Spaced repetition allows graceful deprecation of experiments
An "experiment" in spaced repetition usually involves trying out a new thing (e.g. new way of making a card, new parameters for deck options), and then waiting a few months to see how it goes (e.g. how well you remember the thing, how you feel about doing the reviews). Because of the nature of the spacing (in particular, the increasing intervals), the amount of effort required to "keep an experiment going" decreases over time (as long as you don't add any new cards). This means that even when an experiment "fails", it's not too costly to keep reviewing those cards anyway.
Here's another way of phrasing it: when you run an experiment in spaced repetition, you pay most of the costs near the beginning (when the uncertainty is greatest). As you become more certain of the outcome, the less you need to pay. So even when an experiment is a failure, you don't need to pay a huge cost at the end.
e.g. my cloze deletion "read only" cards weren't so useful and i eventually switched to a dedicated incremental reading deck. but i can still keep reviewing my old cloze cards! and they get less frequent over time, so there's no huge burden to continuing to review them. contrast this to something like note-taking: if you switch to a new note taking system you might need to convert your ENTIRE existing notes to your new system in order to keep the old content useful.
Other examples where terminating an experiment can be costly (near the end of a failed experiment): dating, switching text editors or operating systems (need to rewire your brain), switching jobs (e.g. need savings to be able to quit).
See also
- Spaced repetition experiments take months to complete
- Spaced repetition as soft alarm clock -- graceful deprecation is one aspect of spaced repetition's general softification ability
