throwawayg99

I submitted a bug to Facebook's whitehat disclosure 3 or 4 months ago. Got no response whatsoever, except an automated response. The bug still exists. The bug allows users to post as though they are other users on the timeline. I think that is pretty serious, but I guess they do not.

This is awesome. I looked, but couldn't find out: is there a word sense disambiguation layer somewhere hidden in here?

You forgot marquee!

Why are you suggesting A/B instead of multi armed bandit approaches?

You're right. That's a bit more reasonable, so back to my 10% change in base rate (1% abs.) but with a 90% power:

  > power.prop.test(p1=0.1, p2=0.11, power=0.9,sig.level=0.05)

     Two-sample comparison of proportions power calculation

              n = 19746.62
             p1 = 0.1
             p2 = 0.11
      sig.level = 0.05
          power = 0.9
    alternative = two.sided

  NOTE: n is number in *each* group

Requires about 40,000 samples per test. I would strongly recommend anyone serious about doing this look in to MAB testing, as A-B testing is way too expensive for reasonable scale testing (unless you have a strong a priori hypothesis to test).

Nor me, to collect a 1% change:

   > power.prop.test(p1=0.1, p2=0.11, power=0.8,sig.level=0.05)
     Two-sample comparison of proportions power calculation

              n = 14750.79
             p1 = 0.1
             p2 = 0.11
      sig.level = 0.05
          power = 0.8
    alternative = two.sided

   NOTE: n is number in *each* group

Edit: so that's 30,000 samples. To do the same with a 5% change in the same area is still 1300 samples (alpha=0.8, p=0.05)

I work with a dozen or so people who are involved in this sort of work. I think it is very interesting to see how they rationalize and deal with their moral compass internally.

One of the guys is the most caring, liberal, loving person you'd ever meet; he justifies being involved in this sort of skeezy marketing work as "I can take a small amount from a lot of people and amplify the result to do good with a lot of money."

He genuinely believes this. A lot of the other guys simply try not to see the "punters" (potential customers [1]) as real people, they are disconnected through the impersonal nature of the internet.

[1] http://dictionary.cambridge.org/dictionary/british/punter_2