Show Posts - jones  
Please login or register.

Login with username, password and session length
Advanced search  


Latest Stable Nxt Client: Nxt 1.12.2

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - jones

Pages: [1] 2 3
Nxt General Discussion / [video] Explaining Nothing At Stake
« on: August 12, 2015, 10:37:25 am »
I tried to very briefly, (4 minutes), go over the basics of proof of stake blockchains and how the nothing at stake attack works, and what can be done about it.


Thank you to Kushti and the Consensus Research team for their work, this is based off of the reading of many of their papers.

I hope everything is covered in the correct way and I didn't misspeak, get anything wrong, or leave anything out, if you notice any errors, please notify me and I will make some edits.

General / getAccountTransactions debacle
« on: August 06, 2015, 03:39:51 am »
So I have been trying to get updated with the new things in the Nxt API, though I'm still confused mostly with the deprication of getAccountTransactions

so I see that now we are supposed to use getBlockchainTransactions, but we don't have getBlockchainTransactionIds which we have had in the past, any particular reason for this?

also does the blockchainTransactions imply that this request only returns transactions that aren't being phased at the moment? so we have to getPhasedTransactions to get the currently phased transactions along with blockchainTransaction to get all of the current active transactions?

--- new content

the website is now live and I added an introductory video, its a thing now :)


Now I need your help to spread this event to as wide an area as we can, try to get some developers interested, do those social media things I hear so much about.

Thanks all


I've been thinking about this for a few days now and I think it could work very well for nxt.

Generally I believe that the two things that Nxt could benefit most from right now are more developers, and a bit more exposure and recognition as a platform for building things on. This idea manages to do both in a fairly cost effective way.

Basically the idea is to round up a bunch of developers, and give them the nxt api, some guidance and inspiration (a slack-like chat to share ideas with maybe), and 48 or so hours to make something on top of nxt using whatever programming language they want to. Then the community tries out all of the entries and votes on the best, which wins some sort of prize, then a few other people win other prizes based on how good their projects were.

I would be willing to organize and run this event I just need a little help in two areas, one being a bit of prize money for the winners, and two being some help spreading the word and gathering developers who want to participate in the challenge.

I'm thinking somewhere around 10k/20k nxt for the winner and a few other k for runners up. I think it would be worth doing if were were able to attract around 10ish developers, though the more we can get the better.

Nxt Plugins / [PLG] Next Block Predictor Plugin
« on: July 01, 2015, 06:47:37 am »
A very simple box in the left corner of the screen that tells you how much time is left on the current block.

Download: http://jnxt.org/block_counter.zip

* I estimate that it is within a few seconds of being correct about 76ish% of the time, but sometimes it can be somewhat off, just due to how the calculations are made.

Decided to spend most of today refining my idea of using nxt tokens from a browser plugin to automatically log users into websites without needing the dreaded username+password combonation. Two liters of mountain dew later, it is now 2am and I am pretty satisfied with where it ended up for the day.


I put together a small demo of what it can do right now.

Remember this is just a messy demo type version, when it is included in the full jay v.2 it will be cleaner and do some other cool stuff.

Nxt Plugins / Code auditng?
« on: May 30, 2015, 02:46:24 pm »
Because plugins are not sandboxed, it would be wonderful if we could set up a sort of system for many independent developers to look over the source code inside plugins and sign off on it once they deem it safe.

This will help against malicious plugin creators as well as plugins that might have slight security issues in need of patching.

Personally, I think it would work best if the plugin announcement in this thread specifies an available bounty for trusted auditors to look over the source, the auditors then go through the plugin and search for vulnerabilites and sign off on it if there are no issues.

If an audior signs off and there and another one finds a vulnerability, the first auditor is not paid his share of the bounty, this is to avoid auditors from just oking everything instantly.

I've been meaning to do this for a while, and decided it was time when 1.5.9 came around.
I've set up an Nxt package on the archlinux user repository, basically this allows for the installing/updating and automatic checking of the sha256 hash with a single command (for me its 'packer -S nxt'), then all that's needed to start NRS is the 'nxt' command.


Not too many people use arch, so this is mainly for myself as it speeds up my ability to update to new NRS versions, but if anyone finds it useful, it is available.

Nxt General Discussion / Nxt Projects Updates Hub
« on: May 04, 2015, 03:52:20 am »
This thread is meant to be a place where all nxters with ongoing projects can post what's happening on their projects to gain better visibility.

There is so much going on in the Nxt space, but because we are just so decentralized, it's hard for an one person to be up do date on everything that's going on. Past that is is very hard to come up with any type of collective marketing for nxt as someone would need to constantly be asking many different people to make social network updates.

I believe this is a good solution because nxters generally hang out in general discussion here, and with a single thread, everyone wins.
- projects get an easy stream of publicity and a place to make updates.
- average people get to see what's going on with all these projects
- nxters running social network pages or news type websites can use these as posts or press releases.

