RE: Why I am Super Excited about EOS
Herein lies the problem with EOS, it's untested technology and a Chinese security firm discovered multiple vulnerabilities in the EOS blockchain allowing an attacker to execute malicious smart contract code and remotely take over a wallet node aka really bad. Hardly an Ethereum killer, because Ethereum has some of the smartest minds in crypto working on it (through their own projects or Ethereum itself).
The hype surrounding EOS is going to be its downfall. Based on my unrelated experience in development and databases, I remember when NoSQL databases hit the scene, specifically MongoDB. The problem with MongoDB was that they achieved their breakneck speed by sacrificing a few things (namely data integrity checks). This resulted in some users losing their data because the database would accept it and offered no way of checking if it actually was written.
I do wonder what EOS has sacrificed to achieve the 0.5s transaction speed. It seems so fast, I have to wonder what integrity checks have been sacrificed to get that kind of speed.
They are foregoing Proof-of-Work and (to some degree) "decentralization".
I know there are quite a few people that are fanatically thinking any coin that isn't driven by PoW is a "shitcoin", and that PoW is the only real way of driving a crypto currency. At the other hand, PoW is an extremely inefficient way to drive a blockchain, and we're also seeing quite much of a mining centralization with coins like Bitcoin due to ASIC production (i.e., how many percent of the hashing power is now done through the AntMiner S9? I have no idea, but if I was to guess ... maybe 75%. Theoretically there could be some backdoor in the S9). I do have some problems seeing any benefits at all with PoW at this point.
Then there is the decentralization part - the EOS network will be driven by "some few" block producers. Yes, stakeholders may cast votes, but at the other hand ... pfft, see how many Steemers are actively using their voting power? In theory a corrupt block producer can be booted out of the network very quickly, in practice I believe the list of block producers will be fairly constant. Perhaps there will be one or two "swing positions", but that's all. Is it "decentralized enough" that 21 entities may decide everything (with maybe two entities being "swing block producers" and needing to be very careful keeping up a good image, and 19 entities that can do whatever they want)? Actually, I tend to believe so, but there are for sure many crypto-fanatics that will seriously cringe at such a governance model.
I don't believe any blockchain with totally free and unrestricted transactions will work out in the long run. People could be utilizing that for free backup of (encrypted) personal data, like photo collections, and people with a grudge against the project could clog the blocks with spam. In Steem, transactions are for free, but there is a bandwidth limit - one has to be a major token holder if one wants to put significant amounts of data on the blockchain.
EOS has a similar bandwidth limit. For all intents and purposes to me it results in free transactions for genuine holders and users. So there is a mechanism to protect against spam while keeping things free for the rest of us. It's a good model that seems to work inside STEEM anyway.
Considering Dan is light on the details especially when asked for a post-mortem on the fixes, I am not convinced that Dan is being transparent and honest here. Reminds me of the recent Verge (XVG) hacks where the creators falsely said it wasn't as bad as people said they were (which was a lie). Accountability in cryptocurrency is a huge problem it seems.
There was no "mortem" for there to be a "post-mortem". You can convince yourself by paying attention to github repo and dev channel discussions.
EOS has sacrificed the Proof-Of-Work mining on the altar of progress. It also has increased specification requirements for elected miners to enable multi-threading.
I'm not saying there won't be problems. I expect them. But I trust EOS to overcome them and deliver on it's promises within the first 6-12 months.
Bugs in an undeveloped computer coding software?!! NO WAY!
"vulnerabilities in the EOS blockchain allowing an attacker to execute malicious smart contract code and remotely take over a wallet node"
This lie cracks me up, do you even realize that there is no EOS blockchain yet? So who's getting attacked?