Question: should we have indentation-insensitive variants of indentation-sensitive syntax?

[osa1]

Some languages like Haskell and Koka provide alternatives to indentation sensitive syntax that are not indentation sensitive, often with braces and semicolons. (note: I'm not sure if they have an indentation insensitive syntax for all indentation sensitive syntax)

Some languages like Python don't.

I'm not sure what are the use cases for having indentation insensitive variants of indentation sensitive syntax. Is this necessary?

Given that Python gets away without it I'm inclined towards not wasting time with this.

[osa1]

Indnetation insensitive syntax can also be used single line if statements or match branches, and similar.

Currently we have these syntax:

match <expr>:
    <pattern>: <simple expr>

# or
match <expr>:
    <pattern>:
        <block>

I want to get rid of the :s, but that causes ambiguities in the single-line syntax:

C (a, b, c)

Which can be parsed as: C as the pattern, (a, b, c) as the right-hand side, or C(a, b, c) as the pattern.

With semicolons and braces, we can have just

match <expr>
    <pattern>
        <block>

and for single-line match alternatives use braces:

match x
  C(a, b, c)
    <expr>

match x
  C(a, b, c) { <expr> }