1. Project name
2. Update description, less than 140 chars so people can tweet
3. A picture or two, optional but suggested
4. A longer press release type test blurb explaining in more detail the update and project.
5. Any other useful links or anything else

Any people that use social network services can also post links to versions of these updates on that network so that others mah rebroadcast the message to gain wider reach.

I use the term project very loosely, so pretty much anyone with any nxt related announcement is welcome to share here.

Also comments about updates are welcome as long as it is easy to distinguish updates and comments, update makes should probably use bold (please no smaragda-like all caps red typeface)

Jay Framework / The road to Jay ver. 2
« on: April 14, 2015, 03:28:23 am »
Jay has been out for a week and a half or so now, and I'm very happy with it so far.
With an article in cointelegraph, an asset being actively traded, and over 300 views for the intro video so far, things are going well.

At the moment, the big issue in my opinion is that there is only http://jnxt.org/wallet to use to test Jay, and it can do so much more than this. I will be releasing the AE side of JayEX within the end of this week, allowing for a Nxt AE trading with the security of the localhost NRS client, the inspiration of secureAE, and the chart ideas and layout idea from poloniex.
I also plan to have an index of all Nxt transactions on a single page using Jay for testing purposes as well as quick transactions with minimal ui. Along with this I have a vague idea for a wallet I'm calling chaos wallet that I might play around with in the next week or two.

In addition to the adding of different wallets to the Jay landscape, I also plan to push forward to Jay version 2 where the client and jay.js will have a few more features:
1. Handling of encrypted messaging and encrypted message to self
2. Implementation of the rest of the transactions in the Monetary System
3. Attempt to make bridge functions for MS to make trading on MS more like trading on AE
4. Add Multi-language support to the Jay Client and get translations for the client and documentation.
5. Add auto-updater to the Jay client to allow for easier upgrades of the system in the future
6. Add a checksum check off of official Jay alias 'jayversion' on Jay Client update to assure version downloaded is correct.
7. Add Voting system transaction types
8. Add Phased transaction appendage as well as the validate phase transaction
9. Include the P2H specification in 1.5 as well
10. Implement the new prunable message standard to be included in 1.5
11. Add anything else that comes up before the mainnet hard fork to 1.5
12. minor bug fixes, naming changes, and style/format changes to improve usability.

My hope is that Jay Version 2 will be released around the time that 1.5 hard fork occurs, but there is much to do.

Nxt Community News and Announcements / [ANN] Jay Framework Version 1
« on: April 02, 2015, 03:59:03 am »
As with the countdown that's been progressing all of this month, Jay version 1 is finally ready to go.

jay.html sha256: 19fecf2a296bbf730219fe84cc5fba1c7bceeaf54a4d85ea4eb48828438f6ac1

Jay is an Nxt web wallet that acts in two parts.
1. A small download that keeps your passwords held locally and safely
2. A library that allows for wallets to be made easily and hosted on any websites without worrying about security.

This release has the basic outline of the project with the Client, and most transaction types available.
Next version will include the rest of the MS transaction types as well as a way to send encrypted messages through the Jay Client easily.

Asset - Jay - 8688289798928624137

The asset represents the Jay Frameworks usage of SuperNET services,
For instance the asset gets 10% of the fees from an instantDEX trade that is facilitated on a Jay-based InstantDEX wallet.
Similar deals will be made for other services.
So once these wallets are made and start to generate fees, dividends will be made.
I expect these dividends to start a couple months from now, June-ish.

General / Why isn't referenced transaction full hash an appendage
« on: March 23, 2015, 02:24:50 am »
In transaction layouts, referenced transaction full hash takes up 32 bytes of each transaction, and is filled with zeros most of the time.

It would save a lot of space on the chain to move this to an appendage, we have 4 bytes of flags to be used for flagging appendages, and we would save 23 bytes on each transaction.

The only reason I can think of for keeping this large chunk of data in each transaction is either that it was made before we had appendages and was just never transferred.
Or the more interesting that this feature will be more used in the future, for TF/EC/other and it is kept in for now for convenience.

Just some thoughts I have while digging through all of the transaction types and formats.

Also if anyone knows what a hub does, or will do, or why its a transaction type currently. (I'll just look through the source for this one).

Jay Framework / A Framework example
« on: March 23, 2015, 12:26:38 am »
I was playing around with Jay Framework today and get SendMoney to print out a TX_ string to use with Jay client.

So I ended up making a basic client with 13 lines of code:

Code: [Select]
<script src="util/jquery.js"></script>
<script src="js/jay.js"></script>
$("document").ready(function() {
    $("button").click(function() {
        $("span").text((new Jay()).sendMoney($("#acc").val(), $("#amt").val()));
amount: <input id="amt" type="text"> <br/>
recipient: <input id="acc" type="text">
<button type="button">Create Tx</button><br/>

Just an example of what will be possible with the Jay Framework, 10 days left :)

Nxt General Discussion / interesting service idea, fee cutting
« on: March 12, 2015, 06:04:00 am »
Lets say I'm a whale with 10 million nxt, odds say I will forge every 100th block or so.

