I'm sorry if I sounded harsh or rude. It was not intentional. I understand that CPAN modules are made and supported on a volunteer basis, and that finding time and attention for them can be difficult, even at the best of times.

Actually, I've never considered Perl6::Roles for production use. Any module author who feels that a module they submitted is "a toy" or "not fit for production" or "obsolete" would do a huge service to the Perl community by documenting this in the module's POD.

My vehemence was related to the amount of time I have spent wading through too many choices on CPAN, trying to select a good contender.

This all really comes back to the "too many choices on CPAN" problem that has already been well chewed elsewhere. Breifly, there are so many modules on CPAN that it can be hard to choose a tool. We need to make it easy for people to distinguish what modules to use. CPAN ratings, AnnoCPAN are an attempt to address this problem. I've seen suggestions that CPAN be regulated and only certain modules be allowed on. I won't advocate restricting CPAN or even removing modules that maintainers consider obsolete or dead. I think its better if they persist, that way someone else can learn from them, adopt them or adapt them to meet their needs. A module, created as a toy to test an idea and then abandoned, may be taken up by someone else and be transformed from toy into masterpiece. If we remove "obsolete" or "dead" modules from CPAN, we prevent this from happening. The problem is that you can accumulate a pretty big pile of cruft. The best way that I know of to deal with the cruft is explicit labeling.

I should really take the time to formulate a clear meditation on this subject.

I didn't mean for my comment to seem like a raging rant aimed solely at you (or TheDamian). Your post just triggered an epiphany about one of the ways that our beloved CPAN sucks and perhaps a simple way to make it a bit better.