-snip-
- the weight of block-chain that can reach and go over 1TB (it start to become to big for a lot of users)
Running a full node isn't mandatory for most user who requires basic functions such as sending and receiving. Additionally, users could implement block pruning in the future which would reduce the size. Only a few nodes are needed to run the full copy.
- the computing power required to confirm every more and more transaction (not for all again)
The computing power does not need to grow to confirm more transactions. Node wise, I believe there is a prototype that can achieve 4000transactions per second on a CPU that is 3 generations older. It should improve with the recent CPU release.
- the probability that the check system could go all to less companies that can substain power and storage
I don't get that.
- the probability that anyone can generate a private key from my public key.
No one can know your public key except those who have your private key.
- the probability that anyone generate a bruteforce attack generating a lot of keys (public,private) and steal bitcoin
Highly impossible and expensive. There is 2^160 keys that exist, it would take the power of several suns and extremely efficient computing power to stand a chance and it would still take thousands of years. However, it would be possible if the wallet has a weak RNG implementation, generating non random and easily guessable private keys.
Thank you
Welcome.