To be honest Rakulang is quite comfortable for Ethereum dev. Traditionally a lot of tools and utils were written in JS (the well-know Web3 and others) and when I tried to do/implement smth similar in Raku I was very surprised: Raku has many handy features out of the box (big numbers, binary buffers) and also is very flexible for the bindings. I mean if you have experience with Perl's SWIG, Raku's NativeCall is much more clear and powerful - so you can easily write binding to any C/C++ lib, like it was done for secp256k1 and original Keccak-256.
Anyway, I think Raku is very good for Ethereum as well for any other blockchains.
Perl has a reputation (justifiably IMO) for the default being unreadable/unmaintainable code, particularly because almost everything was done with regular expressions.
The author is referring to this when juxtaposing it with it being used for a financial system.
Raku is not the old Perl, so I think it is unfair to bring up the patterns of Perl 5.x and earlier. Pretty exciting to see a real use case for Raku/Perl6, and for it to be as heavy as Ethereum code imho.
I only know the IRC Tcl community; I have a different experience, but not that I was caring much about the community anyways. They will not deter me from using a good language.
I know, I know, it's not Perl anymore, but still. Life is stranger than fiction.