So, things moved on from when I last looked at Python & Ruby -- it seems they've also embraced the 'never mind the quality, feel the width' attitude -- but they are the wrong targets.

See Re^4: Would you stay with Perl if there were no CPAN?. In particular note:

"only 2% of them seem to be downloaded/released with any regularity, and indeed about the same 2% look to be the only ones I could imagine more than a handful of people ever finding useful, ever, just based on their problem space."

and

"Reading between the lines, they're [Haskell developers] trying to optimize for minimalism, efficiency, and elegance long-term, even in the published libraries, in exchange for some of the "benefits" of more "flood algorithm"-y approaches... As a result, the vast majority of Hackage packages implement thousands of "known" algorithms and standardized protocols/interfaces, making them very useful to scientists and other users of "hard" comp-sci. While not preaching "one way to do it", in most cases there is only one choice because it is so definitively/obviously optimum, there's no reason to ask the question if you really understand the problem space."

and

"don't compare Perl to Python or Ruby (same case with PHP) anymore, the three of those are so far behind the .NET ecosystem, the Java monster/monstrosity, and the less visible but ubiquitous JavaScript juggernaut, that if you want to talk about growing the Perl userbase by embracing and extending the other language communities, you should try to target the 90% of the the "trained" professional programmers who use the plurality languages/systems, not the other 10%."

I can't say it better. More packages won't help, unless those packages are authoritatively written and used by experts in vertical markets that are in current demand and growth. If nobody is using Perl; there is nobody to write those packages.