All of this could theoretically work better with our current technology - Internet, digital documents, file exchange systems; no need to pass around physical devices. Except somehow, it doesn't.
(I think in the Star Trek universe you would just reclaim the PADD in the replicator once you're fully done with it rather than move it around to some central "PADD library", right?)
RE replicators, you're right! I don't know why I haven't thought about it. PADDs are very simple as electronics go (even as today's standards go). DS9 showed people recycling dishes and food through a replicator, so the same would work for PADDs.
Feels like a very good use of ML on the part of the mobile OS makers would be to decide, based on place and time and activity history, which of your active widgets should appear above the fold on the lockscreen of your phone.
(E.g., when near a bus stop, next-bus times should appear. First thing in the morning after the phone has been inactive all night, the "summary of the calendar events for today" widget should appear. Weather should appear ~5 minutes before you usually leave the house, if-and-only-if there's something useful to know, e.g. rain predicted for today. Etc.)
I know iOS has "Siri Suggestions", which is similar, but these are activity-history-corrleated tap-to-open links for apps, not specific actionable at-a-glance information.
I think Google Glass tried to do this?
But my iPhone does automatically pop up my grocery store membership card on the lock screen when I'm at the grocery store, and my boarding pass when I'm at the airport. I think that's currently done via Wallet and may not help with bus schedules, but a more curated approach does have some value.
The thing is, mobile OS makers, especially the ones with huge investments in data and ML, have a conflict of interest. A large majority of them are in the business of monetizing your attention and the data they can collect about you, not just helping you focus on what's most important to you.
This is a really good case for free/open OSes, and ML projects. But the resources needed to develop free and open software are much scarcer because they tend to be harder to monetize / not as profitable.