I didn't say that AISC ought to be used here, just pointed out that it has allowable stresses for a buckling situation that is not well addressed elsewhere (or at least, I've never seen it anywhere else). The reference to 0.66Fy in the original post is typical AISC allowable stress, so the suggestion doesn't seem out of line. Reference to that method, or similar buckling criteria, will also answer the question of "Does it matter whether bending is about the major or minor axis?" If one of those British Standards also gives an allowable stress for flat bars turned on edge, that would be handy to know as well, but I've not used them to know.
The AISC code in question is a free download. The way the code is written, it is very convenient to adjust the factor of safety if the application requires it as well.