The Cypherfunks[FUNK]: a coin for a global band! Talking v2. Join! 80+ songs

[soundposition]

Here are some interesting, (perhaps useful) articles I came across in the last hour doing research for the Cypherfunks..

Hidden surprises in the Bitcoin blockchain and how they are stored: Nelson Mandela, Wikileaks, photos, and Python software
http://www.righto.com/2014/02/ascii-bernanke-wikileaks-photographs.html

Putting the Blockchain to Work For Science
http://www.b8coin.com/2014/05/putting-blockchain-to-work-for-science.html

[n00bnoxious]

Putting the Blockchain to Work For Science
http://www.b8coin.com/2014/05/putting-blockchain-to-work-for-science.html

There's a coin in development called NDcoin, designed to fix the issues of using non-deterministic algorithms within the blockchain - these issues being exactly why coins like Primecoin, Riecoin et al have failed.

I actually spoke with someone involved in Gridcoin recently, and was most unimpressed. Their entire system seems to be dependent on people WANTING to donate their machines to science, but the coin itself doesn't actually succeed in monetising anything, as they haven't worked ANYTHING into the system by which coin ownership is incentivised, and thus there is no reason at all for anybody to do anything but dump the coins, which is destined to as previously with Rie, Prime and the rest, destroy the coin's value.

NDCoin takes a different attitude to this: They are building a decentralised miner-rental system. The idea being you can submit complex computation to the network by spending these coins, and you will be rewarded with your algorithms being processed by the miners, in return for which they receive the amount you paid in a fully distributed fashion. It's extremely clever, and offers good incentive for the big 3: ownership, sales and purchase.

I think this is in danger of zipping off on a tangent, so I'm gonna leave my discussion of these coins at this. It's very interesting though...

[simondlr]

Putting the Blockchain to Work For Science
http://www.b8coin.com/2014/05/putting-blockchain-to-work-for-science.html

There's a coin in development called NDcoin, designed to fix the issues of using non-deterministic algorithms within the blockchain - these issues being exactly why coins like Primecoin, Riecoin et al have failed.

I actually spoke with someone involved in Gridcoin recently, and was most unimpressed. Their entire system seems to be dependent on people WANTING to donate their machines to science, but the coin itself doesn't actually succeed in monetising anything, as they haven't worked ANYTHING into the system by which coin ownership is incentivised, and thus there is no reason at all for anybody to do anything but dump the coins, which is destined to as previously with Rie, Prime and the rest, destroy the coin's value.

NDCoin takes a different attitude to this: They are building a decentralised miner-rental system. The idea being you can submit complex computation to the network by spending these coins, and you will be rewarded with your algorithms being processed by the miners, in return for which they receive the amount you paid in a fully distributed fashion. It's extremely clever, and offers good incentive for the big 3: ownership, sales and purchase.

I think this is in danger of zipping off on a tangent, so I'm gonna leave my discussion of these coins at this. It's very interesting though...

Following a more "appcoin" model it seems. Very, very interesting times!

[soundposition]

Putting the Blockchain to Work For Science
http://www.b8coin.com/2014/05/putting-blockchain-to-work-for-science.html

There's a coin in development called NDcoin, designed to fix the issues of using non-deterministic algorithms within the blockchain - these issues being exactly why coins like Primecoin, Riecoin et al have failed.

I actually spoke with someone involved in Gridcoin recently, and was most unimpressed. Their entire system seems to be dependent on people WANTING to donate their machines to science, but the coin itself doesn't actually succeed in monetising anything, as they haven't worked ANYTHING into the system by which coin ownership is incentivised, and thus there is no reason at all for anybody to do anything but dump the coins, which is destined to as previously with Rie, Prime and the rest, destroy the coin's value.

NDCoin takes a different attitude to this: They are building a decentralised miner-rental system. The idea being you can submit complex computation to the network by spending these coins, and you will be rewarded with your algorithms being processed by the miners, in return for which they receive the amount you paid in a fully distributed fashion. It's extremely clever, and offers good incentive for the big 3: ownership, sales and purchase.

I think this is in danger of zipping off on a tangent, so I'm gonna leave my discussion of these coins at this. It's very interesting though...

Couldn't someone in this case with a lot of fiat money just buy it all up?

Sorry.. I'll try to stay focused, but I do think this is relevant.

[n00bnoxious]

Couldn't someone in this case with a lot of fiat money just buy it all up?

Yes, but to do so they'd have to whale-swipe the exchanges, which would cause everyone to remove their sell orders and put them in much higher... Thus busting the price up hugely.

