You are viewing a single comment's thread from:
RE: Bitcoin visions
I've always had a positive feeling about Ethereum, though there are some things that are bad with it ...
- "Proof of Vitalik". Well, it was a quite bad situation with the TheDAO-hack, damned if they did or damned if they didn't, quite many have fanatic opinions on this, but I think it was probably the best decision to refund that contract through a hard fork. However, Vitalik was a bit too fast saying "of course we'll fix it". The decision power over the ethereum platform is in reality quite centralized. (maybe even more in Ethereum than in Bitcoin).
- Some technical artifacts ... for instance, 256-bit definition of a "byte". Well, maybe in fifteen years it will prove to be a smart decision. Today ... no.
- The biggest elephant in the room, IMO: what's the point with smart contracts if it's trivial for the developer to put in hidden backdoors - either by accident or deliberately? I do believe that both the TheDAO-vulnerability and the Parity-vulnerability was accidentally put in and not maliciously - but anyway, nobody saw it until it was too late. It seems to me that the contract language isn't designed for being a robust and simple contract language.
You definitely have a lot of knowledge my friend and you communicate very well. Let me ask you about IOTA. Is the Tangle technology better than Blockchain in your opinion?
I didn't find the time to read the whole whitepaper yet, but I started at it - and I also started setting up a wallet.
From the little I read in the whitepaper, it seems to be some kind of sharding - and sharding is on the roadmap both for Bitcoin Cash and Ethereum, so if I'm right it's not so novel after all.
If I've understood it right, every node needs to keep the full history of at least the coins they own plus more. If this network is supposed to be used for microtransactions between embedded devices then this may be a choke-point, it may be quite a bit of data with time.
The nodes in the network (low-power embedded devices) are supposed to secure the network through proof-of-work, I have a hard time understanding how this can give sufficient security.
The tangle is largely unproven technology. As of today, the network hinges on a black-box central coordinator. That taken into account, I fail to see how their tangle has any advantages as compared to holding all the transactions in a central database.
Indeed, security is the biggest elephant in the room, they have decided to make their own cryptoalgorithms - and people have already been poking holes in it.
There seems to be social issues as well, such as censorship of all critics in their subreddit, and the reactions from the team when the security issues hasn't been up to expectations.
Just like Ethereum decided to define the byte to be 256 bits, the IOTA team has also made a weird technical decision, defining a bit to have three possible values.
You mention them creating their algorithms and the surety issues they have seen and need to fix. Does this make them better in any way? If they fixed the security issues first I mean. Also, I’m trying to understand how the tangle can be sustainable if the transaction weight grows with speed overtime? Wont this bog down the system eventually? If new transactions need to reference the old transactions, wont this lead to a slow network overtime?
In computer engineering, one should avoid reinventing the wheel whenever one can. There exists cryptographic libraries that are used in many other software products. The probability that some grave security flaw will exist undetected by "white-hats" for a longer time is much bigger when using an ad-hoc home-brewed crypto library.
It's generally of course a nice thing that security flaws gets detected and patched - but if 14 critical bugs are found and fixed in software product A, and meanwhile software product B is equally much used and no critical bugs are found there, then that says something about the quality - it's much more likely that one will find 14 more bugs in A than in B.
(Y) great reply @tobixen