Qora | 100% POS | Assets | Names | Voting | Open Source - page 30.

Djinou94

legendary

Activity: 1036

Merit: 1000

Quote from: CIYAM on February 18, 2015, 07:41:31 AM

We are hoping to get a Bitcoin clone to implement AT (it has been designed in a way that it should be able to work with Bitcoin's Script engine with just one new op code being required).

If this is achieved then AT would end up being the "standard" method to do such ACCT transfers and a marketplace could evolve in each supporting blockchain.

Original Bitcoin Cool

CIYAM

legendary

Activity: 1890

Merit: 1086

Ian Knowles - CIYAM Lead Developer

Quote from: calvinstm on February 18, 2015, 08:15:59 AM

But before that can happen, I'm sure we need to get a smaller coin onboard first (to show that we have a viable and stable working system).

This is our main focus for now - improving AT (there is now both an assembler and a debugger) and getting ACCT to work across two blockchains.

calvinstm

full member

Activity: 229

Merit: 101

Quote from: CIYAM on February 18, 2015, 07:41:31 AM

If this is achieved then AT would end up being the "standard" method to do such ACCT transfers and a marketplace could evolve in each supporting blockchain.

Wow!! Very encouraging news! It'd certainly be awesome to see AT becomes the de facto standard for ACCT transfers!

Do we have a candidate already?

It will be a major win for AT system if we can convince one of the two big guys (Litecoin or Dodgecoin) to incorporate AT into its core and serve as base currency in the AT system. Grin

But before that can happen, I'm sure we need to get a smaller coin onboard first (to show that we have a viable and stable working system).

Happy with the direction AT is going! Grin

Glad to have you here, CIYAM!

CIYAM

legendary

Activity: 1890

Merit: 1086

Ian Knowles - CIYAM Lead Developer

We are hoping to get a Bitcoin clone to implement AT (it has been designed in a way that it should be able to work with Bitcoin's Script engine with just one new op code being required).

If this is achieved then AT would end up being the "standard" method to do such ACCT transfers and a marketplace could evolve in each supporting blockchain.

calvinstm

full member

Activity: 229

Merit: 101

Quote from: CIYAM on February 18, 2015, 07:03:23 AM

Quote from: calvinstm on February 18, 2015, 06:39:03 AM

So a fully decentralized bitcoin-based exchange is certainly feasible, don't you think?

The problem is basically achieving ACCT with Bitcoin - the technique that forum member TierNolan outlined is what the ACCT AT was based upon but it relies upon nLockTime which is "non-standard" (until after the time has been reached) so although possible it would be a bit clumsy to do with Bitcoin.

With Qora to Burst it will have a simple UI and work without requiring trust or any centralised website.

I understand having an intermediary coin is also a clumsy way to solve the problem, but I just thought it might be easier than doing a real ACCT with Bitcoin.

I don't know.

It seems like we're having a race to see which group will come up with a working decentralized exchange first (SuperNet, MGW, InstantDEX come to mind)..

I really hope to see AT becomes the dominant player.

calvinstm

full member

Activity: 229

Merit: 101

Speaking of GUI, I asked Qora Dev a question concerning GUI a long time ago (when we had weekly 'features' votes... One of the things was "Wallet Upgrades")... He told me GUI would eventually go OS. But looks like he forgot to OS certain things before he left..

Quote from: qora on June 28, 2014, 02:18:30 PM

Quote from: calvinstm

Hi Qora Dev,

I have an idea for the wallet upgrade option today. I was thinking, what if we have a custom plugins and skin/theme system for Qora wallet? Grin

So 3rd party developer can create different plugins to expand the features of Qora wallet (and artists can create custom themes). To accomplish that, I'm thinking you will probably need a scripting engine (maybe based on Lua). Wink

For instance, there is a folder called "Plugins" and another folder called "Theme" inside wallet folder. At the startup, the wallet will load all the plugins and skin/theme inside those folders.

Do you think it's something that can be added to the Wallet Upgrade option in voting list? Cheesy

If you don't like my idea, just ignore it. No worries. Wink

Another thing, could you tell me how much time (approximate) is needed to do wallet upgrades? I think it should take no more 1 week. Am I wrong in assuming this?

