Hacker News new | ask | show | jobs
by gus_massa 3454 days ago
> You sound very negative towards conducting more experiments on this, when the results have been unexpected every time it's been tested.

This is wrong. There is a strong selection bias, where a many team tried this and only those that got a "successful" measurement get press.

It's very difficult to get the list of all the unsuccessful (unpublished) experiments, but someone recollected a list em-drive test http://emdrive.wiki/Experimental_Results The important column is the last one. More than 1 means that if it's correct the device is breaking the current laws of physics. Anyway, I count 5 zeros in that list. [And I think that the other are experimental errors.]

1 comments

A relatively strong argument for experiment pre-registration as a prerequisite for eventually publishing.