I create a service that takes asset issuance parameters along with 100 nxt into a wallet with a few thousand nxt

Then I use a forging predicter to wait until my 10 million will forge then broadcast the asset for the person. Now I have put 900 into the 1000 of an asset issuance, and got the 100 from the person as profit.

Then anyone who is willing to wait a few hours can make assets for 10x less and everyone still profits.

General / nxtlib-php, five functions to allow easy nxt integration
« on: February 14, 2015, 03:35:49 am »
As suggested by Damelon, and required by another project I want to work on, I have created this library.


It is split up into three different classes -

class Curve25519:
There wasn't an implementation for a standalone curve25519 ECDH in PHP, so I ended up having to make my own :) ported from the js version in the nxt client

class Converters:
Conversion functions for changing between data types and creating public and private keys to use with curve25519

class Nxtlib:
Where the fun stuff happens, contains the objects curve, for curve25519, and convert, for converters, as well as some additional necesarry functions.
To initialize you need to include the ip of the node you want to connect to.
functions of Nxtlib:
- request($requestType, $params)
- signBytes($message, $secretPhrase)
- verifyBytes($signature, $message, $publicKey)
- generateToken($websiteString, $secretPhrase)
- parseToken($tokenString, $website)

with node requests, signing and verifying bytes with curve25519, and generating and parsing tokens, nxt can be integrated into a web service fairly easily.
Licensed with MIT licence, use however you'd like :)

Nxt General Discussion / Nxt Vapor Wallet - Cold storage with style
« on: February 12, 2015, 04:55:27 am »
Three days ago I had an idea for an offline wallet for generating transaction bytes then broadcasting them within a single downloadable html file.

After the first proof of concept came out it was suggested that I add some features to make it usable for cold storage.

It can now:
- make normal payments
- do public key announcements
- lease effective balance transactions
- generate passphrases from user input and randomly
- stay under 400 kb in one downloadable html file
- do all of this offline
-  broadcast transaction bytes within the wallet, or copy paste them in cases of offline mode.


There might be a few errors still, (aren't there always) but it passed all of my sanity checks thus far. :)

overall it turned out to be a good proof of concept, and it gave me more good ideas to work on this weekend

Community Funding Committee (CFC) / proposals?
« on: February 08, 2015, 03:44:39 am »
When will it be possible to propose something with this or to somehow ask for funding?

I have a good idea for something that the CFC would work well for :)

Nxt General Discussion / [VIDEO] A walk through nxts voting system
« on: January 30, 2015, 02:25:11 pm »

My first attempt at screencasting describing all the different parts of the voting system, what it does, and what you can do with it.

General / an idea - tx fees as a proof of work
« on: January 18, 2015, 01:00:42 am »
I think CfB mentioned something like this somewhere, so I got to thinking about the implications.

from BCNexts prospective fees should be used only as a spam prevention, not for network security. For this reason, it would be better for the end user if the spam prevention didn't require any nxt spent at all.

each transaction would require a PoW puzzle to be completed for the TX to be considered valid and be included into the system. The difficulty could scale over time to keep up faster computers, the puzzle would need to be just as difficult on gpu and asics, so something ram intensive might work.

this system would make it so no nxt fees wild be rewarded to the forger of a block, which most believe would make nxts consensus easy to be attacked because there's no incentive to forge than to secure your incentive in the currency that you have a large stake in.

it is an interesting concept in my opinion, no fees could do some cool stuff.

I've done a good amount of research on what the best parameters are for the min difficulty and max difficulty, this will walk you through which one to pick.

First of all we can say that the range for difficulties is between 1 and 255
1 being a 1/2 chance of minting per hash
255 making it necessary to break the algorithm used in order to forge

To calculate the number of hashes needed to mint a currency we can use:
2^256/((2^(256-min-(max-min)*(minted/max_mintable))/unitsPerMint) = hashes

To solve for hashes needed: (for calculating how many hashes a coin will take to mint)
hashes = 2^diff*10^decimals*unitspermint

lets say I'm minting an MS currency that has a diff of 16 (EGOLD)
hashes = 2^16*4^10*1 = 65000*10000*1 = 650'000'000 hashes
650 Mhashes, about 5 minutes for my machine (2Mhashes per second)

To solve for difficulty:
difficulty = log2(hashes/(10^decimals))

for a 1 Ghash per mint currency (4 decimals):
difficulty = log2(1'000'000'000/10'000) = log2(100'000) = 16.6 -> 17
so that would have a difficulty of 17.


"FORKK it"

Built from the ramblings of a discussion on SN slack #general

this next gen crypto technology is available to all with nothing better to do.

- 10000 total FORKK (4 decimals)
- diff from 4 -> 104 (really easy -> really hard)
- diff adjustment every 100 FORK minted
- max per mint of 1 (now you don't need to stress about how many units per transaction)
- no premine (even the developers of FORKK didn't want to be burdened by its existence)
- backed by pure speculation (it will get more difficult to obtain eventually, that makes it worth something, right?)

Be a part of tomorrow, mint FORKK today :)

Pages: [1] 2 3