Lacking Credibility

A “scientific” explanation of homeopathy.  This woman is either lying or completely deluded, either way she’s helping to take advantage of people.  Her misrepresentations of science are harmful, she is spouting meaningless drivel and calling it medical advice and claiming a scientific explanation.  There is a scientific explanation for why homeopathy appears to work, it’s called a placebo and it’s just another delusion.

..evaluating medical information using nothing but Jenny McCarthy’s brain must be a little like running an Olympic wind-sprint while dead… The Peak - I hate Jenny McCarthy - Opinions

SF Erlounge: Testing Erlang

Nick Gerakines from EA/Rupture previewed his TDD talk for Erlang Factory and showed off etap (http://github.com/ngerakines/etap/tree/master) his erlang tap testing framework.  Mostly a test runner with some useful coverage features.  Part of his mochiweb work is porting the tests to etap and generating coverage reports.  (22% or so coverage)

Cliff Moon from Powerset/Microsoft showed off the mock framework he’s using to unittest dynomite (http://github.com/cliffmoon/dynomite/blob/62d5d52cbc852784b50f1f3cee1dfa7e0a0669cf/etest/mock.erl)  This appears to be more useful in erlang than trying to compose a test harness by parameterizing the modules/processes you call.  Especially useful when you need to test code that touches erlang std lib stuff and don’t want to take the performance hit of abstracting it to something testable.

Other discussion were sizes of erlang clusters, Powerset has apparently spun up 200 node erlang clusters on EC2 instances for doing performance testing, Yahoo apparently has gone up to 84 nodes, Engine Yard does at most 8, and other companies seem to be working with <20 nodes.

It’s definitely a useful event even for people who don’t primarily work with erlang.  I’ll definitely be going again.