NXT :: descendant of Bitcoin - Updated Information

[wesleyh]

bit777 asks about Nxt Daemon https://bitcointalk.org/index.php?topic=183798.new#new

Ah thank you ! I was asking for help from devs but no one is moving.
If bit777 add Nxt this will be the biggest promotion ever.

With what exactly can we help? what do you need?
Pin

Just helping him understand how the client work and how to implement it to his casino (considering the high traffic/transactions).
The Nxt client is very different from BitcoinQt.

This is a main problem we see also appear with exchange, people dont know how to implement it in they environment. Can we have somebody other than core DEVs JL and CfB to help people with this?

who is willing to do this?

Pin

repost:


This is a main problem we see also appear with exchange, people don’t know how to implement it in they environment.

Can we have somebody other than core DEVs JL and CfB to help people with this?

Do we have someone other than core devs that could explain how NXT works and help them setup business; this is also in the interest of NXT

wesleyh?

Well certainly they're free to read the code at http://nxtra.org/api - it should be pretty self-explanatory. Setting up nxt isn't much harder than setting up a BTC clone. The problem is making sure it runs consistently and is not stuck or something like that.  While vircurex and bit777 are not programmed in PHP, the concepts are the same.

[bitcoinpaul]

going to a size-of-wallet voting scheme is plain MEDIEVAL - sure this has been brought up here before, but how about this:

Thanks everyone for the great ideas and discussion about voting, it's what I was hoping for

I agree with the above statement and that's why I proposed what I did. My concept of voting is only based on the American democratic process. So that's why I proposed how I did.

I am curious that no one commented about the account restrictions based on blockchain height I mentioned: If a vote was proposed and went public on block height of 45,000 an account would have to have existed for 'X' number of blocks before the block containing the vote (i.e, you have to be 18yrs old to vote in the U.S.), we could take it a bit further and say that that account would also have to NXT in it for a certain amount of blocks before the vote as well (like the 1440 blocks before you can forge). This could surely cut down on the amount of gaming no?

This makes a lot of sense on so many levels. Not only would it prevent gaming the system from large holders, but it prevents people from buying large sums of NXT (for a short period) to try to manipulate the vote.  The more I think about this approach, the more it grows on me.

I like the idea. What's the catch?!

[msin]

going to a size-of-wallet voting scheme is plain MEDIEVAL - sure this has been brought up here before, but how about this:

Thanks everyone for the great ideas and discussion about voting, it's what I was hoping for

I agree with the above statement and that's why I proposed what I did. My concept of voting is only based on the American democratic process. So that's why I proposed how I did.

I am curious that no one commented about the account restrictions based on blockchain height I mentioned: If a vote was proposed and went public on block height of 45,000 an account would have to have existed for 'X' number of blocks before the block containing the vote (i.e, you have to be 18yrs old to vote in the U.S.), we could take it a bit further and say that that account would also have to NXT in it for a certain amount of blocks before the vote as well (like the 1440 blocks before you can forge). This could surely cut down on the amount of gaming no?

This makes a lot of sense on so many levels. Not only would it prevent gaming the system from large holders, but it prevents people from buying large sums of NXT (for a short period) to try to manipulate the vote.  The more I think about this approach, the more it grows on me.

This is a great idea.  A forging account with public key will be able to get one vote.  Or as mentioned above, only an account that existed at a particular block height would get to vote on the particular topic.  That way you could go back in time and put up a vote for a smaller group of founders, etc...

[notsoshifty]

I believe 32K, which can hold 255 transactions, much fewer AM. Cost of AM at 1 NXT seems to be a bargain considering the space in the blockchain it uses.

If we say each block is worth 255NXT for 255 transactions or around 30NXT for AM, that makes AM almost 10 times cheaper. I strongly suggest AM cost of 0.01 is a disaster. It is long term storage in blockchain, distrubuted, redundant, etc.

James

I'm coming around to this thinking too: distributed redundant long term storage shouldn't be cheap.

How about a fee proportional to both:
 - size (bytes)
 - TTL (days)?

If we then had an efficient way to prune the blockchain, then the price of the transaction would at least be related to the cost to the network of storing the message.

Regarding size, once the network becomes busy this should resolve itself: forging nodes would include/exclude transactions on a fee-per-byte basis. But it has no incentive to take TTL into account.

