No, no loss of safety. Was there something in particular you were thinking of? With IO-wrapper effect systems you give up native multishot continuations (but you can always get them back by wrapping in LogicT).
You can look at my table of "A Comparison of Effect Systems at a Glance" to see all the tradeoffs:
> No, no loss of safety. Was there something in particular you were thinking of?
This is second hand information and may apply more so to effectful then bluefin. I have just done some pretty minimal test projects with bluefin and effectful but I was told:
"""
It's not too hard to make segfault if you organize your higher order effects just right.
It's been a while but i think you just return a continuation with the effect you just handled, and then all of a sudden it's not in scope and yet you need to still handle it.
And the tricks they use to make things go fast don't work in this case and things go boom
"""
That definitely doesn't apply to Bluefin. I can't rule out it applying to effectful, but it is careful to do dynamic checks to make sure you can't use effects in the wrong scope. Maybe you're thinking of eff (i.e. what started this thread)? Alexis King described how to get it to segfault in this video: https://www.twitch.tv/videos/1163853841 (now private, sadly).
You can look at my table of "A Comparison of Effect Systems at a Glance" to see all the tradeoffs:
https://hackage-content.haskell.org/package/bluefin-0.2.0.0/...