I know... but what are the best arguments to convince my boss' boss? He is not very technical, but my boss is.

• Other people may find bugs that we can fix before our customer knows that there is a bug to find.
• Other people may even fix the bug they find for us! *
• Other people may add features that we can use! *
• We get a multi-platform test farm for free with CPAN testers.
• The discipline necessary to produce a good CPAN module will improve our code.
• If I'm hit by a bus then there will be other people in the world who know how to use the module, and may even have hacked on the internals.
• We can say that the company "contributes to the open source community". Making ourselves sound nice never hurts.
• It will make me happy. Keeping your employees happy is a good thing.

* Although this doesn't happen that much in my experience.

Update: I forgot the most important question: Why not? Understanding why your boss++ wouldn't want to do it will probably give you the best guidance on how to persuade them otherwise.

I would not use the term "hacked on the internals". May sound suspicious to non-techies with all the bad press about hackers.

Jenda
Always code as if the guy who ends up maintaining your code will be a violent psychopath who knows where you live.
   -- Rick Osborne

Edit by castaway: Closed small tag in signature

Assuming that money would be the motivating factor for your boss not to want to do it, then.. if you explain that by giving your software to the open source community, it will be tested, reviewed, and quite likely improved if necessary. This should save him money.

Of course, open source software allows all of us to save time by not re-inventing the wheel, by sharing resources and knowledge, and in general, working as a team (business buzz word = synergy).

Good luck.

"Assuming that money would be the motivating factor for your boss not to want to do it"

Certainly not. We're in the software business, but we develeop and operate large systems in a specific niche.

We don't want to sell this. It's not our business, and the overhead is too large. There is no money in it for us.

ok, then appeal to his sense of community spirit.

I've had much better luck when we needed specific functionality from Proc::ProcessTable. The module's behaviour on Freebsd significantly differed from the behaviour on Solaris/Linux, and we needed it on all three. A collegue wrote a patch, and the boss immediately okayed sending that patch to the maintainer. Naturaly, since otherwise we'd have to apply that patch each time a new version came out. (Not to mention losing the easy CPAN instalation).

So do you spend company time improving the module, or did you "adopt" it personally? What was the original plan, and how did it turn out?

I have seen very few modules that are "backed" officially by a company. Are there any good examples of that out there? Other than Fotango I mean :)

I adopted File::Tail personally, but I'm allowed to do some work on it on company time (because we use it for company purposes).

As for Proc::ProcessTable the patch was developed on company time, but I don't think my collegue is going to adopt the FreeBSD branch.

I don't work at a software company. Rather, it's a small company that has a web site that happens to require a programmer for many tasks. So it didn't take much convincing to be allowed to release a few module I created at work (such as HTML::Template::Dumper).

One thing you might try is that just by using Perl, you've already benifited enormously from Free Software development. No doubt you've used other CPAN modules to add onto Perl's base functionality, which is another enormous benefit to your company. You're essentially working off perhaps millions of dollers worth of developer time for free. Is it not reasonable to contriute something back into that pool?

"I just want to" isn't a very convincing reason, which is all I can see.

"I just want to" isn't a very convincing reason, which is all I can see.

I've always found it a rather convincing argument myself (assuming that there is no massive cost involved :-) Keeping your employees happy is a good thing when you're in boss-land.

I guess I'm just old fashioned that way :-)

But we can't know if the costs are massive. Plus, it's One More Thing To Think About. Distracting the worker/boss/department from their real job has costs.

xoxo,
Andy