From @masak

Rakudo r31066 works under the assumption that the absence of "is rw"
on attributes means that they cannot be modified, period. This reduces
their utility somewhat.

$ ./perl6 -e 'class A { has $.x is rw; method a { $.x = 5; } }; A.new.a' # works

$ ./perl6 -e 'class A { has $.x; method a { $.x = 5; } }; A.new.a' # fails
Cannot assign to readonly variable
[...]

$ ./perl6 -e 'class A { has $.x; method a { $!x = 5; } }; A.new.a' #
this works, though

My intiution about this may be off, but I read "is rw" as "not
_publicly_ writeable". I think even a non-rw attribute should be
changeable from within its own class.

(Substituting a "%.h" for the "$.x" causes another segmentation fault
in the failing case, by the way.)

From @jnthn

On Sat Sep 13 04​:18​:27 2008, masak wrote​:

Rakudo r31066 works under the assumption that the absence of "is rw"
on attributes means that they cannot be modified, period. This reduces
their utility somewhat.

$ ./perl6 -e 'class A { has $.x is rw; method a { $.x = 5; } };
A.new.a' # works

$ ./perl6 -e 'class A { has $.x; method a { $.x = 5; } }; A.new.a' #
fails
Cannot assign to readonly variable
[...]

$ ./perl6 -e 'class A { has $.x; method a { $!x = 5; } }; A.new.a' #
this works, though

My intiution about this may be off, but I read "is rw" as "not
_publicly_ writeable". I think even a non-rw attribute should be
changeable from within its own class.

The discussion that ensued on p6l, IIRC, meant that what Rakudo is doing
is the Correct Thing. So rejecting this ticket.

Thanks,

Jonathan

The RT System itself - Status changed from 'new' to 'open'

@jnthn - Status changed from 'open' to 'rejected'