At my talk to Sydney.pm, I was surprised to hear how many people had suffered badly through what they called the "Catalyst Bad Months".
The November/December period was a particularly rough one for Catalyst users, as it hit some plugin snags as the module went through a major upgrade, and at the same time suffered from a conversion of Catalyst to Module::Install.
Two of the problems were probably largely my fault.
First was mind-merging with Audrey Tang after YAPC::AU to try and push Module::Install to be even simpler and easier to use than it already was. ("As easy to learn as Ruby" is really what we aim for now). This resulted in some Module::Install overhauling with a new version coming out almost every day for a few weeks. That caused some people a lot of pain.
And secondly, after declaring that doing Module::Install-based Task modules instead of Bundle:: modules was far better, I encouraged Catalyst to move over to doing its Bundle::Catalyst as Task::Catalyst.
The resulting combination caused quite a lot of pain for a lot of people it would seem.
Yet from inside the "echo chamber" (as Andy Lester calls it) it seemed like some minor snafus and were quickly fixed, but not a huge deal. Yet outside, for developers who don't follow the inner workings of the Perl machine, it was a month or more of death and pain.
It's always amazing when you get to meet people in person and find out about all sorts of problems you didn't know people were having, and how badly the things many of us consider nigglies impacts people in real life.
It was also very interesting to hear people's reactions to what one person described as "superstar authors releasing modules on a deadline, before they are ready".
It would mostly certainly seem that once someone starts to accumulate a profile via authoring or what you, they are mostly definitely held to a higher standard, and expected to release modules that are much more complete and mature than other developers.
Make of that what you will.