- Mar 3, 2016
- 10
- 7
-- jl777 left update on iguanacore development on SuperNet channel #btcd ---
LTC and SYS now have corresponding genltc, ltc and gensys, sys files in SuperNET/iguana/coins dir. the convention is ./genCOIN is for running as a full node for that coin and ./COIN runs as a lite node. The default config will be BTCD as full node and all the others as lite nodes. this will allow COIN <-> COIN trading, using combined atomics and liquidity providers.
Still not happy with the networking side. the tradeoff is speed vs privacy, but I want something that is both fast and private. maybe there is a way.
sync times: LTC 20 minutes and SYS2 5 minutes, but those are not fully optimized times. however I think it is fast enough so I will leave it as is
basilisk tech works so well, I will port InstantDEX to use that instead of battle against bugs and performance issues in the current methods.
Given the choice of spending a lot of time debugging or porting to an improved engine, I like the latter approach better. At first the trading will be limited to simple pairs so we can find any issues with that. once it is validated, then we can activate combined trade sequences
Now if there is any doubt as to the value of iguanacore, keep in mind that I lost a day trying to get bitcoind to duplicate the iguanacore functionality. And failed
The only way I could make bitcoind do the things iguana does was to push the raw data from bitcoind through iguanacore...
iguanacore 1.6MB in codesize and syncs any chain in 1hour or less vs bitcoind that cant give you address balances for any non-wallet address. iguanacore can calculate balance for any address as of any height and I need that ability
I guess I could add insight blockexplorer or something like that, but I decided that trying to get bitcoind to be fast and nimble, well, easier to just directly add coin support to iguanacore. So I added LTC and SYS today. this found a few bugs and unimplemented things, like alert message and auxpow, but I think as more and more coins are supported, such missing things will become rarer and rarer.
the atomic swap statemachine sort of works, but everything has to be all in a happy state, so I am not feeling that it is robust enough and I went a bit crazy with basilisk and it supports arbitrary number of simultaneous remote transaction requests using a generalized metrics based system that verifies consensus (or >50% or whatever desired threshold) for each request
and the requests can be nested, ie a remote request can spawn other remote requests, recursively
now if you think about the power of this, you will see that it compares favorably with centralized (via blockchain) scripting systems. At least if you want to be able to run something on your computer that maintains decentralization and have many coins at the same time.
and all signing is done on your computer so it is as secure as near as I can tell with having a full node
What this means is that you will be able to do basic coin operations and statemachine driven tradebots securely for many coins on any caliber of computer. Maybe there will be some upper limit if you try to do 100+ coins at the same time, but I havent minimized resources used for the lite node operation so that is a temporary limit
iguanacore is the platform that provides the highest performance decentralized solutions, in the smallest footprint.
But it is not easy to understand and people dont invest a lot in something they dont understand, so it is easier to invest in "clone of something that got a big marketcap" or simply put money into any major new project as more than a quarter of them will do more than 4x, so it seems relatively good ROI
And my goal is not to make some pump type of thing, but rather to solve the scalability and mass market adoption issues. With basilisk it will be possible to even make a hosted website that allows people to see people with a litenode are seeing, just wont be able to do any money things. Well you could, but not securely so please dont. just install the liteclient, no need for big blockchains
One way to get the best of both worlds is to have a VPS that is running full iguana nodes and run a basilisk lite client locally. You can set your VPS to be your only peer and since you will be trusting your VPS, it is a reasonable setup and short of your VPS being breached, your privacy is pretty good
but there is no need to have your own VPS as you can rely on the iguana nodes. The basilisk nodes will be the liteclients, but to confuse you it will be possible for your computer to be a full node (iguana) for some coins and lite node (basilisk) for other coins
@sigwo: liquidity is nice for making a peg, but there are ways to achieve it without that much liquidity.
if you send me the parameters for INFX (based off of gensys or genltc) then I can see if iguanacore will parallel sync INFX
For those wondering what is special about BTCD if iguanacore is working with all the other coins, the answer is that I will only add BTCD staking to iguanacore and also whenever there needs to be a blockchain, ie posting of availability as LP, such notices will use BTCD. And I am working out a way to make BTCD transactions nearly as secure as BTC (if you wait for enough confirmations). PAX will be using BTCD and BTCD will be the preferred coin for fast clearing chips for pangea
I know it has been a long wait, but iguanacore is now doing things that nothing else does and my guess is that nothing else will be able to do it for a long time. With parallel sync solving most of the scalability issues and basilisk solving the market adoption issue, all that is left is to have mass market solutions. a DEX isnt exactly mass market outside of crypto, but it is the automated currency conversion engine that is needed to make all the crypto an equivalent store of value, ie if you can at anytime convert any crypto to any other crypto at low spreads and atomically, this enables many new solutions. combine fiat based pegs to this and all the infrastructure is there for making any sort of dapp
however there is no need to wait for some magical dapp dev to adopt the platform and create something useful, as I am already solving the DEX and poker, with the PAX doing fiat pegs. I am also preparing for adding privacy to all this, but that will happen toward the end as the exact transaction and dataflow clarifies.
I could do privatechains without too much work, but I dont really see the value for that as even with a private chain, there still needs to be a solution that uses it. I think a privacy enhanced anything to anything DEX is a more useful thing, but maybe there will be something that having a privatechain will solve
I came up with a way for people to test the DEX without needing any counterparty that is also testing, but it will require running a native iguana at first, so if you want to help the best thing is to get up to speed on installing native iguana. It depends on a single external module: pthread, so should be relatively easy to build on unix based systems
LTC and SYS now have corresponding genltc, ltc and gensys, sys files in SuperNET/iguana/coins dir. the convention is ./genCOIN is for running as a full node for that coin and ./COIN runs as a lite node. The default config will be BTCD as full node and all the others as lite nodes. this will allow COIN <-> COIN trading, using combined atomics and liquidity providers.
