resolution and verification: to be suitable for mining, an operation needs to be hard to solve, and easy to verify. This corresponds to a very precise class of functions known as one way functions. Unfortunately for you, there is only a very narrow set of operations that are actual one-way functions. The SHA256 algorithm used by Bitcoin is one of them. But the average operation your hypothetical user wants to do is more likely not a one-way function, or your user would have to jump through crazy conceptual hoops and be a math genius to manage to break down his problem into a system of one-way functions. So for the bulk of whatever people will want to do with your network, the whole network will have to recompute each solution found by a miner to decide if the solution is correct, which is stupendously wasteful.
- and the list goes on and on...

To summarize: your scheme doesn't allow confidentiality of data because it isn't possible, doesn't allow to share the work between miners equitably or reward the miners proportionnaly to the work done because it's not decidable, and doesn't allow to verify the solutions because no rational miner will accept to spend 99% of his mining time verifying the solutions of other miners.

So yes, this is a completely unrealistic idea. It is almost as hopelessly unrealist as ... well.. Curecoin.

On the other hand a scientist can just rent a bunch of Quad-GPU instances at spot on EC2 and get all their work done for cheap, in all confidentiality, with full turing-completeness and no crazy constraints on the type of operations.

That being said, there is one thing EC2 doesn't do well and that your scheme probably does better: rip off naive beginers from their hardly earned Bitcoins.


"Likely"... "Probably"..
You have a vibrant imagination, it's touching.


Damn, that one is unanswerable. How smart. You surely look like the type of self-less modern hero who is going to play a key role in enabling distributed cancer research.

Btw, thanks for getting all emotional in the wrong thread with the wrong profile.
So now we know that you are Joe_Bauers (so cygnusxi = Joe_Bauers), and that both ScienceCoin and CureCoin are scam attempts by the same clueless teenager.

We'll definitely have something like the lkml in the future. In the meantime, I will keep everyone updated here.
Ultimately, it would be preferred to devote 100% of the resources to the network, and perhaps, in line with what you suggest give the SC's as a reward for work performed.

Right now, I really need some additional devs to get this off the ground. I can't imagine we'll be able to use any more than 20% (if that) of the original Bitcoin code, so this will really need some serious skills to get it where it needs to be.

So, everyone please spread the word!  You can contact me by PM here - or at [myusername] at sciencecoin dotorg

i wish you had a mailing list or something to keep us uprised

I`m not sure how 2 mining scenarios will produce 1 coin, but that remains to be seen..

The best way to achieve this would be... mine the block, be forced to mine x% of a folding block, then give out the reward.

Thanks to everyone for the support so far!  Since this is still in the planning phase, I am looking for suggestions and features to help make this a success.

Once we finalize the requirements, the hard part (programming) will begin... I'm a mid-level programmer, so I will need some excellent programmers to help get this off the ground. Anyone who wants to be involved, please let me know.

go do what you do and buy some more hookers and crack. whatever it is you contribute...

ill save this post to put it in the about section of the next type of research coin that comes out. to remind people of the endless pile of idiots that think they know what is and isnt possible, who constantly screw things up by discouraging good ideas.

please , explain to us why this is impossible. if youre so damn smart. explain how there is no way to code it, and no way to make it work. dont just toss empty insults. back it up.

dont forget it is likely that a software team at stanford university is probably working on something like this right now. explain to me how they wont be able to figure it out, since you know more then the entire collective knowledge of stanford.

youre a fking tard

It's amazing how anyone can come up with complete unrealistic horseshit ideas, and still manage to convince a handful of naive dudes...
What's coming next? World peace coin? Global warming coin? Nuclear fusion coin? Time travel coin?

Guys - you can also get involved here to use your computing power while we wait for Joe_Bauers' project https://bitcointalksearch.org/topic/curecoin-foldinghome-team-2000-dgc-in-bounty-73013-193084

 

Watching this quite closely. My 570 GTX is useless for mining either SHA-256 or Scrypt while my 7950 is chugging away.

Also, mining for actually useful information is a fantastic idea! Hopefully this will gain some traction, as I will immediately shift to this if it does well.

 When your project is ready dont forget to sponsor the folding team "Cryptocoin Network"  The first folding team to get crypto rewards

I am generally sceptical to alt-coins, acknowledging only Bitcoin and Litecoin. This sounds like a genuinely brilliant idea. My GTX 560Ti performs on par with HD 6570 when mining Bitcoins, while at Folding@Home it runs circles around a moredatelty-size mining farm with 5-6 AMD cards.
ScinceCoins would unlock this large sector of hardware and potentially bring it into the crypto-currency world.
A thread worth watching...