Like the small-town mayor who suddenly finds herself running an entire state, our ambitions for Stack Overflow keep growing. Our original idea of making the Internet a better place to get expert answers to your programming questions suddenly seemed too small. Programming questions? We asked. Why just programming questions? Why not every question under the sun? And who says we can’t run for Vice President of the United States of America?
We tried making our software available as a hosted white label product called Stack Exchange. We thought that other people would create awesome sites on every imaginable topic. Some people did (yay!), but it wasn’t the flood of high quality sites we were hoping for.
So we’re making a few changes. Briefly:
Stack Exchange will now be free.
We’re changing the way that new Stack Exchange sites are created to move to a more democratic, community process.
The content of these new, community-created Stack Exchange sites will be publicly owned under a Creative Commons license, instead of being owned by individuals or businesses.
If you’ve already created a Stack Exchange site, be sure to read the announcement in more detail to hear about our transition plan. Don’t be alarmed; we’d never do anything to mess with Stack Exchange sites that are already working.
Getting better answers
As programmers, we’ve gotten used to the clean, fast, reliable answers that you get from Stack Overflow, so whenever we try to get an answer to a tax question, or a Siberian Husky question, or an iPhone question, it’s incredibly frustrating to find old conversations, trapped in forum and discussion software, instead of answers. Forums are optimized for conversation and shooting the breeze, not for getting answers, so they suck when you actually need some information.
During the last week of meetings, we’ve been talking about our company ideals, our core values, and our core goal. We came up with a new, very ambitious company goal:
Make the Internet a better place to get expert answers to your questions.
That’s a pretty big task, but we know that the great software we created is up to it.
Charging for Stack Exchange hasn’t exactly worked
Our first idea was Stack Exchange… call it Stack Exchange 1.0. We thought we’d make our software available on a SaaS basis, a.k.a. “white label Stack Overflow,” so that anyone could start a site on a new topic in exchange for money.
When we launched Stack Exchange, we imagined thousands of sites would start to sprout up on every possible topic. Harley Davidson belt buckles, mathematics, unicorns, you name it.
However, by setting a price ($129-$5000/month, depending on traffic) to cover the cost of servers and bandwidth, we discouraged a lot of people from making sites that might have been great. And by allowing anyone with a credit card to make a site, we got a lot of ghost-town sites that nobody visited. We also got a lot of duplication: multiple sites on the same topic, competing for the same people and preventing one other from hitting critical mass.
Bottom line, it just wasn’t working. We’ve been in beta for half a year now, and we only have a handful of sites that get enough traffic to provide quality, timely answers to difficult questions.
Jeff and I got the teams together in New York to figure out a better way to do it. We had been getting great ideas from smart people all over the world, and we sought advice from some of the people we admired the most. We spent several long days hashing out the issues, figuring out what our real goals were, and trying to find a better way to spread the Stack Overflow way of doing things to the world.
Essentially, we decided that Stack Exchange was failing because:
Only people with money to burn or a business plan could create sites.
Those people didn’t necessarily have the ability to bring an audience.
In other words, it was simultaneously too easy and too hard to create a new, working Stack Exchange site.
Instead of trying to build hundreds of communities from scratch, we decided just to ask our existing audience, “What else do you want to talk about?” We’ve got 6.7 million people visiting every month; they must have something else they want to talk about besides programming and unicorns.
Fortunately, a recent road trip uncovered a long list of investors who believed in our mission, so we were able to raise enough money to make Stack Exchange absolutely free. (The details of that investment are not quite ready to be announced, but we’ll let you know as soon as they are).
So now we’ve got the audience, and we don’t need the money, so all we need is ideas.
Want to create a Stack Exchange community? Propose it! If your idea gets sufficient support from a community of dedicated users, then it gets created. It’s that simple.
The New Stack Exchange Site Creation Process
> > > > _I’m just a bill. Yes, I’m only a bill. And I’m sitting here on Capitol Hill. Well, it’s a long, long journey To the capital city. It’s a long, long wait While I’m sitting in committee, But I know I’ll be a law someday At least I hope and pray that I will, But today I am still just a bill. —I’m Just a Bill_ (from Schoolhouse Rock) Dave Frishberg
A Stack Exchange Q&A site only works when it has critical mass: enough people have to go there every hour so that questions get answered. A big part of our new process is to make sure that a site doesn’t get created until we have some reason to believe that it’s going to get that critical number of people showing up to make it work.
