Basically I've realized I tried to get out of economics last spring and failed "I would love a data science job BigCorp" "This job says economist on the tin but its really a data science role" "Oh cool" Narrator: it wa not a data science role
-
-
Show this thread
-
In the process I think I have burned out hard and I'm playing with more radical solutions
Show this thread
End of conversation
New conversation -
-
-
this is like saying Voldemort out loud. you're gonna get it now.
-
I got ratio'd hard
-
every damn time. hey watch this "RUNTIME ERRORS ARE NOT THAT BIG OF A DEAL IN THE MAJORITY OF WEBSHIT SYSTEMS!" ::ducks::
-
can keep going on like this "MOST BUGS ARE LOGICALLY AND SYNTACTICALLY CORRECT EXPRESSIONS THAT IMPLEMENT CONTEXTUALLY WRONG BEHAVIOR"
End of conversation
New conversation -
-
-
what's the use case for programming in python
-
or java or ruby or whatever
-
Well I can do that already but maybe I should learn Haskell and I am asking why that might be good
-
ah yeah. use case for haskell: more enjoyable for doing stuff
-
minimally fucky path to getting things done
-
also: OO+inheritance polymorphism is a mistake, haskell makes it easier to avoid building enterprise disaster codebases
End of conversation
New conversation -
-
-
Make invalid states not representable. I compose types for solutions like I'm a 1984 Party language designer trying to make invalid thoughts inexpressible.
-
This is one of the only pragmatic arguments I have seen, although maintaining invariants is hardly unique to FP.
-
I would argue that strongly typed FP is best at it, though. Fmap and monadic composition really make Either and Maybe easily usable without which much more error prone boilerplate is required
-
Oh, also sum types and pattern matching on them (with compiler warnings on partial matches) are huge.
-
Also applicative parsers for things like command line flags, optparse-applicative alone justified Haskell. See also composable parsers as used for language design in general
End of conversation
New conversation -
-
-
charging higher rates as a consultant
-
basically all "data science" is done in python these days
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.