2. I have 2 TB storage but I guess average might be well above 1 TB today. At least 5x increase is reasonable in 6 years
Unless new technology comes up, no it isn't.
If you prove we will not see over 10 TB HDDs in mass production in the following 6 years, then you can get some big blockers more sceptical about scaling above 2 or 4 MB. So go ahead.
3. Rasperi PIs are dead path for the most successfull crypto imo. You dont release best games with minimum system requirements so you can please everybody. You stick to something like top 75% of user PC market and only this way you can release top product played by most. Releasing India game with basically no minimum system requirements will not give you a better seller...
You guys seem to lack some logic though. Who cares about the Raspberry in particular; take any model of a processor that you want. Where is the limit? This was discussed at a workshop IIRC; new nodes being unable to ever catch up to the network. You can't deny it now though, essentially you would push X amount of people with systems that are unable to handle the network anymore -> smaller number of nodes -> decentralization harmed.
As previously said, this is definitely going to happen the question is if it is going to be negligible.
Nodes with average system specs catching up pretty nicely (4 core 2 GHz CPU), until checkpoint it is never issue there is network speed biggest botteneck. Only about the last about half year takes most time because only then all transaction signatures are checked. Fortunately today standard 4 core CPU rig can use all 4 cores to catch up in few days at most. Obviously in future number of cores average rig will have is going to increase, and Bitcoin devs can put the new checkpoints in every minor version (compared to major version today) so you will need to really spend most time on last about 2 months of syncing only and not about half a year as of now... pretty easy fix helping scalability for new nodes syncing from 0