Written in response to this.
OmiseGo fails at PoS security
In PoS proof of stake, the coins (i.e. the "stake") distribution = security.
OmiseGo was distributed in pre-sale to limited number of people that even required AML/KYC, so not permissionless sale. Obviously, AML/KYC doesn't prevent sybil attack as anyone can buy identities/people to buy tokens for them.
Effectively they used a permissioned sale with a cap (making it easy to displace other people for coin grabs) to distribute coins to be used for PoS where coin distribution provides security. It also prevented wider decentralization by introducing this barrier to entry which not everyone can provide.
- 65.1% - capped permissioned ICO = decentralization fail
- 20% - premine they keep = decentralization fail
- 9.9% - premine team keeps = decentralization fail
- 5% - airdrop on eth (also premined/ICO'ed eth) = too little to make a diff
I've already seen them advertise how many accounts have OMG without mentioning it's only 5% total.
We know ~30% (enough for security failures in standard PoS) is in their control and another 65% in permissioned hard to get into sale = decentralization failure.
The fraction they keep is minimum as it assumes they didn't participate in their own sale as that's easy to do (source)
Centralized company in complete control
- The company and its members hold at least ~29.9% of coins
- The company has full control over $19m from capped sale
- Most of the value comes from connections in Asia Omise has
This means Omise can do any changes (e.g. taking others money or censorship) with a fork and then:
- Decide which blockchain is called OmiseGo (OMG)
- Use premine to damage value of any original chains dissenting from Omise company decisions
- Hold updates and ICO funding hostage to keep all value on their chain
- Control of codebase where they can set changes to default
- Force all the business connections of Omise to only operate with chain of their choosing
All of these centralization results were demonstrated by Ethereum in the past successfully to subjectively & forcefully confiscate $100m+ and censor an unknown user.
Decentralization is measured by being resistant to attacks from an individual or organization. Omise, themselves or under influence of external force, can single handedly destroy the value and thus incentives for the entire network at any time or even use their stake to attack it.
There is nothing effectively decentralized about governance decisions on their network with enormous advantage given by massive premine and ICO.
OmiseGo control is obviously centralized as well
Plasma is unsecure & not a scaling solution
- Fraud proofs do not work because they make it profitable for everyone to cheat the system (thread)
*Child chains don't survive 51% censorship attacks - security failure (source)
Majority of plasma usecases aren't secure (source)
Ethereum is the worst network possible to connect to since it's literally the only main blockchain to have ever proved complete centralization (source) and there's pretty much nothing they can't modify (source). Also ethereum is way too slow to handle child chains:
tldr: Every part of OMG is centralized, calling it decentralized is fraud.
Ironically, everything OmiseGo hopes to do has already been done in decentralized manner elsewhere (example from 2014) and could be done far better. Not doing it correctly makes it a design failure even before launch.
There is only 1 reason OMG has todays evaluation - it was marketed to the most tech-illiterate community in crypto next to onecoin - ethereum community, with 0 talented developers, 0 innovation, and 0 tech since no one intelligent would ever choose to be associated or waste time on proven centralized fraud of a platform. Thankfully we had many developers switch to other cryptocurrencies after. But thanks to excessive marketing trust based centralized designs like ICO's allow, they were able to get an entirely new completely tech-illiterate community & exactly the type of community that happens to be stupid enough to throw most money at ICO's like OMG.
This is not investment advice. Technically unsecure or bad projects can be completely independent from price. This is more of why I won't be using OMG or eth for any reason.