Stories
Slash Boxes
Comments
NOTE: use Perl; is on undef hiatus. You can read content, but you can't post it. More info will be forthcoming forthcomingly.

All the Perl that's Practical to Extract and Report

The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More | Login | Reply
Loading... please wait.
  • I am kind of sceptical when it comes to people doing a lot of CPAN distributions. There's a pretty long way from an idea of a possible CPAN module to its implementation. For a reasonably complex module, this can take months with all the documentation and tests, interface-considerations etc. And that's only the first step. It has to be maintained afterwards

    Essentially I have a few unfinished modules lying around. My most ambitious project was probably porting the Aalib to Perl. I was mostly done with it (even had ways of writing new drivers in Perl). But then I got scared and thought about its implications if people ever used the module.

    Also, the majority of my modules is XS. Whenever an external C library is involved, I chose to include it in the distribution. That way, I don't necessarily have to make a new distribution if the C library gets updated in a possibly incompatible way.

    Furthermore, I think that a module should be properly checked on different operating systems. Before ActiveState started to bundle up new CPAN additions automatically, I was doing that myself which was annoying.

    This all means: maintaining five widely spread modules can be a fulltime job. Uploading one new module after the other is therefore insane. Whenever CPAN authors state in their documentation that they might not have enough time to fix bugs or whatever (by pointing to their vast amount of other modules), it's time for them to reconsider their policy.
    • My reply is remarkably old, so I apologize, but I feel I must clarify my original statements.

      Basically, I'm not spinning out CPAN modules for the hell of it; I can't tolerate "Hero Coding". On the other hand though, my modules were very small and, to me, seemingly inconsequential ones. Maintaining them has been very easy, mainly because there isn't much to go wrong.

      My original intention of this journal entry was to say that most of the code I've written could have been abstracted to a CPAN module, but w
      --
      -man Michael A Nachbaur