So we have people on one hand saying 1TB or some more, and then on the other end you're saying 100PB(?!). Whose estimate is closer to reality?
If it actually is in the 1-2TB range I think that's pretty manageable to be honest. Most people want to use thin clients anyway.
It depends. If you think this is going to replace Visa (or perhaps larger, by including micropayments that are currently handled "off-chain" with Visa) in its present form that then that is probably in the PB range. If you are thinking about Bitcoin today and in the foreseeable future then that is not PB.
Anonymint to his credit wants to "think big" and focus on the former. I think that's entirely reasonable, but I also think it is fairly useless for him to be doing so on this forum, because there is little indication that Bitcoin-derivied technologies are going to do that and certainly he is going to get very little useful collaboration here because few if any in the Bitcoin space are currently making any serious effort to implement it.
The rest of us are (or at least the Monero team is) focused on incremental improvements on what already exists and adding value in the near future. So adding some level of privacy to Bitcoin which currently has essentially none, improving scaling even if that doesn't mean scaling to PB, fixing bugs, improving usability and performance, etc. It is entirely possible that all this work will be made entirely obsolete someday by what Anonymint envisions, but it won't be soon.