Our new system was inspired by the way that new Usenet newsgroups were set up in the 1980s. Unlike the free-for-all in alt., where you had binaries, unicorns, and alt.swedish.chef.bork.bork.bork, the mainstream newsgroups (comp., rec., talk., etc.) mostly consisted of serious, qualified, relevant sites on every topic imaginable.
That system worked pretty well. Usenet rapidly expanded from two nodes to thousands. And it serves as the inspiration for our new, modernized process.
Like our friend the bill on Capitol Hill, every new site has to survive a rigorous vetting process before it gets created:
1) The Discussion Phase
On a new meta site, people will gather to discuss potential ideas for new Stack Exchange sites. The goal of the conversation is to beat around some ideas for what kinds of sites might work, and ultimately, to collaboratively create a detailed proposal.
2) The Proposal Phase
A proposal consists of these four things:
The topic—what’s the site about?
The target audience—who do we expect will visit?
A list of five exemplary on-topic questions
A list of five exemplary off-topic questions
The Discuss links lead to pages where anyone can propose, discuss, and vote on the parts of the proposal itself. Over time, a collaborative proposal emerges.
Anybody can vote on proposals. If a proposal gets enough votes, the site moves to the commitment phase.
3) The Commitment Phase
During this phase, people who are interested in a potential site are asked to electronically “sign” a commitment to help make the site a success. They are committing publicly to participate actively in the site, by asking questions, answering questions, developing a system of tags, and generally helping the site get off the ground.
Each individual person can only commit to one site per month.
As people commit, signing the e-petition, a thermometer will go up showing the level of commitment. When the thermometer gets to 100%, the site moves into beta.
How does this thermometer work?
Intuitively, if Jon Skeet says that he’ll participate in the Sock Puppet Stack Exchange, that commitment is a better sign that the site will succeed than if we get a commitment from a random Internet user who has never participated in Stack Overflow. Sure, they’re both wonderful people, I’m sure, but Jon Skeet has proven that he likes to participate in Stack Overflow so it’s a good bet that he’ll participate in SockExchange, too.
Furthermore, we want to make sure that each new site has enough users who already grok our system of badges, answers, questions, tags, voting, community wiki, reputation, etc., so that the site gets off to a good start.
That’s why the commitment thermometer will not be precisely a one-person, one-vote petition. Instead, we’re going to require a selection of _existing _users with certain badges and reputation that proves that they’ll participate. For instance (and I’m making these numbers up), we might require that a site get at least 100 commitments from people with the Teacher badge, at least 20 from people with the Enthusiast badge, and at least 50 from people with a reputation of 1000 or more on some of our sites.
Over time, we’ll adjust the thresholds upwards or downwards as we discover what it really takes to get a site off the ground. For example, if we discover that new sites are getting created but they’re not using tagging correctly, we could raise the requirement for the number of Organizer or Taxonomist badges.
4) The closed beta
If a site gets to 100% commitment, we’ll email everyone who committed and notify them when the closed beta will begin. During this closed beta, they’ll be expected to seed the site with enough interesting questions, answers, tags, and a site-specific FAQ. They’ll appoint temporary moderators and publicize the site.
5) The open beta
During the open beta, the site will be open to the public at a temporary domain name (topic.StackExchange.com). The site will be all black and white, and include an animated-GIF “under construction” triangle showing men at work, from 1996.
This phase will last between 60 and 90 days. At the end of that period, the site will need to reach a minimum critical mass to continue and move on to full citizenship.
6) Full citizenship
We’ll set strict criteria (number of new questions per day, number of registered users, percentage of answered questions, number of people who vote, etc.) to define a site that we consider to be successful. If a site meets those criteria for 90 days, it graduates to full citizenship.
A citizen site gets its own top-level domain, chosen by its community. There are elections for moderators, we’ll have a graphic designer make the site look great, and let the community pick a logo.
If a site does not have enough activity at the end of 90 days, it will be closed down. Any existing Q&A will be archived and made available for download, but the site itself will not remain live. Small, unhealthy sites do nothing but draw traffic away from other sites, splitting audiences, so we don’t want to keep them around.
We don’t have any yet! Although we’ve figured out the basic skeleton of the new site creation process, we’re depending on you, the public, to refine these ideas and make them great. We’ll be talking to you at meta.stackexchange.com about this new site creation process and looking for your feedback, ideas, and suggestions for how to make Stack Exchange even better.