Thank you!

PS: You're the best!

This could be done easily when the GUI is opensource.
It already uses a system to change to the OS theme.

CIYAM

legendary

Activity: 1890

Merit: 1086

Ian Knowles - CIYAM Lead Developer

Quote from: calvinstm on February 18, 2015, 06:39:03 AM

So a fully decentralized bitcoin-based exchange is certainly feasible, don't you think?

The problem is basically achieving ACCT with Bitcoin - the technique that forum member TierNolan outlined is what the ACCT AT was based upon but it relies upon nLockTime which is "non-standard" (until after the time has been reached) so although possible it would be a bit clumsy to do with Bitcoin.

With Qora to Burst it will have a simple UI and work without requiring trust or any centralised website.

calvinstm

full member

Activity: 229

Merit: 101

Quote from: CIYAM on February 18, 2015, 06:35:46 AM

The account would be determined as a hash of the actual machine code for the AT (which might also take into account the account of the creator and perhaps some other metadata such as its description, etc. to ensure there should be very little chance that your AT create tx fails due to an existing AT with the same account id).

All initial data for an AT is part of the AT creation tx (so is of course stored in the blockchain) but subsequent state data does not need to be sent around nor stored in the blockchain (just the hash of it does).

AT is actually an extremely "lean and mean" design (and IMO the most efficient Turing complete solution on offer).

Thank you! I got it.. So nothing is being changed or stored at runtime (both data and code remain persistent). Even the account hash is determined based on an amalgam of data.

That's brilliant! Thanks for the explanation.

calvinstm

full member

Activity: 229

Merit: 101

So I just thought of this...

Let's say we have a portal/gateway that sells an AT-enabled token CIYAM-Bitcoin (which represents a Bitcoin IOU in the trading network) and set the exchange rate of Bitcoin to CIYAM-Bitcoin to 1:1. This portal is solely responsible for converting Bitcoin to AT-enable Bitcoin IOU (CIYAM-Bitcoin) and vice versa.

Doesn't that mean we will have the means to trade BITCOIN against any other AT-enabled cryptocurrency (Qora and Burst for starter)?

So a fully decentralized bitcoin-based exchange is certainly feasible, don't you think?

CIYAM

legendary

Activity: 1890

Merit: 1086

Ian Knowles - CIYAM Lead Developer

The account would be determined as a hash of the actual machine code for the AT (which might also take into account the account of the creator and perhaps some other metadata such as its description, etc. to ensure there should be very little chance that your AT create tx fails due to an existing AT with the same account id).

All initial data for an AT is part of the AT creation tx (so is of course stored in the blockchain) but subsequent state data does not need to be sent around nor stored in the blockchain (just the hash of it does).

AT is actually an extremely "lean and mean" design (and IMO the most efficient Turing complete solution on offer).

calvinstm

full member

Activity: 229

Merit: 101

Quote from: DeMap on February 18, 2015, 06:27:43 AM

Yes i am.
You are actually asking the right person the right question. Why are you asking? Huh

Talking about Arbitrary Messages can we please get rid of this Base58 encoding in API calls.
Look at the Qora Base58 implementation (@https://github.com/Qora/Qora/blob/master/Qora/src/qora/crypto/Base58.java ) and you will see it was never designed to process large amounts of Data like a 4kb Arbitrary Message.
In fact it's ridiculous slow and memory consuming. Can we please switch to HEX encoding for the next version?

LOL

Are you sure you're not Qora Dev in disguise??? Cheesy

I was starting to have my suspicion. Cheesy

There is no way you're a newbie with 4 posts.. lol

Funny you should mention it because I actually talked to Qora Dev about size limit (message size)... I cannot recall what he told me, but I am pretty sure it was never designed for large amount of data...

HEX is a good choice.. In fact, I think you should be Qora's lead dev.. I hereby nominate you! haha.

