Default offerings, target audiences, and the future of Fedora
Ever since I started contributing to the Fedora Project I’ve always loved the work (still do!). The operating system is flexible (just a bunch of puzzle pieces, really) and that flexibility allows people to build pretty much any type of system they need to get their work done. Everything being licensed under a free and open source license makes all those puzzle pieces very easy to work with as well. In short, what I see in Fedora is what I’ve come to expect from all software solutions.
An almost common topic on the Fedora Board is the “target audience”. I dread these conversations because the conversation has never made much sense to me. To me, Fedora is flexible enough to be whatever anyone wants it to be. If you need to use Fedora on a server to serve up web pages or email it can do that. Need it to be your primary operating system on your laptop? Yep, it does that well too. So why do we spend so much time, energy, and effort on the “target audience”? I have some theories, which I won’t discuss here, but in short we really shouldn’t be trying to pidgin-hole ourselves when our flexibility is one of our major selling points.
For years now we’ve made some assumptions about our “target audience” and what that “target audience” wants. The assumption of what they want ends up being the default offering and everything else becomes a Spin. I’ve been on the Board now for a year (just entering my second term) and I’ve yet to hear any evidence that says that this is what our “target audience” wants and why we need to push the hard work of other SIGs down to second-class citizen level. It doesn’t matter that Fedora runs on tablets, headless ARM devices, servers, cloud environments, and pretty much anywhere else people can think of putting it. We still think that everyone wants the GNOME Desktop Environment with all the bells and whistles for every installation or that you can just manually add and remove packages or by use a kickstarter file at installation to fix these issues.
So before it gets brought up in the next Board meeting (taking place in just over twelve hours from now) I figured I’d take the opportunity to explain my thoughts on the idea. Why don’t we let the SIGs determine what’s important and let them build the releases that create Fedora offerings? The Desktop SIG could put together what they wanted as could a Server SIG, ARM SIG, cloud SIG, etc. The new download page could display several desktop environment downloads, server ISOs, cloud images, and perhaps several ARM images to allow specific hardware goals to be more easily met. The Project could focus more on the core packages and then help the SIGs develop their releases. This isn’t exactly different from what we already have, as far as logical people layouts are concerned. Today there are people who work under several SIGs and sometimes the work they do also falls across those lines of interest. The biggest change would be that Fedora would no longer be only known for its desktop OS but rather it would be known for its flexible OS that is pre-built for many situations making it easier for people to adopt it for their use.
Lets add another foundation to our current four. Let’s make it five with Flexible.