I also think there are better places to put this kind of money, including on projects that we are certain have obvious potential to change the world for the better.
What I was getting at was the very idea that we absolutely have to know what the return is before we start. Just because we know the potential return doesn't mean that it's not research (as in your fusion example), but just because we can't identify a return ahead of time doesn't mean there won't be one.
Also, I don't know if there have been any tangible benefits from the LHC. Precision manufacturing? Improvements in large-scale, multi-jurisdiction project management? Data analytics techniques? More efficient superconducting magnets? I don't know if those are actual side effects of the project and, if they are, I don't know that the LHC was the only way to get them.
Edit: or, like the quantum physics underlying our electronics, maybe we won't know for 50-100 years just how important that proof was.
I remember taking typing (on manual typewriters!) in high school before the personal computer.
I don't know if it was then, or 20 years later when I was taking a desktop publishing course, but I remember being told that, just like dashes, spaces come in 2 widths. The en-space, which goes between words, and the em-space, which goes between sentences. (There are other widths used in kerning, the spacing between letters.) The two-space convention of typing was an approximation of that.
This part I know I got from that desktop publishing course: As a result, the two space convention should be followed only when working with monospaced fonts, proportional fonts that don't offer an em-space, or software that isn't smart enough to use the correct space in the appropriate places and you're too lazy to do it by hand. (I used find and replace for that last case when em-spaces were available.)
Now, I can't say I really care. :)