Author

Topic: NXT :: descendant of Bitcoin - Updated Information - page 659. (Read 2761642 times)

legendary
Activity: 1470
Merit: 1004
** Cryptsy Update **

I just heard back from Paul (met him at the Miami Bitcoin Conference), they are looking to implement Nxt in the next few weeks.
I talked with Big Vern @Skype and I introduced CfB to him - he said he will talk to Paul about it (to get help from CfB if needed).

Fantastic. I did a virtual introduction with all the different devs including CfB and Jean-Luc a few weeks ago as well... they have the resources, now it's just a matter of finding time to implement. Nxt keeps taking another step.

Let me make a suggestion to keep this ball rolling.  We should ask the Cryptsy folks to document what was required to get NXT plugged into their system.  Then we should turn right around to allwelder or some of the other Chinese NXT members and see that this info from Cryptsy gets communicated to BTX38.com in China.  We need to be on both of these exchanges as soon as possible.  It is the only way we are going to get exposure to a new pool of capital to buy NXT and get our exchange volume back up.  

+100, let's be really proactive about getting on exchanges, this should be Marketing Committee's priority (once they are established)
legendary
Activity: 1162
Merit: 1005
I run public node on Android TV stick
Expect some posts like

Quote
Why my node on Panasonic microwave oven forges so little blocks? Should I upgrade it to Zanussi?

 Grin Grin Grin

+1000  Grin
full member
Activity: 238
Merit: 100

not exactly see you in maths
...

no, your comparison is between leased-effectiveBalances scenario and your shared-fee-group scenario, where the only difference is the number of forging nodes.

my comparison is between current state (non leasing) and your shared-fee-group scenario.  How are they not identical?
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
this I don't understand at all - what exactly was your algorithm for calculating weights?

As stated in my other posts my initial algorithm was not a good "model" (and needed "log" to give the expected results).

It wasn't taking into account the idea that your "ticket number" actually "changes with time" (my updated version does that and produces very much the same as "log" results that the earlier version got without actually using "log").
full member
Activity: 168
Merit: 100
I propose another idea that I think can be quite easy to make.

Putting for example conditions of a contract to build a public building in the blockchain of nxt.

before the blocchain has reached a number of blocks (eg 15 days)

Offers must be entered in the blocchain

When the number reaches the agreed block bids are opened and the work is allocated.

Does Anybody dares to develop the idea?

I am interested to participate.   Smiley
full member
Activity: 154
Merit: 100
That's why WE need to start spreading the Nxt even in cities, in schools and in every place.



Let's create a Nxt virus and spread it via e-mail powerpointpresentations  Grin
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
there really isnt much point in this mechanism.  what you are describing is basically a rube goldberg contraption that produces the exact result of current NXT account/network forging

I would have to agree - and there is no point complicating further something we haven't seemingly even properly worked out (in terms of the modelling).
legendary
Activity: 1470
Merit: 1004
Just talked to bittrex, they have only 86 signups from nxt at the moment..

You should all go and sign up.

Quote
Go to https://bittrex.com/Account/Register and use SUPPORT_NXT as the invite code. This gets you past the invite code queue. Place some trades, enabled 2FA, or try any of our other features to help us stress the site before the final launch. If you find a bug, we're handing out bug bounties as well. Having a great exchange helps us but also helps the NXT community.


Done. Testing trading some DOGE until they add NXT.  Grin

Received this from Bittrex, please don't bother signing up any more until we have clarification they can add Nxt.

"Sigh, we just hit a huge technical hurdle that is going to delay this;  I'm not sure we have an easy way around this without doing a lot of core engine restructuring which is super scary.  

I will keep you updated Sad"


I've asked them for more specifics.

Just received this from Bittrex

Never mind... After hanging out 3 hours last night with some folks in your IRC channel, crisis adverted I think.

Thanks to those that helped last night.
full member
Activity: 221
Merit: 100

Greed could be the fatal flaw, if that 50M accounts persist to hoard we are all doomed, simply because that accounts don’t need a higher price to sell, they get rich anyway, no matter what the price is.


