Exactly. Incremental reading was touted as some ultimate productivity/pleasure hack but it's impractical for the real world were humans have to synchronize with each other's calendar. It may work for Piotr because he doesn't have a schedule to follow, at all.

When I need to read and learn, it needs to happen within a timeframe. Not "some day" when it shows up in my queue again.

That's usually not a problem. You can always schedule the reading of an element manually. And if you do that for the first repetition, it will not have any negative impact on the algorithm.

You can also put stuff in something called "pending queue" and then just pull it from there once you're ready for it.

Only for the subsequent repetitions the algorithm will take over. This is like reviewing notes basically.

Also, you can manually advance repetitions if needed, for example before an exam.

It works very well for me and I constantly have to (and want to) learn lots of stuff.

What you describe fits the 'Plan' feature, not incremental reading. You don't sync with anyone other than your future self...

If you have deadlines for certain knowledge, then this can still be managed; it just requires some manual overruling via 'advance' or subset reviews.