[gammalost]

My point is that something is not useless if it has the potential for future use. I would, for example, not call the example in the article (in any stage of its development) useless because it has the potential. Same for any possible paradigm-changing work. If it is proven (in the strictes sense) to not work, then it could be deemed useless.

Now if something is profitable is a whole different thing.

[arcwhite]

The thing is you don't know that it's got actual future usefulness. You might think or hypothesize that it does, but for various reasons it may turn out to be a dead-end.

Almost every endeavour has non-zero potential future utility - even as a counterexample or lesson to learn from. I think that definition of "useful" is probably too broad to make an argument against, and so not a useful definition.