[hondadriver]

The quiz is called AWS Icon Quiz. "AWS has terrible icons" is only the title of the Hacker News post.

[ryanianian]

It's still priming the responses. If the interest is in drawing a truly 'scientific' result, the hypothesis/wanted-conclusion shouldn't be in the prompt used on participants.

[birdtime]

I think the point was that the fault may lie in the poster on HN and not the author of the survey, provided they're separate individuals.

[Slippery_John]

The questions and candidate answers themselves are also heavily biased. The survey is clearly not intended to be an unbiased scientific poll.

[ryanianian]

Sure - either way it's not a "fair" survey. If I were making this survey trying to be scientific and somebody posted it to HN with this title I would be quite upset at them for priming a large number of my participants.