[freehunter]

You sound very negative towards conducting more experiments on this, when the results have been unexpected every time it's been tested. You're displaying the kind of dogma that science has been fighting against since the dawn of time. Either it's true or it's not, but we can't tell until we test it. And so far, it's only been disproven in theory, not in practice. That's worth more experiments.

[cjensen]

Occam's Razor says it is not true. If you read the linked article, it is saying, in the most polite and disinterested way possible, that it is not true although they allow, in the driest of terms, that it could be true. I expect if the same authors analyzed the Loch Ness monster or Big Foot, they would also disinterestedly point out the unlikelihood of either and point out the how awful the proofs put forth are while admitting there is no categoric proof that the monsters don't exist.

Pons and Fleischmann were straightforward in their error. This is bozo territory.

[bicubic]

I don't expect the em-drive to work, but the parent post has a point. We've had repeated experiments yielding unexpected results. It's worth an inquiry; if nothing else, to thoroughly explain what went wrong so the same mistake is avoided in future experiments.

Outright dismissing new ideas, no matter how far-fetched, is very much the antithesis of the scientific principle. You mustn't forget that everything we take as indisputable fact today, was an outrageous far-fetched theory at some stage.

It was barely yesterday that Barry Marshall was ridiculed for proposing that stomach ulcers are bacterial, because everyone 'knew' that bacteria can't survive in such an environment.

[mhneu]

Many possibilities of what went wrong are well-explained in the parent article, drawing on experience of other scientists that make low-force measurements. (E.g. forces on rig due to electrical current flow, or liquid flow.)

Unfortunately it seems likely we won't learn much by finding the possible sources of error - the sources are already well understood by people doing low force experiments.

[freehunter]

Wouldn't it be worth repeating the experiment controlling for different factors at the very least to rule out some of the "many possibilities"? If there are "many possibilities" for why something is happening, by definition we don't know why it's happening. Science says repeat the experiment until we know why it's happening, or at least until we can't rule anything else out.

[euyyn]

> it's only been disproven in theory, not in practice

There's a hell of an experimental body that led to (and supports) our current laws of motion. These laws aren't "just theory", and they definitely aren't dogma. Are these experiments probing the laws in a region they haven't been tested before? Can these unexpected results be reproduced outside of the framework of a cool engine for space travel?

E.g. there's a difference between measuring for the first time the spectrum of antihydrogen, which we predicted with the strongest confidence would be the same as hydrogen's; and measuring the spectrum of hydrogen with Rock&Roll sounding within, because "we can't know if any specific music genre will have an effect until we test it".

So yeah, if people want to spend their own resources testing this, the more power to them. But the way it's been done makes it look like they're more interested in a cool positive result than in unveiling the truth, and that mindset leads to things like the N-rays.

[gus_massa]

> 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.]

[justinjlynn]

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

[mhneu]

The only unexpected "results" happened in experiments with a lack of controls, so while nothing can be ruled out, it would not be surprising if all the data so far is explained by well known forces (like flow in the liquid supply lines.). It's on the previous experimenters to do proper controls.