I remembers, almost a decade ago, I got into an argument with someone here, who thought that significant whitespace languages were somehow SO fundamentally different, that they couldn't be translated back and forth to a bracket delimited language. (This is one of the hazards of not getting a formal education, or of not having enough curiosity to self educate at a deep enough level.)
There probably is something to the point that whitespace languages need more context to know what is going on? Moving code, in particular, is a bit more tedious with it. That said, I'm curious why anyone would think it can't be just as "mechanical."
Languages where blocks are indicated by indentation are not context-free (you have to keep the current and previous line indentation level as state). Everything about them is more tedious because you can't do structured editing.
And I know many become an indentation shift. This can be a lot more intrusive then just moving a paren. I personally find it harder to match indentation shifts. (Excepting simple one liners, of course.)
I usually do the alignment with tab: my Emacs wisp-mode knows the indentation levels in use in the file, so it’s just shift-tab to go back to the previous scope and tab to nest more deeply.
One observation I made is that wrapping an expression around several lines is often less intrusive, because I can simply insert a line before the block that uses only half the indentation.
