But, isn't it a temporary solution? What will happen once that increase isn't enough to cover the same security it did before a certain halving in the future? Unless we double the block size each halving and the miners always fill their blocks with others' transactions we retain the same security given that the price remains the same.
I don't understand. Why is everyone so particular about *maintaining* security? That is not the point of PoW. The point of it is to make it prohibitively expensive for any adversary to launch an attack against Bitcoin. It is NOT an issue until we reach the levels where it becomes critical and profitable for an adversary to attack Bitcoin. We are very far away from that stage.
I think it is both unrealistic and unreasonable to demand the same level of security at all times. The resources needed to attack Bitcoin is astronomical at it's current stage, to the point where it is ridiculous to be attacking Bitcoin, and we still have block rewards. I don't see the need for block size to be increasing because there is nothing to be compensated for. Bitcoin is secure enough at the current stage, and it would suffice to have a linear increase for the near future.
Aren't we using a system where each user can verify by their own the validity? Sure, few pools control most of the computational power that is offered to the network, but the average person should always have the right to verify what they're receiving. If you make it too expensive, there's discouragement.
We are. I don't think the argument here is to make it prohibitively expensive for normal people to run a full node... I think you misunderstood my "sustainable" block size increase?
Let's be a bit more pragmatic. Would users prefer to run full nodes over an SPV node at it's current stage. Full node offers virtually little to no benefits to most and the security/privacy of it is overblown. There is simply no reason for people to still be running a full node, unless they intend to run a full node without any tangible benefits to themselves. At that point, then it probably wouldn't matter to spare a couple bucks more to get a larger storage.
Having lesser people run full nodes isn't a problem. It only becomes a problem if it becomes centralized to the point that it poses a threat to the network. Full nodes are important, I never denied that. A capacity increase doesn't result in the majority of the users that were previously interested in or otherwise able to run a full node to suddenly not be able to do so.
Technological improvement has made manufacturing cheaper, and in a larger storage density. It simply doesn't make sense for us to not be keeping up with it as well. 1MB as a limit was arbitrarily defined, and it stuck for so long before Segwit got activated. Even the SW block sizes are underestimating the true capacity of the network.