I would be interested to read a layman's explanation of why this - or a simple, limited version of this - is so difficult.
From my point of view as a completely coding-incompetent user, it seems like Zotero currently does [something] that produces (Smith, 2001), so I don't understand why it is so difficult to add the capacity to do [something a bit different] that produces Smith (2001).
It seems from reading this thread, others, and the github issue that there is a lot of thinking 18 steps ahead about all possible ways that some part of this might be slightly imperfect in some circumstances, while not providing a useful-though-sometimes-imperfect feature that would be beneficial while work proceeds on creating the perfect thing (or even if such work never is realized).
I don't think "allow for relatively effortless switching between the three main types of citation styles (note, author-date, numeric)" is 18 steps ahead, and implementing a future poorly as a first cut creates technical debt that makes it harder to implement it properly later on.
As a user, I would be delighted to have a narrative citation option that worked (approximately) properly only in author-date styles, and would accept that I might be making it more difficult for myself to subsequently switch to a different type.
(Much like, when I use \citet in LaTeX, I know that I may be in for some finding-replacing and annoying compilation errors if I switch to a different citation style later.)
I understand your point about potentially making things "harder to implement ... properly later on", I would just point out that it is not clear to me that, after 16 years, "later on" will ever arrive.
Upgrade Storage
From my point of view as a completely coding-incompetent user, it seems like Zotero currently does [something] that produces (Smith, 2001), so I don't understand why it is so difficult to add the capacity to do [something a bit different] that produces Smith (2001).
It seems from reading this thread, others, and the github issue that there is a lot of thinking 18 steps ahead about all possible ways that some part of this might be slightly imperfect in some circumstances, while not providing a useful-though-sometimes-imperfect feature that would be beneficial while work proceeds on creating the perfect thing (or even if such work never is realized).
Am I being unfair? What am I not understanding?
As a user, I would be delighted to have a narrative citation option that worked (approximately) properly only in author-date styles, and would accept that I might be making it more difficult for myself to subsequently switch to a different type.
(Much like, when I use \citet in LaTeX, I know that I may be in for some finding-replacing
and annoying compilation errors if I switch to a different citation style later.)
I understand your point about potentially making things "harder to implement ... properly later on", I would just point out that it is not clear to me that, after 16 years, "later on" will ever arrive.