I really don't see it as a problem. Even if 500 million NXT don't move, the other 500 million would still be in circulation. That's more than enough. Bitcoin only has 12 million so far  and how many are in circulation? Very low number.

Hoarders can't "kill" nxt. If anything, hoarding pump prices up as less coins are in circulation.
We are in war with Ethereum (for those who have not understood it yet).

We need them engaged!

They must be insane if they don't get engaged now. If I were one of them I would have quit my job to fully work for Nxt.
full member
Activity: 168
Merit: 100
I don't get your idea, or I get it. I don't know.

Instead of leasing forge power, you commit with your account to share fees among others in the same "share fee group" if you forge a node. You still try to forge a block on your own, but you commit to share the incentive with others if you are successful (with special conditions like committing to run the node for some time, ...). Could this be done with AT and would that make sense?

there really isnt much point in this mechanism.  what you are describing is basically a rube goldberg contraption that produces the exact result of current NXT account/network forging

The idea is that all forgers repeatedly get small incentives for trying to forge. It is a pure psychological reason.

In the end, the fee distribution is the same.

exact!

But it is very important to keep alive the interest of forging every day because we want two things;

every day we must keep network secure
every day we must keep maximum decentralized network


That's why WE need to start spreading the Nxt even in cities, in schools and in every place.


it is true every time I like more Nxt ecosystem and community spirit.

 Smiley Wink
sr. member
Activity: 376
Merit: 300
People have been wondering about some of the figures that my simulator has come up with so far - but there is actually a much worse picture here than I have so far shown as my simulator has been using "log(balance)" rather than just "balance" as the "stake multiplier".

So get ready for something that approaches more closely how Nxt actually works as it does not use log(balance):

blocks = 1440
a: 10
b: 10
c: 10
d: 10
e: 10
f: 50
wins( a ) = 48
wins( b ) = 50
wins( c ) = 50
wins( d ) = 41
wins( e ) = 52
wins( f ) = 1199
best_streak( a ) = 2
best_streak( b ) = 2
best_streak( c ) = 2
best_streak( d ) = 2
best_streak( e ) = 2
best_streak( f ) = 49
best_combined_streak = 3

In particular pay attention to the figure for "f" in regards to both "wins" and its "best streak".

this I don't understand at all - what exactly was your algorithm for calculating weights?
sr. member
Activity: 376
Merit: 300
That's not how the nxt forging algorithm works.
You are looking for the maximum of random number * balance to get the winner where you really should calculate the minimum of random number / balance and let the minimum win.

I think that how you choose the winner is irrelevant - it is only the probabilities that matter here.

I think the rule of choosing the winner is relevant - it affects the probability distributions...

Also, could you run your simulations (with the correction of BloodyRookie) in the situation when the rich guy has around 0.35 of all NXT and there are around 10 poor guys with equal stakes? Some quick computations that I've just did indicate that the ROI for the rich guy would be maximal in this case...
sr. member
Activity: 376
Merit: 300
Here is the source code to my simulator:

Code:
#ifndef USE_LOG
         weights[ i ] = ( rand( ) % 1000 ) * balances[ i ];
#else
         weights[ i ] = ( rand( ) % 1000 ) * log( ( double )balances[ i ] );
#endif
         total_weight += weights[ i ];
      }
...
      for( size_t i = 0; i < balances.size( ); i++ )
      {
         double adjusted_weight = weights[ i ] * 1000 / total_weight;

         if( adjusted_weight > best_target )
         {
            winner = i;
            best_target = adjusted_weight;
         }
      }


That's not how the nxt forging algorithm works.
You are looking for the maximum of random number * balance to get the winner where you really should calculate the minimum of random number / balance and let the minimum win.


Yes, BloodyRookie is right (according to my today's conversation with CfB).
legendary
Activity: 1470
Merit: 1004
Version 0.3.4 of Nxs is how up on GitHub

All API / UI server requests are now done through POST (with a config option for GET for backwards compatibility with Nxt versions prior to 0.8 ).