Think about that, friend.. Cheesy

Vrontis

sr. member

Activity: 351

Merit: 250

Quote from: DeMap on February 18, 2015, 06:27:43 AM

Quote from: calvinstm on February 18, 2015, 04:52:36 AM

Are you aware that we also have Arbitrary Transaction for uploading arbitrary bits of data to the blockchain? Wink

I wonder if we can use that to our advantage...

Yes i am.
You are actually asking the right person the right question. Why are you asking? Huh

Talking about Arbitrary Messages can we please get rid of this Base58 encoding in API calls.
Look at the Qora Base58 implementation (@https://github.com/Qora/Qora/blob/master/Qora/src/qora/crypto/Base58.java ) and you will see it was never designed to process large amounts of Data like a 4kb Arbitrary Message.
In fact it's ridiculous slow and memory consuming. Can we please switch to HEX encoding for the next version?

This has been applied by Qora and it will be followed for the next releases that the AT team will handle.
Qora rules will be followed.

DeMap

newbie

Activity: 9

Merit: 0

Quote from: calvinstm on February 18, 2015, 04:52:36 AM

Are you aware that we also have Arbitrary Transaction for uploading arbitrary bits of data to the blockchain? Wink

I wonder if we can use that to our advantage...

Yes i am.
You are actually asking the right person the right question. Why are you asking? Huh

Talking about Arbitrary Messages can we please get rid of this Base58 encoding in API calls.
Look at the Qora Base58 implementation (@https://github.com/Qora/Qora/blob/master/Qora/src/qora/crypto/Base58.java ) and you will see it was never designed to process large amounts of Data like a 4kb Arbitrary Message.
In fact it's ridiculous slow and memory consuming. Can we please switch to HEX encoding for the next version?

calvinstm

full member

Activity: 229

Merit: 101

Quote from: CIYAM on February 18, 2015, 05:29:57 AM

To clarify a couple of points - indeed AT stands for Automated Transactions (the "Turing complete" blockchain transaction solution created by CIYAM Developers which is detailed here: http://ciyam.org/at).

As ATs are stored in a blockchain (which is a public ledger) there is simply no way that it can store secrets, however, it can stored "hashes" of secrets which can later be revealed to match (this is the basis of the Atomic Cross-Chain Transfer AT).

Complete anonymity is never going to be possible with ATs but the technology would allow for "mixing" (of same amounts) that would make it impossible to connect inputs to outputs (but development of such an AT will be done after we have implemented and demonstrated ACCT between Qora and Burst).

One key feature of ATs is that they have their own account (so you can send them funds and messages and they can do the same). The AT machine code cannot be changed once the AT has been created so provided you (or someone that you trust) can vet that the AT does what it has been purported to do then it can act as a trustless 3rd party.

Thanks so much for the clarification and explanation! I'm gonna read up on AT's documentation so I'll have a better understanding of how it all works.

I have one quick question (sorry if it sounds ignorant), where is this initial hashed data (ie. account used by the program) being stored? Is it part of the AT transaction (resided in the same block as the AT machine code)? Or is there a secret storage space where those hashes are stored?

Thanks again for bringing AT to Qora!! Really can't wait to see the first ever Atomic Cross-Chain Transfer in action!

Congratz!

Vrontis

sr. member

Activity: 351

Merit: 250

Thanks CIYAM for clarifying, as we are getting close to AT release your explanations would became much more needed in order to let Qora members gain the advantage of AT and it's functionalities.

CIYAM

legendary

Activity: 1890

Merit: 1086

Ian Knowles - CIYAM Lead Developer

To clarify a couple of points - indeed AT stands for Automated Transactions (the "Turing complete" blockchain transaction solution created by CIYAM Developers which is detailed here: http://ciyam.org/at).

As ATs are stored in a blockchain (which is a public ledger) there is simply no way that it can store secrets, however, it can stored "hashes" of secrets which can later be revealed to match (this is the basis of the Atomic Cross-Chain Transfer AT).

Complete anonymity is never going to be possible with ATs but the technology would allow for "mixing" (of same amounts) that would make it impossible to connect inputs to outputs (but development of such an AT will be done after we have implemented and demonstrated ACCT between Qora and Burst).

