Topic: NXT :: descendant of Bitcoin - Updated Information - page 1008. (Read 2761642 times)

Funny how we are taking about plugins (more security issues) when one fatal flaw  that existed in all versions prior to 0.6 could have completely ended Nxt experiment

https://nextcoin.org/index.php/topic,3884.0.html

Had this flaw been discovered by someone more nefarious, that would have been pretty much the end (40 million Nxt stolen for bter, for example).  

very nice just like building blocks....

I am not sure how it is possible to implement a decentralized filesharing or website configuration.
Remember that there will be hundreds of nodes, any of which could end up forging the next block. The one that forges the block would scan all AM's in that block and if it finds a plugin invoked, it will make sure it hasnt been tampered with and then will invoke it. There also needs to be a way to add this event to the blockchain with some sort of transaction id that can be verified by all the other nodes.

So all the data for the plugin needs to fit in an AM. I guess I need to clarify that.

Sorry about not being clear enough. Since I wasnt clear enough, I will authorize 500 NXT for you, but following submissions need to be something that fits into the NXTplugin model. All the input needs to be available in the blockchain (including AM) and there also needs to be a way for other nodes to verify that the action has been done, eg the one way hash of email content and destination email address.

James

Edit: I think my usage of plugin is what was a bit confusing. I am talking only about NXTplugins, eg. external actions that can be invoked by NXT VM (turing scripts) from the AM that it outputs.

25000 NXT bounty for a method and Java implementation that verifies NXTplugin process has not changed. Need an OS independent way of finding the executable code of a previously registered NXTplugin. This means we can constrain the creation method (linker output), probably need to do this for unix, Mac and Windows separately.

I am looking for a practical solution that will allow realtime verification by the NXTcore to make sure that the NXTplugin has not been tampered with. NXTplugins will have to be opensourced and publish signatures for specific compilers. This signature is then verified prior to any usage of that plugin by the NXTcore.

BEFORE we would ever consider submitting this to jean-luc, we of course need to test it like crazy. If the code is changed at all, we assume it is tampered. This probably means we cant do any dynamic linking, and need either static or relative jumps. Not sure though. Just finding where the code is might not be so easy. Figuring out how to get a ptr to the Java process will probably be pretty difficult. Any reasonable one way hash function is fine for this bounty, just want to get the system issues out of the way so we can validate in realtime that a plugin has not been tampered with.

As long as the source is reviewed for Evil Bobness and the code that is executing generates the same signature, I think we are getting close to where we can trust it almost as much as a hardcoded plugin where the plugin is actually part of the NXT core. Once we have the ability to have NXT plugins that are external to the NXT core, that is when things can really take off. We still need a formal validation process before it is approved for inclusion in the approved list of plugins, but maybe we can sidestep that issue by just having web.xml entries?

PLEASE if anybody can find a security flaw in this method, post ASAP. Remember Evil Bob is very evil

James

+1440
It shows that Nxt is not just a coin, but crypto-PLATFORM !

Are there any mining services that let you convert your newly mined coin directly to NXT instead of BTC?

I am currently studying some useful apps, so here is my attempt of Nxt plugin
Name: Nxt File Sharing

Description:
The File Shares (Windows) plugin uses the popular Samba (SMB) daemon to host files on a local network. These files can be accessed from any computer in the network using the network name of the device hosting it and the applicable share name. This plugin is used to coordinate filesharing with Windows-based devices on your network.
https://wiki.arkos.io/wiki/File_Shares_(Windows)

Source:
https://arkos.io/downloads/

+ many other plugins:

https://arkos.io/genesis/plugins

Websites
- The Websites app allows you to add, remove and edit websites to your arkOS server. It automatically downloads the configuration data for each app type and configures them in the background.

Databases
- The Databases app allows you to manage a variety of databases depending on the types that are installed on your system. These databases can be used for individual plugins or webapps that you use with Genesis. Different database types can be installed by going to the Applications pane from the Settings menu, then downloading and installing them as you would another plugin.

Peer of NXTio.org is Rank 11 in https://www.peerexplorer.com/index.php

https://nextcoin.org is down again

EDIT: https://nextcoin.org comeback!

I found an SMTP server in Java: https://code.google.com/p/subethasmtp/

I want to offer a bounty of 20000 NXT to modify it to become a hardcoded NXT plugin proof of concept, NXTsmtp

The goal is to get this to where we can submit it to jean-luc for inclusion in an upcoming release, without making him yell at us for sending him junk

For now, there is too much that is undefined, so I just want a few simple things. Need a way to make its input be contained within an AM. For now we can limit to 1000 byte message + destination. The idea is that at some point in the future, there will be NXT VM scripts that output an AM which should trigger sending an email.

A reasonable format needs to be defined for the AM, to start with I am thinking a code for plugin ID, followed by size, followed by plugin specific data. Since we havent defined the layers yet, any well defined format is fine. We will just tweak it when we get better format definitions. A way to create the AM from text file is also needed and this code is intended to be included in clients, eg. read text file, format AM data, submit AM.

The second part is to figure out where in the NXT core we need to put a check for AM triggered plugin invocation. This code also needs to get a transaction ID (which would be the hash of content and email address) and put it into the blockchain so all the nodes can verify it was sent from the forging node.

