I have yet to see someone that gives a detailed technical Counter Argument to the Guys here which making me have doubts about ethereum "Long Term". Following this thread now
do you have a summary of what these doubts are long term? 8 pages is a bit much to read.
Just read the 1st Page and 1st TPTB_need_war first replay on the thread. Basicly i was very enthusiastic about CASPER and Scalability potential of ethereum.
Until i read some good technical counterargument, i'm reducing my investment in the ethereum project long term, which is best now since prices seems to be good enough for me.
I just read: http://www.multichain.com/blog/2015/11/smart-contracts-slow-blockchains/
they have some very good arguments, but does it mean ethereum is doomed? hard to say until they reach that point and possibly come up with another solution. you got to give it to them on the marketing front.
That multichain.com article is talking about the lack of parallelization, but as I (and CIYAM and even Vitalik) have pointed out (both upthread and in my video), parallelization can be fixed by simply restricting the inputs of scripts to data from already confirmed blocks (i.e. do not allow scripts in the current block to impact scripts in the current block).
Edit: by parallelization we mean the ability of a validator to validate more than one script simultaneously. Distinguish this from partitioning where we have validators who trust other validators (since not all validators validate all scripts).
That multichain.com article does not address my point about partition boundaries being impossible for scripting. That is my point and I am sure no one else has thought of it. I am the first to make that point. And this is a damning point because it means all validators (full nodes) must verify all scripts. And from that comes the other points I have made about why mining is forced economically to become centralized (and this applies to Bitcoin as well). Thus I have argued don't pretend that crypto currency has decentralized validation (verification) because it can't. And thus I argued just make the validation centralized and try to put the decentralized control else where in the design. Which is what I have contemplated for my design. Which I am working on.
Ethereum could hire me to solve their problem perhaps. I wouldn't charge $18 million.
Fuserleer has been trying to teach the masses that miners of a successful coin always get rich. It is a necessary evil.
Not true in my contemplated design, because the mining is unprofitable.