Nxs now has a sidebar for navigation (similar to the nxtra.org NXT Wallet's) because the advanced Nxt features such as Messages, Aliases, Asset Exchange, etc. won't fit into the navigation bar at the top. Aliases have now been implemented and can be registered and updated on the Aliases page. If you choose "Nxt Account" as your type, the Nxt account number URI will automatically be wrapped with "acct:" and "@nxt" so the full URI is something like "acct:0000000000000000@nxt". This is an attempt to help push a standard for Nxt account number alias URIs in solidarity with wesley's NXT Wallet app. In the future, these account numbers will be able to be entered as the recipient in a transaction as well.

Also, aliases now show up in their registering and updating transactions on the Transactions page.

Along with the new side bar, the pages interface has been completely revamped to show only one page at a time instead of the old navigation system where you could toggle widgets on and off and display multiple widgets at a time. The initial version with toggling widgets made too much of an attempt to emulate Nxt's default NRS GUI instead of rethinking the interface altogether.

Rate of requests for new data is now adjustable in "assets/js/config.js" allowing Nxs to potentially run more smoothly on older computers. Repo available here along with installation instructions and the full change log can be found here. New screenshots can be found here.
]

Anyone tried this?  Looks awesome, great work Regulus.
legendary
Activity: 1367
Merit: 1000
Very strange. On 0.8.xe versions after several hours my node always forks. That was not the case on 0.7.x. I usually forged >1 block a day but on 0.8 i got nothing  Cry
Downgrading.

mine doesnt fork, but every few hours my 0.8.1e forging NRS node just hangs on a block and stops receiving new blocks.  I upgraded to 0.8.2e, same thing.

May be it is one case. The forks i was getting were forged mostly with my account with very high target in blocks. So the reason may be the same - loosing connection with the network.
hero member
Activity: 600
Merit: 500
Nxt-kit developer
I run public node on Android TV stick
Expect some posts like

Quote
Why my node on Panasonic microwave oven forges so little blocks? Should I upgrade it to Zanussi?

 Grin Grin Grin
legendary
Activity: 1778
Merit: 1043
#Free market
I don't get your idea, or I get it. I don't know.

Instead of leasing forge power, you commit with your account to share fees among others in the same "share fee group" if you forge a node. You still try to forge a block on your own, but you commit to share the incentive with others if you are successful (with special conditions like committing to run the node for some time, ...). Could this be done with AT and would that make sense?

there really isnt much point in this mechanism.  what you are describing is basically a rube goldberg contraption that produces the exact result of current NXT account/network forging

The idea is that all forgers repeatedly get small incentives for trying to forge. It is a pure psychological reason.

In the end, the fee distribution is the same.

exact!

But it is very important to keep alive the interest of forging every day because we want two things;

every day we must keep network secure
every day we must keep maximum decentralized network


That's why WE need to start spreading the Nxt even in cities, in schools and in every place.

full member
Activity: 168
Merit: 100
I don't get your idea, or I get it. I don't know.

Instead of leasing forge power, you commit with your account to share fees among others in the same "share fee group" if you forge a node. You still try to forge a block on your own, but you commit to share the incentive with others if you are successful (with special conditions like committing to run the node for some time, ...). Could this be done with AT and would that make sense?

there really isnt much point in this mechanism.  what you are describing is basically a rube goldberg contraption that produces the exact result of current NXT account/network forging

The idea is that all forgers repeatedly get small incentives for trying to forge. It is a pure psychological reason.

In the end, the fee distribution is the same.

exact!

But it is very important to keep alive the interest of forging every day because we want two things;

every day we must keep network secure
every day we must keep maximum decentralized network
jr. member
Activity: 59
Merit: 10
i work on a bigger project for the english an german community.. if someone can support it with a donation... please pm me..
i will show u on what iam working


casx


legendary
Activity: 1162
Merit: 1005
Still testing 0.8.1e. Noticed some small bug. I run public node on Android TV stick and private forgoing node at home laptop. I'm connected to my public node with other PC. So, sometimes I notice, that in public node there stays unconfirmed transactions, which does not exists in private forging node. After refresh of public node, everything ok, unconfirmed transactions list became the same.
Jump to: