👍

🌵

👎 Unlike the others on this list, this is the only one that actually makes sense curried, And if it's used that way, I would guess that nCopiesOfFoo is the more likely desire than fiveCopiesOfVal.

@davidchambers: You've been our strongest advocate for consistency, and I admire that. But I think there is a more important consistency here than "where parameter n goes", and that is the one that says our parameters are ordered from the least likely to change to the most likely to change, for the best experience possible with currying. In rare cases (prop/propOf), we'll expand our API when that choice isn't obvious.

Damn, A few seconds late.

Sorry, I've been both sick, and lacking Internet, managing to see a few issues on my phone, but not really to engage.

In that case, how about we revert to the old behaviour but rename the function R.repeat? As with a function like R.take, it's clear that one of the arguments must be an N. We can thus keep the preferred argument order while making the xxxN family consistent.

No objections.

I like your focus on naming consistency; it's kind of like having JDD around to poke us on performance here and there: While I know you're almost always thinking about these things right, it's often not something I would have thought about myself.

👍