Topic: Closed Thread - page 133. (Read 677703 times)

Correct me if I'm wrong, but the point of a needing a balance for such things is to (only) mitigate the risk of a SA, is so the cost of such an operation is to expensive to be practical.

No system (that I'm aware of) is 100% protected against the outlier cases of a sybil.  I think, you could potentially monitor a few metrics and use them for calculating the value of a node.  If a very-healthy node is expensive to maintain (constant uptime, with restarts of servers being very brief, and a fast connection) you can mark the node as a "really good one".  If someone wants to cheat the system by creating an array of super-nodes, it should be expensive if the network of people and services using the network is large.  Also, if you have a very rigid process for validating blocks and TXs, you could slap heavy penalty "weights" on nodes so that can't forge for a LONG time.

I've been thinking about reputation in a trustless network in the back of my mind for a couple years but for a different type of application.  My gut is that there is an alogrithmic "sweet spot" where you can award healthy, but poor, nodes with the ability to mine blocks.  I'd have to get out the calculator and fudge some network statistics coz I don't have a massive network to test on, but I think this can be done.

The "problem" with my request is that a simple solution might not be simple to derive.  Using a balance as a means to determine who can forge is easy-- just look at a number on an account and then apply a little math to it.  However, statistically analyzing a graph-network, distributing consistent (verifiable) results and then deriving a meaningful output is difficult.  But, once you've found the magical formula, you're golden.

Sorry for this and my former post.  I get ideas and I post them a little faster than I should when I login here BEFORE I get my AM coffee.

😴

And how would you like to prevent a sybill attack with many "healthy" nodes without balance ?

I think I'm seeing something similar.  I have over 1%, forging away, on a node that has been running non-stop for over a week.  For a while I was getting about 25% of all blocks.  Now, I'vd not seen my account hit a block for days.

My account is the 9689... one.  At my heigth of generation, I was getting every other block.  Now... Nothing and I'm on a rock-solid system with a good network connection.

Instead of resetting counters, I suggest running statistical averages on each node and provide more blocks for the healthiest nodes.  I know changes are coming to the PoA algorithm.  I hope this is the approach you take.  I would.

Also, if you can derive a good metric to determine the health of a node, the. You can get rid of the need to monitor user balances.  Also, I think wouldn't even care if the account is sending/receiving funds.  If there is an account on the network, associated with a node that has 99.999% uptime and a dedicated fibre line, it deserves to hit blocks, even if it's balance is 0.

That's just my opinion.  Keep up the good work.  Thanks devs!

Hi to all, i have followed guido for installation of client in vps, it works some hours and after the block suddenly  stop of update and i need to restart  manually the service (evidently i lost all activity points), i´m using debian 7 64bits vps, the node version is 0.35 and i´m too using the latest linux version of nodecoin,

thank you in advance

Thank You!

Merry Christmas for devs!

thank you

Good decision. I'll help to test if necessary.

Yesterday I found a bug in the updated version and decided not to release the client in a real network, I'll be waiting when return staff in the office and we'll run the test network for testing as suggested earlier users of the community, it will be wiser. After stable operation of the test network we will carry on a real network. We changed the algorithm POA and I can't risk it on a real network, only after thorough testing and fixing bugs will be issued a new client.

When da new client ?

Hi, i would like know how is going nodebank development, and say than bounty thread is with little activity, i have bough a quarter of million and created a node, i like this project a lot, we need more often updates


thank you and i would love to help project

again, the node is long-term project.
See the OP.

I love Node! To the moooooon!

"4120890003132688172" - Node Address

I've written about it before

where is the roadmap?the price is too low

how about the China exchanges ,like OKCOIN,HUOBI,BTCCHINA?

"You can't see the wood for the trees."
Only too true.

Most people have no idea of anything like this nor do they care anything like this.
They just want to move money and to use their money as they like.
Which is the target? 100 people? 1000 people? or maybe tens or hundreds of thousands, millions of people?

Keep it simple.
Simple and effective.

Can you fill the questionnaire?

About the Node:

Operational - ? Months
Consensus Mechanism
Block Production TIme
Recommended Confirmation Time
Worst Case for Confirmation Time
Transaction Load
Anonymous Transactions
Transaction Fee Costs
Network Security Costs
Share Supply
Share Divisibility (precision)

Coin features: Yes/No/In Development/Planned:

Profit Sharing
Market Pegged Assets
User Issued Assets
Order Matching
User Registered Accounts
Messaging
Cross-Chain Trading Support
Multisignature
Smart Contracts
Hardware Wallet
Mobile Wallet
Decentralised Exchange
Pruning
Voting System
Instant Transactions
Automated Transactions
Mesh Networking
Account Control
Marketplace
Brain Wallet
Alias System
Encyrpted Blockchain Messaging
Merchant Tools
Inflation
Average Volume (mil USD)

lol http://www.wired.com/2015/01/bitstamp-offline/
and btc-e still the greatest exchange