Still not happy with the networking side. the tradeoff is speed vs privacy, but I want something that is both fast and private. maybe there is a way.
sync times: LTC 20 minutes and SYS2 5 minutes, but those are not fully optimized times. however I think it is fast enough so I will leave it as is
basilisk tech works so well, I will port InstantDEX to use that instead of battle against bugs and performance issues in the current methods.
Given the choice of spending a lot of time debugging or porting to an improved engine, I like the latter approach better. At first the trading will be limited to simple pairs so we can find any issues with that. once it is validated, then we can activate combined trade sequences
Now if there is any doubt as to the value of iguanacore, keep in mind that I lost a day trying to get bitcoind to duplicate the iguanacore functionality. And failed
The only way I could make bitcoind do the things iguana does was to push the raw data from bitcoind through iguanacore...
iguanacore 1.6MB in codesize and syncs any chain in 1hour or less vs bitcoind that cant give you address balances for any non-wallet address. iguanacore can calculate balance for any address as of any height and I need that ability
I guess I could add insight blockexplorer or something like that, but I decided that trying to get bitcoind to be fast and nimble, well, easier to just directly add coin support to iguanacore. So I added LTC and SYS today. this found a few bugs and unimplemented things, like alert message and auxpow, but I think as more and more coins are supported, such missing things will become rarer and rarer.
the atomic swap statemachine sort of works, but everything has to be all in a happy state, so I am not feeling that it is robust enough and I went a bit crazy with basilisk and it supports arbitrary number of simultaneous remote transaction requests using a generalized metrics based system that verifies consensus (or >50% or whatever desired threshold) for each request
and the requests can be nested, ie a remote request can spawn other remote requests, recursively
now if you think about the power of this, you will see that it compares favorably with centralized (via blockchain) scripting systems. At least if you want to be able to run something on your computer that maintains decentralization and have many coins at the same time.
and all signing is done on your computer so it is as secure as near as I can tell with having a full node
What this means is that you will be able to do basic coin operations and statemachine driven tradebots securely for many coins on any caliber of computer. Maybe there will be some upper limit if you try to do 100+ coins at the same time, but I havent minimized resources used for the lite node operation so that is a temporary limit
iguanacore is the platform that provides the highest performance decentralized solutions, in the smallest footprint.
But it is not easy to understand and people dont invest a lot in something they dont understand, so it is easier to invest in "clone of something that got a big marketcap" or simply put money into any major new project as more than a quarter of them will do more than 4x, so it seems relatively good ROI
And my goal is not to make some pump type of thing, but rather to solve the scalability and mass market adoption issues. With basilisk it will be possible to even make a hosted website that allows people to see people with a litenode are seeing, just wont be able to do any money things. Well you could, but not securely so please dont. just install the liteclient, no need for big blockchains
One way to get the best of both worlds is to have a VPS that is running full iguana nodes and run a basilisk lite client locally. You can set your VPS to be your only peer and since you will be trusting your VPS, it is a reasonable setup and short of your VPS being breached, your privacy is pretty good
but there is no need to have your own VPS as you can rely on the iguana nodes. The basilisk nodes will be the liteclients, but to confuse you it will be possible for your computer to be a full node (iguana) for some coins and lite node (basilisk) for other coins
@sigwo: liquidity is nice for making a peg, but there are ways to achieve it without that much liquidity.
if you send me the parameters for INFX (based off of gensys or genltc) then I can see if iguanacore will parallel sync INFX
For those wondering what is special about BTCD if iguanacore is working with all the other coins, the answer is that I will only add BTCD staking to iguanacore and also whenever there needs to be a blockchain, ie posting of availability as LP, such notices will use BTCD. And I am working out a way to make BTCD transactions nearly as secure as BTC (if you wait for enough confirmations). PAX will be using BTCD and BTCD will be the preferred coin for fast clearing chips for pangea
I know it has been a long wait, but iguanacore is now doing things that nothing else does and my guess is that nothing else will be able to do it for a long time. With parallel sync solving most of the scalability issues and basilisk solving the market adoption issue, all that is left is to have mass market solutions. a DEX isnt exactly mass market outside of crypto, but it is the automated currency conversion engine that is needed to make all the crypto an equivalent store of value, ie if you can at anytime convert any crypto to any other crypto at low spreads and atomically, this enables many new solutions. combine fiat based pegs to this and all the infrastructure is there for making any sort of dapp
however there is no need to wait for some magical dapp dev to adopt the platform and create something useful, as I am already solving the DEX and poker, with the PAX doing fiat pegs. I am also preparing for adding privacy to all this, but that will happen toward the end as the exact transaction and dataflow clarifies.
I could do privatechains without too much work, but I dont really see the value for that as even with a private chain, there still needs to be a solution that uses it. I think a privacy enhanced anything to anything DEX is a more useful thing, but maybe there will be something that having a privatechain will solve
I came up with a way for people to test the DEX without needing any counterparty that is also testing, but it will require running a native iguana at first, so if you want to help the best thing is to get up to speed on installing native iguana. It depends on a single external module: pthread, so should be relatively easy to build on unix based systems
