I don’t believe this article is good enough to merit a complete response. If you wish to build a readerbase and large following by doing this I suggest you make improvements to the structure, research, writing style and critique.
Some things to pick up immediately:
- You don’t seem to know the difference between control chains and side chains
- It is evident that you were not aware of 90% of the things you’re writing about before you wrote this ‘article’, and are trying to critique something that you have no background in.
- Your arguments are not arguments, and where you try to argue, you are often incorrect or misinformed, for example you state the random clustering isn’t random because it accounts for other factors. I thought this too, but I decided to research and learned that no, it is in fact random, but the size of the random pool varies each round based on the factors which you thought would not make it random. If you want to write these articles, please do more research first.
- You think that self optimization must mean that code is sentient. I should have stopped reading right there.
- You think transaction searching is automatically devoid of privacy and therefore is not “Satoshi’s vision”, I should have stopped reading here, too
- You admit yourself you don’t know what pattern matching is, but for some reason think you are qualified to critique such a system. This is why you do not understand the, as you call it, ‘program synthesis’. Google deepcoder. This is a form of program synthesis but uses pattern matching, and is much farther along than algorithmic program synthesis.
- You don’t understand the basics of cryptography, private key generation, as you question how to know a random seed is truly ‘random’. In Bitcoin, a private key is just a random number between 1 and 0xFFFFFFFFFFFFFFFFFFFFFFFF FFFFFFFEBAAEDCE6AF48A03BBFD25E8CD0364141. This number is derived from the order of the elliptic curve that Bitcoin uses, secp256k1. You verify its randomly selected by simply checking the source code.
A Markov Chain procedure is just a probabilistic state machine , imagine a washing machine , in the real world a washing machine is deterministic meaning that it knows exactly what’s it going to do when it starts , a probabilistic machine is the contrary when it starts at IDLE it has 25% to go to the DRY state and 75% to the WASH state , this process of moving from one state to another is described using a Markov Chain in mathematics .
- The quote above was particularly useless. I cannot understand the inclusion at all and it epitomises what I mean when I say you were previously unaware of the things you were writing about when you wrote this article. There is no discussion around it whatsoever.
- You state that no sane researcher will use an open source distrubuted computing platform to train datasets. Maybe you’re right. But you somehow looked over the fact that MATRIX also proposes public-private chain interoperability, making this a moot point.
I don’t wish to go on. I’ve already written more than I came to. The section about GAN was particularly bad and the entire article is riddled with spelling, grammar, and even definition errors. Anyone reading should take note that the author is anywhere near qualified to do what they are setting out to do here for any project let alone MATRIX, especially when trying to argue against this.