I just released a new module, Google::OAuth. I've been pretty prolific publishing modules in the last few months. And I understand that CPAN tests each distro on a wide variety of platforms. But is there any other way for a publisher to get feedback?
Specifically, I'd be interested in the following:
- Is the documentation readable? Even if a distro requires a particular expertise, the documentation needs to be accessible enough to identify its intended audience.
Is the module useful? Although I seldom see complaints, I often find modules that are designed to a specific problem, and working backwards to a general solution is usually harder than starting from scratch.
In terms of my personal experience, my favorite modules are those that I've subclassed to solve the problems that I frequently encounter.
Are there any obvious shortcomings? The default installation of Google::OAuth is inappropriate for a production environment on a shared server. But since users can easily get started with the default installation, a SECURE INSTALLATION is treated as an advanced topic.
This approach makes sense to me. But I'm still worried that security hawks would disapprove. It's hard to effect the right balance without any feedback.
I haven't ranted about PHP for a while. I still suffer from a perception that PHP progammers are essentially script kiddies, and shared PHP code is delivered in shrink-wrap (metaphorically speaking) with inaccessible internals. It follows, then, that perl is more appropriate for professionals. Most of my modules incorporate this philosophy. But no idea whether this is the same world the rest of you see.
Google's developer site makes no mention of perl. On the basis of the listed API's, developers might assume perl is incompatible with Google integration. It's not likely that my distro is going to have as much influence as all that. But I really hope it will sail as far as possible on its own merit.
Thanks for any shared thoughts and feedback.