Somewhat related to my last post, here’s a Infoworld post that contains a lot of vehemance about OSS “freeloaders” and so forth.
These groups that forsee this don’t have a “freeloader” problem, they have an untapped opportunity. They have a large contributor (and advocacy and testing) pool that they have not tried hard enough to take advantage of. They still believe that everything is centered around themselves.
It’s easy to make such wrongful comments when you didn’t try to build proper communities around projects, because you didn’t know how, and you didn’t work at it. Blaming the community though for one’s inability to build up that collaborative environment? That’s just a horrible thing for anyone to do. More reason to know what you have tapped, and what you haven’t, and where the opportunities are.
If you’re not doing that, you are a closed source business that accidentally gave away code thinking it would make everyone beat a path to your door. That’s the “open source as a bullet point” on your slides approach — not being a real open source project.
Don’t blame your users for your own failures to build community — instead look at what you still have an opportunity to build and what you need to do to really be an open project. Without users your project does not matter. If you haven’t found people that are interested in working on your project, this is something you can fix …. or perhaps the app itself isn’t that interesting. This should also be fixable.
In either way, blaming the coal mine canary for dying seems very short sighted.
I’ve been following your “series” on open source communities quite keenly.
I’d be interested if you have anything to say about the importance of IRC (ie. chat) versus mailing lists. Naturally I have no hard figures, but I’ve found that IRC was useful in allowing the fedora-mingw community to attract new contributors. Seems that people don’t want to commit to subscribing to the mailing list [even though they don't need to subscribe in order to post], but will sit in an IRC channel.
On the other hand, IRC is problematic. It demands attention. It’s not easily searchable, and many are not archived by Google unless we take extra steps. So it doesn’t build a knowledge-base.
There is also the question of whether we should set up topic-specific IRC channels, or piggyback on existing ones (eg. the general #virt channel on OFTC).
What do you think?
Hi!
Yeah, “series” not intended. I should slow down posting
I can mostly only speak for my channels and a few others… I do have a hard time following other channels too closely (I’d love to watch more of fedora-devel and epel, but ENOTIME).
Personally I love it, but you’re right it does take a lot of attention. Lately there have been days where I’ve /not/ payed too good attention (so I could get something implemented), and let other users answer questions, which seems to work… provided you have that base built up, but I always try to read the scrollbacks. If you try to pay attention 24-7 it will kill you, and it can be distracting, no doubt. But you get all sorts of interesting people dropping in, all sorts of new ideas, and I couldn’t give that up.
In my case, I think there’s a lot of folks that /do/ subscribe to the MLs, but it does seem you get folks that aren’t there hop on IRC, which eventually join up. They can be made to mix. I seem to do a lot of asking for people to join the mailing list, and occasionally if someone has a neat idea, I ask them to join the ML. That seems to work.
It sort of feels like IRC support is kind of like a band going on tour and working hard to get the music/name out there, and mailing lists are more like selling albums. Time in the studio is also required!
For talking with the people who are really involved though, It think IRC is great, but I find myself repeating good ideas and discussions on the mailing list, because that has a wider audience. This fixes the Google problem.
“Bob had a good idea on IRC just now…” … which also makes folks want to join IRC!
There’s a lot of folks who can’t get IRC at work, or don’t have time to follow it, so all the ideas/planning/new-stuff needs to be on the mailing list too.
I’d guess some projects could just get by with a mailing list, but for debugging and stuff, or people working out ideas in real time, I like IRC a good bit.
Topic-specific stuff? Depends how specific I guess. I think OFTC was an interesting point — most users (and potential developers) are on Freenode, so to really get an interest in virt (and build the love), I think a Freenode presence would be good to have.