Perhaps it's possible to get lucky and find good libraries using nothing but Google and patience, but I'm baffled that anyone would deny the value of something like CPAN.
CPAN's structure enforces good practices: (e.g., tests, docs, bug tracking).
CPAN provides not one but two excellent search pages. It also contains many reviews, and makes it trivial to see the source of any library or app before you download it.
it goes beyond CPAN - it would have been nice to have a "batteries-included" version that packs in most useful libraries (like Python's Image, lxml, etc libraries).
The very argument that choice > standardization is something I dont get - even in Rails, you have the choice of not using the default templating engine, JavaScript framework, etc. But it doesnt mean that they dont package it in anyway. This accelerates adoption, since it gets out of your way - and later when you mature as a developer, you can of course customize it wildly.
It may or may not also have the happy side-effect that the quality of the packaged libraries increase, because of a much larger user-base. I'd much rather that people fork mature libraries than hack the one-millionth version of an XML library that just does 2 things.
I'm not denying the value but I do think too much value is put in it. I certainly wouldn't use the lack of something like CPAN as an argument not to learn an otherwise fine language.
CPAN's structure enforces good practices: (e.g., tests, docs, bug tracking).
CPAN provides not one but two excellent search pages. It also contains many reviews, and makes it trivial to see the source of any library or app before you download it.
CPAN helps enable sprawling, group-built libraries (Catalyst, Moose, DateTime).
CPAN has a whole mini-network of related sites that provide secondary help (installation testing and reports, annotation for docs from users, etc.).
I really don't get what you don't get. It's obvious to me why a good centralized library repository is a win for any language.