Am Fr., 2. Dez. 2022 um 16:10 Uhr schrieb Lassi Kortela <xxxxxx@lassi.io>:
>
> > The name is then a bit misleading.  The established symbol-xxx
> > procedures work on symbols.
> >
> > Consider changing its name to something like append-to-symbol.  Or
> > generic-append/symbol.  Analogous would be append-to-vector,
> > append-to-string, ...
>
> Common Lisp has:
>
>  > (concatenate 'string "fo" '(#\o #\b) #(#\a #\r))
> "foobar"
>
>  > (concatenate 'vector "fo" '(#\o #\b) #(#\a #\r))
> #(#\f #\o #\o #\b #\a #\r)
>
>  > (concatenate 'list "fo" '(#\o #\b) #(#\a #\r))
> (#\f #\o #\o #\b #\a #\r)
>
> We could imagine:
>
>  > (concatenate 'symbol "fo" '(#\o #\b) #(#\a #\r))
> |foobar|

Concatenate is already taken by SRFI 1.  Would probably be confusing
to attach a different meaning to it.

(concatenate 'symbol ...)

doesn't look like modern Scheme to me.  If concatenate-symbol,
concatenate-string, ... is no option (ignoring the SRFI 1-issue here),
I would at least propose

(concatenate (result-type symbol) "fo" '(#\o #\b) #(#\a #\r))

with

(define-enumerate result-type
  (string vector list symbol bytevector)
  result-types)

(In effect, (result-type symbol) expands into 'symbol but with
compile-time checking (so that (result-type zymbol) results in an
early error).)