I think a PL to explore novel concepts will hardly ever make it into mainstream. Those languages from the 90s reused concepts that were developed decades ago bu

I think a PL to explore novel concepts will hardly ever make it into mainstream. Those languages from the 90s reused concepts that were developed decades ago but were eventually put into a form accessible for the masses. New problems will arise since the way computers are used will change which will pose new problems. But it will take decades to filter out which approach is most suitable for practicioners and those resulting languages will have little in common with the academic PLs whose ideas they reuse. I thus find it funny to name haskell and php in the same sentence (those languages don't simply have different target domains but they are from different planets). From the point of view of an academic PL researcher/teacher, that's somehow frustrating of course.