4 ms·

I think the argument against feature flagging this (that this reduces app complexity) seems a bit thin. Normally you want to reduce the number of options in pl

I think the argument against feature flagging this (that this reduces app complexity) seems a bit thin. Normally you want to reduce the number of options in play so that you don't need to think about all the permutations when making future changes — if you introduce a feature in the future, you're not stuck supporting edge cases. But the whole point of this change is that they don't want the backspace behaviour overloaded (if it's bad UX); so it seems unlikely this would interfere with changes down the line? Annoying that adding the toast and creating a browser extension was probably more work than adding the feature flag would've been.

MaulingMonkey 10y ago
> Normally you want to reduce the number of options in play so that you don't need to think about all the permutations when making future changes Also to reduce testing overhead when you invariably don't think of permutations - which is now that much harder as you'll need to manually test with the feature flag enabled, since everyone else will probably be running with the defaults. You think all text manipulation code in and around the backspace functionality in Chrome is static? Seems unlikely to me. Also reduces refactoring overhead - less code, after all. To me, the question is which will happen first: 1) They remove the toast intentionally after everyone's muscle memory has adjusted. 2) They remove the toast accidentally by refactoring input logic and breaking the popup functionality without noticing.