[jl777]

Just found out about http://www.cryptocoinsnews.com/2014/01/15/breaking-news-russia-may-outlaw-bitcoin-russian-voice/

Seems Russia following in China's footsteps. war on bitcoin continues

James

[utopianfuture]

going to a size-of-wallet voting scheme is plain MEDIEVAL - sure this has been brought up here before, but how about this:

Thanks everyone for the great ideas and discussion about voting, it's what I was hoping for

I agree with the above statement and that's why I proposed what I did. My concept of voting is only based on the American democratic process. So that's why I proposed how I did.

I am curious that no one commented about the account restrictions based on blockchain height I mentioned: If a vote was proposed and went public on block height of 45,000 an account would have to have existed for 'X' number of blocks before the block containing the vote (i.e, you have to be 18yrs old to vote in the U.S.), we could take it a bit further and say that that account would also have to NXT in it for a certain amount of blocks before the vote as well (like the 1440 blocks before you can forge). This could surely cut down on the amount of gaming no?

This makes a lot of sense on so many levels. Not only would it prevent gaming the system from large holders, but it prevents people from buying large sums of NXT (for a short period) to try to manipulate the vote.  The more I think about this approach, the more it grows on me.

This is a great idea.  A forging account with public key will be able to get one vote.  Or as mentioned above, only an account that existed at a particular block height would get to vote on the particular topic.  That way you could go back in time and put up a vote for a smaller group of founders, etc...

+1 I like the idea. But if you go back to the first block, there aren't anyone else to vote except BCNext or anyone else ?

[opticalcarrier]

I believe 32K, which can hold 255 transactions, much fewer AM. Cost of AM at 1 NXT seems to be a bargain considering the space in the blockchain it uses.

If we say each block is worth 255NXT for 255 transactions or around 30NXT for AM, that makes AM almost 10 times cheaper. I strongly suggest AM cost of 0.01 is a disaster. It is long term storage in blockchain, distrubuted, redundant, etc.

James

Offloading off this thread,  I believe Ive good a pretty good scheme over here:
https://forums.nxtcrypto.org/viewtopic.php?style=1&f=17&t=587#p2631

(get rid of that "style=1&" in there if you want to use the default nxtcrypto forum theme.  I dont like the default as in I get more real estate efficiency by preempting the theme type in there.)

[jl777]

I believe 32K, which can hold 255 transactions, much fewer AM. Cost of AM at 1 NXT seems to be a bargain considering the space in the blockchain it uses.

If we say each block is worth 255NXT for 255 transactions or around 30NXT for AM, that makes AM almost 10 times cheaper. I strongly suggest AM cost of 0.01 is a disaster. It is long term storage in blockchain, distrubuted, redundant, etc.

James

I'm coming around to this thinking too: distributed redundant long term storage shouldn't be cheap.

How about a fee proportional to both:
 - size (bytes)
 - TTL (days)?

If we then had an efficient way to prune the blockchain, then the price of the transaction would at least be related to the cost to the network of storing the message.

Regarding size, once the network becomes busy this should resolve itself: forging nodes would include/exclude transactions on a fee-per-byte basis. But it has no incentive to take TTL into account.

Simpler seems better to me. The more requirements, the more complicated, the longer to get it working right.

I think there will be (is?) a bidding process for space in the forged block. As long as the bidding is based on NXT per byte used in the blockchain, then it won't matter if it is for transaction, AM, AS, etc. Mission critical items will need to bid higher, best efforts can use minimum fees, etc. I think it makes sense for the client to dynamically change the fee paid if something doesn't get into the blockchain. The core protocol should use some sort of universal metric to determine what goes in or out, NXT/byte seems logical.

There might be some EBS (emergency broadcast system) payloads that always make it, but not sure how to make it spamproof.

James

[gbeirn]

going to a size-of-wallet voting scheme is plain MEDIEVAL - sure this has been brought up here before, but how about this:

Thanks everyone for the great ideas and discussion about voting, it's what I was hoping for

I agree with the above statement and that's why I proposed what I did. My concept of voting is only based on the American democratic process. So that's why I proposed how I did.

I am curious that no one commented about the account restrictions based on blockchain height I mentioned: If a vote was proposed and went public on block height of 45,000 an account would have to have existed for 'X' number of blocks before the block containing the vote (i.e, you have to be 18yrs old to vote in the U.S.), we could take it a bit further and say that that account would also have to NXT in it for a certain amount of blocks before the vote as well (like the 1440 blocks before you can forge). This could surely cut down on the amount of gaming no?

