To put the thought out there, ‘better’ search and/or relevancy/recency ranking could help mitigate this concern substantially. ‘Sufficiently intelligent search’ certainly isn’t a panacea and can sometimes be an anti-pattern (I would much prefer explicit unindex to just trying to ‘magic it away with search’), That said, it can make a huge difference prior to manual optimization by a user.
Right now the reason why I’m interested in a keyboard shortcut for home and up/down is because the search doesn’t prioritize Home when I type ‘Cmd+K -> H’ and doesn’t put the parents and children of the current note high up in the Command Prompt. If the typeaheads and search were ‘smarter’ and did that, I might just use those all the time instead.
Algolia is probably not a fit here, but it’s historically had support for ranking based on recent, context, user behaviour, demographic, etc…
There’s no need to go that far, recency (like fasd) and context (parent/child) might be enough to make a huge dent here and make it far less concerning to have cards like ‘Tasks 07/2020’, ‘Tasks 08/2020’, where currently you have no idea whether the current or past versions will be prioritized in search.
I appreciate that I’m proposing what could be a fair amount of work — and I’m sure that you had something around ranking in mind on your roadmap already — but hopefully this is some useful material and food for thought at least.