One key feature of ATs is that they have their own account (so you can send them funds and messages and they can do the same). The AT machine code cannot be changed once the AT has been created so provided you (or someone that you trust) can vet that the AT does what it has been purported to do then it can act as a trustless 3rd party.

calvinstm

full member

Activity: 229

Merit: 101

Quote from: DeMap on February 17, 2015, 05:09:10 AM

What happend to the GUI source code? I can't find it on GitHub!

Speaking of GUI, I thought PondSea was working on a new GUI for Qora... What's up with that? Cheesy

Haven't heard any update in a while... It would be nice to see the new GUI design in the next wallet update..

A long time ago, I made a suggestion to qora dev that we should have a theme/skin system that works like a plugin (dynamic dlls). But he said something like... He had a groundbreaking idea for Qora wallet. And now he's gone... Cry

Does anyone have any idea what Qora Dev was talking about??? He never got to reveal his big secret plan for the wallet. How SAD is that?? Roll Eyes

Honestly wish Qora dev would contribute more since he's such a genius...

I admire him very very much. Cool

-----------

Haven't checked the source code but if use a decompiler like Salamander, it should give you some easy-to-read result... Qora has a very simple GUI, shouldn't be too hard to figure out calls.

calvinstm

full member

Activity: 229

Merit: 101

Quote from: DeMap on February 18, 2015, 04:37:11 AM

It's indeed unfortionate that ATs cannot have/process secrets data. For instance you could solve the cryptocurrency anonymity problem if it was different.
Probably somewhere down the road someone figures out a way for ATs to process secret data.

AT team pls think about it Wink

Are you aware that we also have Arbitrary Transaction for uploading arbitrary bits of data to the blockchain? Wink

I wonder if we can use that to our advantage...

Now thinking about it, I think it's going to be tough for AT to be able to process secret data since there is literally no secret on the blockchain. Tongue

Even if the data is encrypted, it can still be cracked.

Wonder how it can be done...

Anyhow, I just feel that a fully decentralized exchange is so close to being a reality.. Really wish someone find a way to get it done asap!

(Can't stop thinking about that since I got lots of coins stuck on BTER, really getting tired of using 3rd-party Exchanges for trading!!! Tongue

)

DeMap

newbie

Activity: 9

Merit: 0

Quote from: calvinstm on February 18, 2015, 04:13:54 AM

Oh!!! Thanks for the reply! I have to admit my knowledge of AT is limited. *embarassed* Embarrassed

I was thinking that if we could incorporate a bitcoin address generator and manage private keys inside an AT program on the blockchain, we might be able to pull something like a multi-cryptocurrency exchange off. Guess it's not possible then. Forgive my ignorance. Grin

It's indeed unfortionate that ATs cannot have/process secrets data. For instance you could solve the cryptocurrency anonymity problem if it was different.
Probably somewhere down the road someone figures out a way for ATs to process secret data.

AT team pls think about it Wink

calvinstm

full member

Activity: 229

Merit: 101

Quote from: DeMap on February 18, 2015, 04:06:31 AM

Unfortunately this doesn't work, because an AT cannot manage a bitcoin address. An AT cannot have secrets because everything it knows is on the Blockchain thus it cannot manage the private Key of a bitcoin address.

Oh!!! Thanks for the reply! I have to admit my knowledge of AT is limited. *embarassed* Embarrassed

I was thinking that if we could incorporate a bitcoin address generator and manage private keys inside an AT program on the blockchain, we might be able to pull something like a multi-cryptocurrency exchange off. Guess it's not possible then. Forgive my ignorance. Grin

Quote from: olcaytu2005 on February 18, 2015, 03:43:30 AM

AT is Atomic Transfer?

AT stands for Automated Transaction and is developed by CIYAM team. Think of it as a scripting engine in the Qora Core that is capable of interpreting codes/applications uploaded to the blockchain.

CIYAM team is going to develop some "use cases" for Qora after they're done with the new Wallet version.

I think the first application will be the atomic cross-chain trading between Burst and Qora (if I recall correct)

Topic: Qora | 100% POS | Assets | Names | Voting | Open Source - page 30. (Read 113833 times)