[soundposition]

Couldn't someone in this case with a lot of fiat money just buy it all up?

Yes, but to do so they'd have to whale-swipe the exchanges, which would cause everyone to remove their sell orders and put them in much higher... Thus busting the price up hugely.

So like.. What if a bunch of corporations wanted to decimate the 'coin world'. Could they 'whale-swipe' the exchanges and then be the majority holders of this "hugely pricey" coin?  Money for the rich?

[n00bnoxious]

So like.. What if a bunch of corporations wanted to decimate the 'coin world'. Could they 'whale-swipe' the exchanges and then be the majority holders of this "hugely pricey" coin?  Money for the rich?

Yes, but then the price would stagnate unless they used their coins, and other BTC to pump it.... That's exactly how pump groups work. They buy up a coin hugely cheap, then use other funds and their high-stakes in that coin to manipulate the market until they can dump a bunch for profit.

[mckmuze]

The Bitcoin blockchain is already too big to download. Electrum figured out how to use specialized servers that index the blockchain:

Instant on: Your client does not download the blockchain. It uses a network of specialized servers that index the blockchain.

http://electrum.qc.to/

It is open source..

Yeah Electrum is a thin client implementation. The issue with this is you need a strongly secured network BEFORE these get put in place. Without the pre-existing network security it hugely damages decentralisation, as these clients basically just steal the information they need to maintain balances, and give back nothing at all except extra network traffic. Even PoS staking secures the network, but without running a full node, I don't think this is possible.

How to prove you've made music? Trusted Timestamp:
https://www.btproof.com/

It even says on their website:
Musicians and other digital art producers can prove when they created their work.

Looks awesome!

http://www.proofofexistence.com/
Also found this one, doesn't give certificate error either.

[n00bnoxious]

http://www.proofofexistence.com/
Also found this one, doesn't give certificate error either.

Looks great, but the problem I see with these is they're all data-agnostic and centralised. We'd have to be able to decentralise it AND detect music somehow - whether that's by individuals listening and voting for and against, or automated metadata checking, a combination, or something else entirely...

[groggin]

    just a thought: don't even think about loading up the bc with music, fill it instead with .torrents, trhey're only a few kB, link to public tracker, files are hosted by artists and fans, who upload more torrents ... just have to figure out how to auto-mod the input.

[simondlr]

http://www.proofofexistence.com/
Also found this one, doesn't give certificate error either.

Looks great, but the problem I see with these is they're all data-agnostic and centralised. We'd have to be able to decentralise it AND detect music somehow - whether that's by individuals listening and voting for and against, or automated metadata checking, a combination, or something else entirely...

Yeah. Difficult problem. You basically have to build out consensus networks outside from a basic ledger. Something like reality keys almost. https://www.realitykeys.com/.

I reckon we first try to improve the economics of the coin, keep it simple, and keep the discussion going on how to incorporate music into a blockchain: either through consensus, or cryptographically. Both are hard, but we are breaking new ground here. Let's change the world!

[groggin]

                                         hypernotes
                                         melody
                                         sonic boom
                                         harmonEy
                                      

[n00bnoxious]

I reckon we first try to improve the economics of the coin, keep it simple, and keep the discussion going on how to incorporate music into a blockchain: either through consensus, or cryptographically. Both are hard, but we are breaking new ground here. Let's change the world!

100% - economics and new coin code are priority 1, the rest can be added later. Just the process of coding the new coin will give us all time to have a good think and investigate into usable technologies (or even new ones if we're feeling really adventurous :-D)

Yeah. Difficult problem. You basically have to build out consensus networks outside from a basic ledger. Something like reality keys almost. https://www.realitykeys.com/.

It's more than possible to include extra information in each block (Darkcoin recently did exactly this to store their Master Node votes) - perhaps we could build a system which allows individuals to submit a single file by concensus per block, as groggin mentioned perhaps utilising the Bittorrent protocol. These tracks could then be streamed from all around the world via magnet/DHT once they had reached a certain level of propagation. The bandwidth requirements of the protocol would be marginally bigger, but no more than running a music player like Spotify constantly, and wouldn't have any kind of centralised server requirements.

At 1 minute block times, this would allow for a maximum 1440 new submissions every day. Is that enough?

People could perhaps even only stream their own "favourites" list or something, encouraging people to produce music that others enjoy, as they wouldn't have any incentive to produce or submit garbage because it would never propagate, and thus would consistently waste their own bandwidth. It would also further incentivise staking, as having your wallet open would be a necessity while initially broadcasting any new track. It'd be like running your own, totally decentralised radio station!