This makes a lot of sense on so many levels. Not only would it prevent gaming the system from large holders, but it prevents people from buying large sums of NXT (for a short period) to try to manipulate the vote.  The more I think about this approach, the more it grows on me.

This is a great idea.  A forging account with public key will be able to get one vote.  Or as mentioned above, only an account that existed at a particular block height would get to vote on the particular topic.  That way you could go back in time and put up a vote for a smaller group of founders, etc...

+1 I like the idea. But if you go back to the first block, there aren't anyone else to vote except BCNext or anyone else ?

Well haha of course if you go back to the Genesis block But it would an arbitrary number. Need to be an account with a public key for at least the previous 10,000 blocks when the vote block come up. Or whatever.....

[opticalcarrier]

Project Kharon @ Nextcoin.org

CfB,

I have a lot of experience with SVM predictors. Basically given training data and answer, it can produce pretty accurate predictions. Overfitting can be controlled by special constant, so the smallest factor that gives acceptable prediction result is usually a good compromise.

Unlike NN that take forever to train when there are a lot of inputs, the SVM can handle almost arbitrarily large feature set. Millions of features can be used. Prediction is a simple dotproduct of feature vector with linear model.

I think here is finally a technical part of NXT I can really contribute to! Just need to develop a set of features and zombie/not-zombie categorization for each set of features.

James

Great. Post in that thread, plz.

Can't. Login didn't work. Asked to reset password, still waiting on email. Any chance of having thread on nxtcrypto?

James

nextcoin seems to be down right now.  Ive created the Project Kharon Forum & Thread at
https://forums.nxtcrypto.org/viewforum.php?f=56

[superresistant]

Perhaps I can help: I'm a Linux system administrator and a PHP programmer.

Well certainly they're free to read the code at http://nxtra.org/api - it should be pretty self-explanatory. Setting up nxt isn't much harder than setting up a BTC clone. The problem is making sure it runs consistently and is not stuck or something like that.  While vircurex and bit777 are not programmed in PHP, the concepts are the same.

Ok, I told bit777 about it. Thanks.

[utopianfuture]

going to a size-of-wallet voting scheme is plain MEDIEVAL - sure this has been brought up here before, but how about this:

Thanks everyone for the great ideas and discussion about voting, it's what I was hoping for

I agree with the above statement and that's why I proposed what I did. My concept of voting is only based on the American democratic process. So that's why I proposed how I did.

I am curious that no one commented about the account restrictions based on blockchain height I mentioned: If a vote was proposed and went public on block height of 45,000 an account would have to have existed for 'X' number of blocks before the block containing the vote (i.e, you have to be 18yrs old to vote in the U.S.), we could take it a bit further and say that that account would also have to NXT in it for a certain amount of blocks before the vote as well (like the 1440 blocks before you can forge). This could surely cut down on the amount of gaming no?

This makes a lot of sense on so many levels. Not only would it prevent gaming the system from large holders, but it prevents people from buying large sums of NXT (for a short period) to try to manipulate the vote.  The more I think about this approach, the more it grows on me.

This is a great idea.  A forging account with public key will be able to get one vote.  Or as mentioned above, only an account that existed at a particular block height would get to vote on the particular topic.  That way you could go back in time and put up a vote for a smaller group of founders, etc...

+1 I like the idea. But if you go back to the first block, there aren't anyone else to vote except BCNext or anyone else ?

Well haha of course if you go back to the Genesis block But it would an arbitrary number. Need to be an account with a public key for at least the previous 10,000 blocks when the vote block come up. Or whatever.....

One idea is that when a voting issue is raised and up for public debate we can mark the block at that time and only accounts "older" than that mark can vote on that issue.

Furthermore, I think setting up a vote should cost a certain amount of NXT to avoid spamming.

[ImmortAlex]

Nxt is a psychological experiment in group dynamics as well as a cryptoplatform.

+100500
And experiment is successful! Just look at page number

[jl777]

going to a size-of-wallet voting scheme is plain MEDIEVAL - sure this has been brought up here before, but how about this:

Thanks everyone for the great ideas and discussion about voting, it's what I was hoping for

I agree with the above statement and that's why I proposed what I did. My concept of voting is only based on the American democratic process. So that's why I proposed how I did.

