Thanks!
1) So you'd basically like an easy way to take a card that's currently 'new' and not being studied and add it to the first step of 'learning' in a particular profile?
2) Yeah, that's on the grand to-do list but it'll be post 4.0. (to be honest, the % of Pleco users who also use an iPad is a lot smaller than it used to be, but we expect the desktop version will boost demand for this considerably)
3) It's pretty simple - if you use the default 'learning' profile with 'cycle all cards,' there isn't really an algorithm to speak of: once a card makes it to 'review' it simply comes back in a regular cycle, with intervals between reviews getting longer or shorter depending on how many other cards you have in 'review', but with a couple of shorter 'relearn' steps if you ever get it wrong. With the SRS option, it's a minimalist SM2-ish SRS algorithm, correct answer -> multiply interval by easiness factor, incorrect -> reset interval and lower easiness factor.
This is kind of a philosophical thing for me, one I've been pushing since our earliest flashcard days in the '00s - I've never seen much evidence that long-term memory decays in a predictable algorithmic way, and even less so when it comes to something like language where you're constantly encountering words in the wild (or related words that nevertheless jog your memory of the original words), so however brilliant your algorithm may be, it's working on a flawed premise and on incomplete information.
There's also the additional real-world problem that people don't study on reliable schedules - for years we pushed the idea that SRS reviews should not allow you to set a limit on the number of cards to review because if you reviewed cards after their due date then the algorithm says that you're likely to forget them, but we finally dropped that idea recently because people hated that they'd end up with large backlogs of reviews they could never get through.
So a huge number of people on our current SRS - and, for that matter, Anki, which wisely started doing this a long time ago - are not actually reviewing their cards on anything approaching the algorithmically optimal spacing, because studying often + reliably enough to maintain that spacing is a challenge for most people. At which point, we're better off with something simple and humane that doesn't make a big deal out of due dates and whatnot.
There's a certain segment of users who are hell bent on fanatically maxing out their flashcard memorization with algorithms, but I don't believe we were ever really in the running for those people - they moved over to the Anki ecosystem years ago and are now excitedly plugging their intervals into ML algorithms to make them a fraction of a % more efficient - and frankly I don't think it's a particularly good use of our time, when the benefits seem minimal and are confined to a small number of users, and when, as I'm fond of reminding people, this is a dictionary app primarily marketing itself as a dictionary app