I am not sure if subethasmtp is a totally self-contained smtp implementation, if it is not, we cant use it for hardcoded NXTplugin. If it turns out that the subethasmtp is not suitable or there is a better one, then please find and use the better one. I am not sure how many people want a decentralized way of sending email, especially when any node can read it. This is my attempt at getting a reference implementation for a hardcoded plugin. No need to worry about encrypting/decrypting AM that contains email.

James

P.S. I also authorize 500 NXT bounty for the first 10 NXTplugin ideas with corresponding source code that does anything remotely useful, eg. you can think of a situation where you would actually use it. Submissions need a description of the idea and when you would use it and URL to source.

And we have our first entry @ 10,007nxt!  

  For those still forging but have a higher balance, you still have a chance, but you have to lower your balance to less than our current high(low)score in order to be eligible, simply transfer some nxt out to another account of yours, or do some good deeds like donations!

All,

People that have running projects other than marketing please let us know what your project is.

We want to have a view on all projects and consider bounty for them.

https://bitcointalk.org/index.php?topic=453935.new#new


for marketing projects please go here:


https://bitcointalksearch.org/topic/m.4465984

I really think 0.1 is perfect. we need some discouragement from people spamming the blockchain but at this point forgers should be doing it because they care about the network. not because they are making some spare change.

Winter Olympics in Sochi Russia leave a circle for  Nxt .
https://www.dropbox.com/s/8hxexcdwfpwkcb4/winter.jpg

That looks cool! Nice work!
[/quote]

Thanks guys!

You can send 100 Nxt as tx fee when you do transaction.

As we all know, BTCChina boosted the tx volume after they ignored tx fee. So you know it that the decrease of Nxt tx fee will also boost more and more application and tx volume, and I think the forger won't get too little.

It's nexern

WHO OPERATES NXT BLOCKCHAIN EXPLORER?
If you know him, please send him a pm.

http://87.230.14.1/nxt/nxt.cgi?action=10

It is still on version 0.5.11, update it please.

For anybody that thinks that NXT VM is an optional thing, please note OT (Open Transactions) will be working closely with (literally sharing offices?) etherium.

Also
****
Stage 1 is to develop the protocol and refine the contract, fee and security model. Stage 2 is to build the reference client. Stage 3 is to develop the application ecosystem. I would love to see gambling, OT, Ripple, voting systems, wall street contracts and trading system as well as DAOs living on our blockchain and in our app store at launch. I'd also love to see an online game using Ether as an ingame currency.
{Charles}
permalinkparent
[–]Timmerish 2 points 3 hours ago
Are you saying that Ethereum itself will build those applications (gambling, OT, Ripple, etc)? Or that you're hoping other people will build them?
Also, are you saying Ethereum itself will build an app store? What will that look like?
permalinkparent
[–]MihaiAlisieEthereum core team 4 points 2 hours ago
The (D)app store will be like a plugin module/function for the reference clients. Just think of clicking a button, and then looking through a selection of Dapps that can extend the functionality of your wallet to pretty much anything from encrypted chat via Bitmessage to drag-n-drop decentralized file storage. Easy.
permalinkparent
[–]ethereumcharlesEthereum core team 4 points 2 hours ago
We are building the app store and it is inspired from the chrome and android stores. We would love to partner with the bittorrent guys on some things for that store. In terms of who builds the apps, some we make ourselves, some our partners make and some organically come.
****

It is mission critical for us to make sure NXTlayers are implemented, preferably with a NXT plugins solution

James

Thanks for link it led to "indistinguishability obfuscator"
It led to http://www.wired.com/wiredscience/2014/02/cryptography-breakthrough/

********** key excerpt
Sahai and Waters proceeded to show that their indistinguishability obfuscator seems to offer much of the all-encompassing cryptographic protection that a black box obfuscator would offer. It can be used, for example, to create public key encryption, digital signatures (which enable a website to convince its visitors that it is legitimate) and a laundry list of other fundamental cryptographic protocols, including two major ones that were previously unsolved, functional encryption and deniable encryption.

The team’s obfuscator works by transforming a computer program into what Sahai calls a “multilinear jigsaw puzzle.” Each piece of the program gets obfuscated by mixing in random elements that are carefully chosen so that if you run the garbled program in the intended way, the randomness cancels out and the pieces fit together to compute the correct output. But if you try to do anything else with the program, the randomness makes each individual puzzle piece look meaningless.

This obfuscation scheme is unbreakable, the team showed, provided that a certain newfangled problem about lattices is as hard to solve as the team thinks it is. Time will tell if this assumption is warranted, but the scheme has already resisted several attempts to crack it, and Sahai, Barak and Garg, together with Yael Tauman Kalai of Microsoft Research New England and Omer Paneth of Boston University, have proved that the most natural types of attacks on the system are guaranteed to fail. And the hard lattice problem, though new, is closely related to a family of hard problems that have stood up to testing and are used in practical encryption schemes.

******

If I understand it correctly, this means that NXTplugins can indeed be implemented in such a way that it can be trusted even though it runs in a different process from NXT core. I dont claim to understand any of the maths, but it does sound very promising that long term there will be such solutions available.

Even if we get less advanced NXTplugin architecture in the short run, we just need to limit its usage to real world actions that wouldnt be disastrous if it didnt happen or went wrong, eg. making posts on website.

James

HAS ANYBODY HERE FORGED any blocks yet? If you have you may be eligible to win 1,000nxt

Head over to this thread here for more info: https://forums.nxtcrypto.org/viewtopic.php?f=2&t=744&p=3548