Could you breakdown the types of boilerplate you're finding? IME most of it is from having generic (Hinze style) ways of handling datatypes instead of specific ones.
-
-
-
1. Gambit Scheme ports are so much nicer than the disparate and underpowered OCaml channels and buffers. 2. Type-descriptors and macros are so much nicer at generating I/O handlers than either manual typed combinators or ppx transformers. 3. All the monad cruft, gone.
- 20 more replies
New conversation -
-
-
So you're trading off LOC for safety and maintainability (i.e refactoring help). Claim: if the final code size is >5 kLOC anyway then this is likely a bad tradeoff.
-
What if I'm turning a >15 kLOC codebase into a <5 kLOC one by cutting the cruft? What if I'm working on adding proper types to Scheme, whereas adding proper macros to OCaml is hopeless?
- 1 more reply
New conversation -
-
-
Would the same be the case from OCaml to Common Lisp do you think? Or does your Scheme still have an edge over CL?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Good tests, of course. Also not always obvious: refactor *more* often, not less. Frequent, smaller changes.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
your a hero. the last time i looked at doing something with/in OCaml, i ran away to seek a different worthy windmill
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Similar experience when I first wrote a system in OCaml and, later, rewrote in Erlang. (If Erlang were fully reflective, with an IDE to exploit that, i’d have had something approaching the refactoring capability of Smalltalk.)
-
And it tickles/irks me that I now find myself writing in C# daily and wishing I could convince my team to move to F# for the same reason.
- 4 more replies
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.
Read my blog!