Failure Friday: http://t.co/KgX1JM3m5m Not quite as aggressive as ChaosMonkey, but all failure testing is worthwhile.
miniblog.
Yikes. In #python, (-7) % 4 == 1 but Decimal(-7) % Decimal(4) == Decimal('-3'). That's really confusing.
Implemented quicksort in Trifle lisp! https://github.com/Wilfred/trifle/commit/4e7aa895e88fbb7f936927c4e8e14f25c9a618fd Inspired by the minimalist Haskell implementation, which is pretty but slow.
Untrusted, a JavaScript adventure game you play by modifying its source http://t.co/NRzFey3i3S Extraordinarily creative!
Today's HN article linking to Phrack has revealed that my mobile ISP has decided to start filtering 'adult content'. Hmph.
Learning another codebase generally involves fitting someone else's metaphors in your head. This is why devs are tempted to write their own.
After the separation of unicode and bytestrings in Python 3, I'm surprised that os.path.exists (and friends) accepts both b"/etc" and "/etc"
apipkg gives you lazy imports in #python: https://pypi.python.org/pypi/apipkg Neat, but I've rarely felt the need.
Obscure #python of the day: you can call methods on integers with a space. For example, `1 .__hash__()`
Current status: implementing clean functional programming idioms in imperative loops in Trifle lisp, so users can write pure code.
Coveralls (and other coverage sites e.g. Jenkins) are great. Knowing which code is untested helps find bugs and unused code.
"our Java6+GWT+Ant+Subversion stack was failing to attract new developers."--Wave developer. This is why Emacs would benefit from git.
J has 'monads' but they're just functions that take a specific number of arguments. Terminology overloading is unfortunate.
J has 'monads' but it uses the term for functions taking a specific number of arguments. Monads are bad enough without overloading the term!
After some soul-searching, I'm removing truthiness from Trifle lisp:
Showing 976-990 of 1,323 posts