Starting to feel like we might be getting somewhere :-) Definitely feeling the excitement vibe!

EDIT: Another thought: To offset the inflation of PoS, and to de-incentivise spamming the network with garbage, we could introduce collateral payments to make a submission, thus to spam the network would be a considerable expense.

[simondlr]

It's more than possible to include extra information in each block (Darkcoin recently did exactly this to store their Master Node votes) - perhaps we could build a system which allows individuals to submit a single file by concensus per block, as groggin mentioned perhaps utilising the Bittorrent protocol. These tracks could then be streamed from all around the world via magnet/DHT once they had reached a certain level of propagation. The bandwidth requirements of the protocol would be marginally bigger, but no more than running a music player like Spotify constantly, and wouldn't have any kind of centralised server requirements.

At 1 minute block times, this would allow for a maximum 1440 new submissions every day. Is that enough?

People could perhaps even only stream their own "favourites" list or something, encouraging people to produce music that others enjoy, as they wouldn't have any incentive to produce or submit garbage because it would never propagate, and thus would consistently waste their own bandwidth. It would also further incentivise staking, as having your wallet open would be a necessity while initially broadcasting any new track. It'd be like running your own, totally decentralised radio station!

Starting to feel like we might be getting somewhere :-) Definitely feeling the excitement vibe!

EDIT: Another thought: To offset the inflation of PoS, and to de-incentivise spamming the network with garbage, we could introduce collateral payments to make a submission, thus to spam the network would be a considerable expense.

I like where this is going. That gives additional incentive for the coin. You need it in order to broadcast a song to the network. I like the idea as well where staking only occurs if you are contributing to the broadcasting network. And yes, you choose what YOU want to broadcast. So if it's gibberish, a user has basically *paid* to upload a song and it's not getting broadcast.

I feel we're getting closer here!

[n00bnoxious]

I like the idea as well where staking only occurs if you are contributing to the broadcasting network.

This does seem to be closing the trifecta... We have incentive to OWN, incentive to SPEND and incentive to BUY... All of the elements are there. Just got to ensure they're all solid. Also IRL usage is integrated in the core concept - people love to listen to and discover new music.

There's even room for keeping the tipping involved in it. Someone's public address could be incorporated in the magnet metadata, so tipping would be a matter of pressing "Tip" in the wallet, and it'd send a tip DIRECT to the person who uploaded its wallet.

In fact could tipping be a double vote (a "like" being one)? Obvious abuse potential, but considering it costs money it doesn't really harm anything because to keep it high in the charts would cost a fortune...

[simondlr]

I like the idea as well where staking only occurs if you are contributing to the broadcasting network.

This does seem to be closing the trifecta... We have incentive to OWN, incentive to SPEND and incentive to BUY... All of the elements are there. Just got to ensure they're all solid. Also IRL usage is integrated in the core concept - people love to listen to and discover new music.

There's even room for keeping the tipping involved in it. Someone's public address could be incorporated in the magnet metadata, so tipping would be a matter of pressing "Tip" in the wallet, and it'd send a tip DIRECT to the person who uploaded its wallet.

In fact could tipping be a double vote (a "like" being one)? Obvious abuse potential, but considering it costs money it doesn't really harm anything because to keep it high in the charts would cost a fortune...

A tip could indeed count as a vote if address is embedded with the song!

[groggin]

i still have my little gpu pointed at the p2pool, just wanted to post that the wallet is still unstable, falls out of synch 2-3 times a day, last 5 days

pulling in 3 mil a day w/88 khps anyway   

[simondlr]

i still have my little gpu pointed at the p2pool, just wanted to post that the wallet is still unstable, falls out of synch 2-3 times a day, last 5 days

pulling in 3 mil a day w/88 khps anyway   

Indeed. Witnessed similar problems. Low hashrates seems to break the KGW. The variance seems to be a problem. Thanks for keeping the old chain going though Groggin!

[BuckSnot]

We need a good place to exchange midi and sound files back and forth with everyone so we can collab with everyone easier? I have little time & even less talent but I like the ideas going on here ;]

[jwinterm]

We need a good place to exchange midi and sound files back and forth with everyone so we can collab with everyone easier? I have little time & even less talent but I like the ideas going on here ;]

agree BuckSnot, and I think for the time being mega or dropbox would probably suffice for sharing files, maybe post links on subreddit or here I guess...