I am curious that no one commented about the account restrictions based on blockchain height I mentioned: If a vote was proposed and went public on block height of 45,000 an account would have to have existed for 'X' number of blocks before the block containing the vote (i.e, you have to be 18yrs old to vote in the U.S.), we could take it a bit further and say that that account would also have to NXT in it for a certain amount of blocks before the vote as well (like the 1440 blocks before you can forge). This could surely cut down on the amount of gaming no?

This makes a lot of sense on so many levels. Not only would it prevent gaming the system from large holders, but it prevents people from buying large sums of NXT (for a short period) to try to manipulate the vote.  The more I think about this approach, the more it grows on me.

This is a great idea.  A forging account with public key will be able to get one vote.  Or as mentioned above, only an account that existed at a particular block height would get to vote on the particular topic.  That way you could go back in time and put up a vote for a smaller group of founders, etc...

+1 I like the idea. But if you go back to the first block, there aren't anyone else to vote except BCNext or anyone else ?

As long as it is 1 NXT, 1 vote, then it makes sense. Anything else opens up to Sybil attacks and tyranny of majority. Certainly using NXT in accts at the moment vote is started is a good idea.

I am confused why this is such a hot topic. America uses republic system, not democratic for most things. Representatives are elected and they then make all the decisions. Since this system is pretty much maps to $tbd USD = 1 vote, to elect the representatives/senators/etc, not sure what the big controversy over 1NXT==1vote is

Has anybody noticed that it is nearly impossible to get 10 founders to agree on any single thing? They are 10 different people. Also, combined, they own a minority, so if they somehow manage to join forces to try to confiscate everyone elses NXT (that is what everybody is concerned with right?) they will fail and probably cause a reverse.

Tyranny of the masses is far more likely. What happened in America, the more numerous poor people vote to transfer wealth from the wealthy to themselves. America goes from biggest creditor nation to biggest debtor nation in the world

In any case, the Voting System is a non-binding mechanism to gauge public opinion. Certainly for any NXT threatening issue, there should be no doubts that all NXT holders big and small will make a correct decision, on balance, eg. avoiding hard forks.

Also, my experience is that the founders are by and large decent and intelligent people and why people think they would intentionally do damaging things, confuses me

James

[notsoshifty]

Simpler seems better to me. The more requirements, the more complicated, the longer to get it working right.

I think there will be (is?) a bidding process for space in the forged block. As long as the bidding is based on NXT per byte used in the blockchain, then it won't matter if it is for transaction, AM, AS, etc. Mission critical items will need to bid higher, best efforts can use minimum fees, etc. I think it makes sense for the client to dynamically change the fee paid if something doesn't get into the blockchain. The core protocol should use some sort of universal metric to determine what goes in or out, NXT/byte seems logical.

There might be some EBS (emergency broadcast system) payloads that always make it, but not sure how to make it spamproof.

I'm all for simplicity.

Regarding transaction size: yes there will be a bidding war, and the forging node has an incentive to consider transaction size when deciding which transactions to include and which to exclude. It's aim is to push a block which includes whatever collection of transactions maximises its own fees.

But it has no incentives to take TTL into account, and so unless the network imposes some hard rule with arbitrarily-chosen fees (e.g. 0.0001 nxt per day) why would a client not always opt for the maximum TTL?

[PeercoinEnthusiast]

Concerning voting, it would be nice to get a poll up to outline some of the proposals and gauge community support. We could go on debating this for months and never reach a consensus.

[ImmortAlex]

Polls, polls, polls...
Is it true that hash didn't get a nessie for their work on logo?!

[CIYAM]

Polls, polls, polls...

Polls on Bitcointalk are simply a reflection of the number of alt accounts you have.

So why would you want that to decide anything?

[gbeirn]

Polls, polls, polls...

Polls on Bitcointalk are simply a reflection of the number of alt accounts you have.

So why would you want that to decide anything?

hahahahaah true

[ImmortAlex]

Polls, polls, polls...

Polls on Bitcointalk are simply a reflection of the number of alt accounts you have.

So why would you want that to decide anything?

I'm not reflecting in mirrors. My votes never appears in polls. My alt accounts do what they wanna do.