As noted in chatter, davorg was correct about Kage using symbolic referencees but was incorrect in thinking that the syntax $a->{key} does not work when $a isn't a "hard" reference (that is, when $a simply contains the name of some global hash variable).

Of course, use strict would prevent it from working, but that also prevents ${$a}{key} from working.

Let the record show that tye is 100% correct and my assumption that $a->{key} implied hard references was 100% wrong.

Let the record further show, however, that in my opinion allowing that syntax to work with symbolic references is a very bad idea :)

--
<http://www.dave.org.uk>

"The first rule of Perl club is you don't talk about Perl club."