Q: Who owns the content on the new sites?
You do! Unlike previous Stack Exchange sites, the content (questions and answers) of the new sites will be owned by the community and licensed under Creative Commons. We will provide regular data dumps containing all non-private data from each site, like we do with Stack Overflow.
The sites themselves will be owned and operated by Stack Overflow.
Q: Who pays for the new sites?
Q: What is the plan to make money from this all?
We believe that we can have a bigger positive impact on the world if we are self sustaining and not dependent on the kindness of strangers. We do not want to hard-code our revenue model too early. We believe that if our platform creates value for a large number of users, we will have opportunities to make money. Ideally those opportunities will not just make us self supporting, they will also make the site better. We are thrilled that we have patient investors who will support us and are prepared to allow a “native” revenue model to emerge organically as the site grows.
Q: What happens with existing Stack Exchange sites?
We don’t want to harm any communities that have already successfully gotten off the ground. This harks back to our corporate goal:
Make the Internet a better place to get expert answers to your questions.
Community is hard to build, and we want to work with you to preserve it if you’ve already done that with Stack Exchange. If we closed down or competed with the existing, successful Stack Exchange sites, that would conflict with our goals.
Existing Stack Exchange sites will be kept open, under existing rules, for at least three months, and at least one year if you have an active site (defined as ten or more active users per day).
You will not have to pay for these sites, ever.
We’ll give you at least 3 months notice before shutting down any site.
We’ll always make your data available for download.
If your site remains very active, we’d love to work with you to migrate it to the new, community-owned Stack Exchange platform. That would be the best thing that could happen to a Stack Exchange 1.0 site, in our opinion: that way your site can take advantage of our existing resources and expansive community.
Q: How do I know how long my Stack Exchange site will remain open?
Log on to your site as an administrator, click admin, and go to the account tab.
Q: What if a new, community-created site competes with my existing, old-rules Stack Exchange?
If your existing Stack Exchange site already has developed a substantial community, we’ll encourage people to go there, rather than creating yet another Stack Exchange on the same topic. Once again, our goal is to make the Internet a better place to get expert answers to your questions. Competing with existing sites that do a great job of that is not in our mission.
However, if your Stack Exchange site does not have substantial traffic, we reserve the right to create a new site on the same topic.
Q: How do I contact you?
Just email our community coordinator, Robert Cartaino (email@example.com).
Q: I was thinking of making a Stack Exchange site, but haven’t yet done so.
You will have to use the new process to propose and create the site.
Q: I am interested in licensing the Stack Overflow source code.
As a part of our new focus on serving large, internet-sized communities, we are no longer offering the Stack Overflow software.
Q: I am interested in creating a private, internal Stack Exchange inside my organization.
We no longer offer such a product.
Q: I want to create a Stack Exchange to make a support site for my product.
You will have to use the new process to propose and create the site.
Q: I want to make a new Stack Exchange site that fits in with the look of my main website.
Although the existing Stack Exchange framework allows this, the new community process is for making public, community sites, not private sections of existing websites.
Q: Is there a difference between the Stack Exchange code base and the Stack Overflow code base?
The Stack Exchange and Stack Overflow sites run on separate, but very similar, versions of the original software. When Stack Exchange started, they forked a copy of Stack Overflow, then each team continued to develop and improve their respective platforms in separate development efforts.
Currently, the combined Stack Exchange and Stack Overflow teams are in the process of merging the two code bases to take advantage of the best features of the two systems. Sites created under the Stack Exchange 2.0 model will take advantage of the new code base, as will Stack Overflow, Server Fault, and Super User.
Legacy Stack Exchange sites will remain on the Stack Exchange 1.0 platform. No further enhancements are planned for that platform, except for urgent bug fixes.
Q: Why is the plan to close down sites that don’t get enough traffic?
This harks back to our corporate goal to “make the Internet a better place to get expert answers to your questions.” A ghost town, without traffic, does not get people answers, but it does draw a few people away from other sites that might do so. We do not believe that the Internet benefits from putting up placeholder sites with negligible traffic that do not attract high quality communities. And we want the Stack Exchange brand to be synonymous with great community Q&A sites, even if we don’t necessarily cover every topic under the sun.
Q: When is all this coming?
We’re working as fast as we can! We hope to start the new process of proposing sites within four weeks.
Q: I have more questions!