Today Ask Ubuntu celebrates it’s first year in existence. Though publicly launched on the eve of 10.10.10, the site went into private beta one year ago today. At the time Evan Dandrea had seen Area 51 and had proposed an Ubuntu Stack Exchange.
I was just as confused as everyone else. It wasn’t until I spoke with Evan at Debconf 10 where he explained it to me. In fact, you can probably call this the first time the idea of mercilessly removing horrible content from Ubuntu properties got started in my brain.
His gist was this; like with code, there is just no replacement for peer reviewed content that focuses on quality. That’s basically the mission of the site. So with that, I dove in head first and decided that I was going to help will this site into existence.
It took us a while (from end of July until October actually) to find our feet. That’s where we honed down our FAQ, what was ontopic and what wasn’t, we narrowly focused what we would be good at, answering people’s questions. We would heavily leverage the existing wiki documentation, bug reports redirected to launchpad, discussion moved to the forums or IRC. No distractions from the mission, ask a question, get an answer; the rest is just furniture.
So how are we doing?
We are currently the 4th largest Stack Exchange according to traffic (behind the original trilogy of Stack Overflow, Super User, and Server Fault). Here’s where you can sort the criteria. During the release of 11.04 we hit around 45k traffic, which is about 50% of Server Fault’s traffic (in less than a year!).
While all that is fine and good, what about user engagement? Well, currently we have about 19,000 registered users, here’s the breakdown by reputation. (Reputation is a measure of how much other user’s trust you).
Stack Exchanges are unique in that priviledges to run the site are earned by the votes from your peers (which is measured in reputation). The more reputation you earn, the more rights you have to edit the content on the site. A user with 20,000 reputation is basically a moderator, but the important one to me is 2,000. This is the level where you no longer need to have someone peer review every edit, and editing is how content stays fresh and relevant.
I consider everyone with over 2,000 reputation to be a heavily engaged user on AU, someone who has taken a personal interest in making the site succeed. We have 85 people with over 2,000 reputation, meaning we have 85 people continuously improving the site at a high engagement level.
Surprisingly, you’ll see over 18,000 people mostly just consuming the content. This is the userbase we serve the most, but you can see how a relatively small group of people can make something good happen.
And what about the end result? So far our accepted answer rate sits at 81% (which is about the same as the original trilogy sites). We’re constantly looking for ways to improve quality; I sometimes yearn for the day when we could answer 95% of the questions, but hey, with great size comes great craziness of unanswerable questions.
I have a ton of people to thank, you all know who you are, the first pile numbers at about 85 people. We’ve all been putting in crazy hours to make this work. For me personally it’s been an about one hour before work, most of lunch, and multiple hours after work. (Spouses getting Ph.D’s are good for internet participation!)
Now that the first “ooh ahh” year is out of the way, the next comes the grinding run into the playoffs. There’s no doubt in my mind we can
eclipse Server Fault provide better quality for end users, it’s just a matter of time and workin’ hard.
If you’re feeling intimidated by it just dive in and get started, we’re friendly, and remember that reputation is a measure of trust, not exactly skill (where else would I outnumber Kees Cook in anything by 16,000 units?)
You can earn reputation by asking good questions, submitting edits to make content relevant for today, or by answering questions. Once you have the 15 rep required to vote you can very easily determine the quality of the site by just voting a few times a day. The quality of the content is determined by it’s people, so I’m looking for experts, people who want to be experts, beginners, medium level, and whoever to dive in and help someone out.
Here’s to another year!