I was proposing the cgo option because it's already implemented.
I _think_ allocating a slice of contiguous bytes and using unsafe pointers should work fine as long as you are very cautious about structs/vars with pointers into the buffer getting freed by the GC.
I've also received around ~10 of these same emails offering increasing prices to remove my review.
Maybe this is a good investment opportunity? I'm joking, but they did offer me almost twice the price of the product itself, which is interesting.
Something also curious is that the seller initially sent emails starting with with "Hi, [my sister's name]". Her name was not associated with my amazon account at all, and definitely not the name shown on the amazon review. I wonder where they are retrieving the names from for this.
This argument reminds me of the historical arguments made against heliocentrism vs geocentrism. Just because we have widespread belief of one thing to be true does not mean we should blindly accept it and not test things that appear to oppose it.
In my opinion, the tests that have happened with the EmDrive are exactly what _should_ happen to make sure that our knowledge is strong and that we do keep advancing the field.
The 44% figure comes from the Greater London Authority, but should be clarified a bit. There is a BBC article[0] about this which states "... 44% of the city's people are now of black or ethnic minority origins".
I _think_ allocating a slice of contiguous bytes and using unsafe pointers should work fine as long as you are very cautious about structs/vars with pointers into the buffer getting freed by the GC.