⭐ Merited by tyz (3)
Just to remember to this post. Check my signature...
Do you really compare Darwin with the a DAG. Sounds like a little bit awkward. I see a similarity between Tangle and the human race (or mostly animal or living species): every child has a father and a mother. In the Tangle every transaction checks two earlier transaction. Or?
https://i.imgur.com/em13wp6.png
I think, later we need Darwin really to analyze the Tangle - how was "new species" born.
"new species" = many many new economical communities, they don't wish to mix with others, because i.e. they has already conflict of interest with other economies on the Tangle......
Are you wish a such economical environment later for our human race?
https://upload.wikimedia.org/wikipedia/commons/8/8f/Edward_Hitchcock_Paleontological_Chart.jpg
https://i.imgur.com/em13wp6.png
I think, later we need Darwin really to analyze the Tangle - how was "new species" born.
"new species" = many many new economical communities, they don't wish to mix with others, because i.e. they has already conflict of interest with other economies on the Tangle......
Are you wish a such economical environment later for our human race?
https://upload.wikimedia.org/wikipedia/commons/8/8f/Edward_Hitchcock_Paleontological_Chart.jpg
In the current IOTA implementation, nodes only trust (value) transactions referenced and approved by milestones issued by the coordinator.
Similar to Bitcoin, where checkpoints were used until 16.06.2014, the use of this centralized "finality device" is currently still necessary to ensure the security of the network, which is in its infancy.
Maybe you should aware of the fact, that until end this year the concept "In the Tangle every transaction checks two earlier transaction" will be gone. So, no more Darwin...
At this Point, consensus, will be defined by, (FPC) "Fast Probalistic Consensus".
What is FPC?
In fancy terms, the FPC is a leaderless probabilistic binary consensus protocol of low communicative complexity that is robust in a Byzantine infrastructure.
Let us see what all these terms really mean, and first introduce a lighter version of FPC that is easy to implement and contains most of the important features.
We assume that there are n nodes indexed 1,2,..., n in the network . Each node i has an opinion or state. We denote s_i (t) for the opinion of node i at time t. Opinions take value in {0,1} ; for this reason we speak of a binary consensus.
The basic idea is majority voting. In each round, each node queries k random nodes and sets its opinion to the majority of the queried nodes. It's that simple! However, this simple majority voting proves vulnerable to faulty nodes and malicious attackers. It needs an additional ingredient, in fact an additional randomness, to make it robust.
More precisely, at each step, each node selects k random nodes C_i , queries their opinions, and computes the median opinion:
I compare IOTA to Darwin - the LIFE working since some years on our globe... >;DO=
I think in my post, the TANGLE will be forked later, and it will have always a leader (!!!!!) in this leaderless consensus protocol ... as in the LIFE we taste it.
Just some question.
1.) Are this algo' solved already an attack with small transactions? i.e. if some hundred, thousand nodes producing only 1 IOTA (!!!) for random addresses, and later try it to use again?
It was a question of me targeted CfB (sorry BCNext on reddit, but he / she dleted that acc after the conversation ... ... ...) on early 2017 ... in IOTA AMA ... I asked some "anything", sure.
One "anything" of me was solved - MAYBE - with FPC (( but with maybe a leader specie later through a kind of an EVOLUTION of the LIFE of all FORKs of TANGLE's >:DO= )) but I asked this possible and simple attack also.
And:
2.) How long will run this testnet of IOTA, and from which time can we ((the original investors of the IOTA project)) calculate with a start of the well tested REALNET with our original stake? Just to KNOW >;DO=
1. Yes, this has been solved. The new model is that you have Messages with certain types of payloads. So for example you have a Transaction Message, which has transactions as payload and you have an Indexation Message, which has an index and data as payload. Indexation Messages are data only. In this model Transaction Messages are not allowed to be 0 valued anymore, but that's not the actual dust protection. Dust protection currently comes from the fact you are not allowed to send transactions to an address with less then 1 Mi (this value may change in the future).
2. When i get your question right, you are asking for the coordicide network? Already now you can run your own node and test the network which runs without a coordinator. The "Coordicide-Network" handles all of this with a set of modules and functionalities that work together to ensure consensus is achieved. Up to now, few functionalities are missing. In a few weeks - my personal guess: it will happen end of may - for the first time all modules are complete and you can try the new network on your own. The testnet will have gradual incentivized participation and a bounty system. The goal of this network is to test for any bugs or issues that need to be fixed before the final release of the mainnet. Again: My my personal guess is, we can see a final release-canidate end of year.