Short Version:
Some who wrote a pretty good module is too scared to publish. Should I publish for him?
Long Version:
A few entries ago, I mentioned that I was working on a C extension for Ruby for the 'df' command. I posted an initial bit of code out in comp.lang.ruby, and a guy named Mike Hall jumped in and eventually came up with a working solution fairly quickly (and it was a different approach than what I had taken, so it was a different design as well).
We exchanged a few emails, I was a guinea pig for Solaris testing, and he did the Linux testing. Soon, he had version .30, a version I definitely considered ready for release (it needs work to support BSD and Windows, but that can be added later). However, here's where the dilemma began.
I asked him to publish it to the Ruby Application Archive, to which he responded something like, "ooh...scary". I took that to mean he was too nervous about publishing his work. It's still not out on the archive.
I can understand a bit of apprehension about putting your work out there for all to see. I mean, most of us probably felt that in the moments before we uploaded our first module to CPAN. You want the module to be *perfect*, and you dread that some idiot is gonna email you with a line like, "What is this piece of junk you dared to post to the Archive? Who do you think you are? Come back when you've reached Guru status". Of course, that doesn't happen, and people are generally very polite about pointing out bugs or asking for enhancements.
However, I have already sent Mike reasons for uploading his module to the archive - it's useful, others (besides myself) have requested it in the newsgroup and it works just fine. It's a handy sys-admin toy. Not everyone is going to come out and ask for it in the newsgroup, but if they see it's available, there's a good chance they'll download it. The alternative is to troll around comp.lang.ruby and send an email every time someone *does* ask for it. And what happens if, God forbid, something should happen to you Mike? That work, and any work he may be doing on it now, will be lost.
So, here is my dilemma. If I can't convince him to upload it to the archive, should I upload it to the archive myself? Obviously, he would be listed as the author, so it's not a question of taking credit (though I'd like to think I was a contributor at least). I'm inclined to say no and keep trying to convince him for a while, but I thought I'd see what others thought.
Any ideas on how I can calm his fears and get him to publish? I haven't heard back from him after I sent an email encouraging him to publish. What happens if I never convince him?
Opinions welcome.
UPDATE 2-May-2002 - Looks like he posted it today. Hooray!
I think you should let him make the final decision to publish. He'll feel more committed to it, then. You've sent him private encouragement; I'd put a little bit of public encouragement on the newsgroup. Take opportunities to mention he has the module (with parenthetical statements like, (Mike, when are you going to release?)).
Make sure he's fully bought into the licensing scheme. I don't think anyone should release code if they're not clear on allowing everyone to do what they want with it. (Sell it on CD-ROM, use it in a corporation, make modifications without returning them to the author, and other things people sometimes don't like.)
If you do it for him, he might conceivably get upset or feel pressured. You could volunteer to walk him through the process, though.
I've never released anything publicly (except some code here in my journal), but I've hung around long enough to know what to do and to have no reservations when the time comes. Actually, I've got one three line patch in Perl, but that's it.
