Blog – Stack Overflow

One year ago today, we launched Server Fault — our Q&A site for professional sysadmins and IT folks.

To commemorate the occasion, we’ve launched a new Server Fault specific blog at

blog.serverfault.com

You can expect to find blog posts there related to us documenting our own efforts of running and scaling our network of websites — the same sort of stuff you’ve come to expect from the server category on this blog. We want to actively give back to the community by documenting everything we’re doing on the sysadmin front — both by discussing it on the blog, and asking (and answering!) relevant questions about it on Server Fault whenever possible.

In fact, we believe in dogfooding so completely that I’m thrilled to announce we will have not one, but two full-time sysadmins — both hired from the existing Server Fault community of users. The first is Michael Gorsuch.

(the second sysadmin is still being determined, but a little birdie told me might end up being somebody on the first page of the Server Fault user list)

I’d like to also thank our hard working community moderators — Stefan Plattner, Kara Marfia, and Denny Cherry — who so generously contribute their time to keep Server Fault on topic and free of noise.

Moderation is the job of the whole community, in part, but having excellent moderators makes things that much easier.

Server Fault was our first foray into expanding the network, so it was involved in a lot of related “firsts” that we do every time we launch a new site now:

• Coming in March: IT Stack Overflow (Jan 2009)
• IT Stack Overflow Update: Naming Is Hard (March 2009)
• Logo Design Contest for serverfault.com (March 2009)
• Logo Design Contest Winner for serverfault.com (April 2009)
• Server Fault Private Beta Begins (April 2009)
• Server Fault Public Beta Begins (May 2009)

We also did two Server Fault themed podcasts:

• Stack Overflow Podcast #75 with Tom Limoncelli of Everything Sysadmin
• The PowerScripting Podcast

In the next year, we’ll be pursuing a bunch of other ideas to keep Server Fault in tune with the greater sysadmin and IT pro community in as many ways as we can — for example, we promoted this year’s LOPSA conference as a house ad. If you can think of anything else that we should be doing that benefits the greater sysadmin community, please let usk now.

I know Server Fault has saved my bacon with expert answers to questions I’ve asked more than a few times. That’s a testament to everyone who participated over the last year — it’s because of you guys and gals that this thing even works at all. Here’s to many more years of collectively becoming better sysadmins!

Late last week we announced the public beta of the Stack Exchange API — a way to write apps that work with Stack Overflow, Server Fault, Super User, Meta Stack Overflow — and any future Stack Exchange 2.0 sites we launch together as a community.

To ensure that we get lots of feedback on our public API beta, and produce some quality apps that fully exercise the API, we’ve decided to have a little contest. With some totally awesome prizes!

First Prize

top of the line 30″ LCD monitor (either Dell 3008WFP or Apple Cinema HD — whichever we can get to you easiest)

Second Prize

Herman Miller Mirra chair

Third Prize

Intel CULV Netbook, similar to Acer Aspire 1410.

Special Prize (for best library / wrapper)

Adjustable height, motorized GeekDesk

The contest will run for the duration of the public API beta, which we expect to be about 8 weeks, roughly. The winners will be announced when version 1.0 of the API is formally released.

If that sounds appealing to you …

visit stackapps.com and start building awesome stuff with our API!

To make sure we’re all roughly on the same page for this contest, let’s set some ground rules:

• Contest open to every man, woman, and child on planet Earth, except those men, women, or children living in places where contests like this are somehow illegal.
• Only applications and libraries/wrappers listed on the apps tab of stackapps.com are eligible for consideration.
• The application or library/wrapper must be written using our API, and work universally against all of our sites — at least those sites we have made public and have an active API at the time the contest ends.
• While we do have a special prize to recognize the best library/wrapper, to be eligible for the first 3 prizes you must build an application.
• If you live in an area of the world where it is logistically impossible for us to get your prize to you — like, say, because your nearest Herman Miller dealer is 3000 nautical miles away — email us when you win and we’ll make something work.

A few notes on how we’ll be judging this contest:

• The entire Stack Overflow, Inc team will ultimately decide the winners based on order of awesomeness. And lest you think we don’t know awesome when we see it, we built Stack Overflow. I’m just sayin’. (But seriously, please understand that our decisions will be based on a variety of factors, some of which may be entirely subjective.)
• We will look at the number of votes your app or library/wrapper gets on the apps tab of stackapps.com. Doesn’t have to be a zillion votes, but we’d definitely like to see you convince your peers that your app deserves to be in the top (n) of stackapps by popular vote.
• We will look at the number of requests for your API key. Was your application used by a lot of people? Or at least a reasonable amount?
• We will look at your application itself. Does it look cool? Does it work? Is it reasonably documented and understandable? Can other people find it?
• We will look at your application’s code. Is this app a reasonable example of how to write clean code against our API? Is it open source so other programmers can learn from it? Does it accept outside contributions? Being open source isn’t required, but it does get you some extra brownie points.

If you’ve read this far, clearly you’re invested, and you deserve one of those totally awesome prizes. Now go build yourself some apps!

After a whopping 18 month gestation, we’ve finally implemented the Generalist badge.

The Generalist badge works as follows:

Provided non-wiki answers of 15 total score in 20 of the top 40 tags.

The top 40 tag list is surprisingly diverse across the entire Trilogy. Just check out the top 40 tags (as of this writing) on Stack Overflow:

One thing the badge doesn’t say, is that there must be at least 200 questions in all 40 of the top tags before this badge is awarded to anyone. That’s why you won’t see it on meta for a bit longer, or any new sites for about a year. I don’t feel you can accurately measure a generalist until the top tag list settles down.

The intent of this badge is to reward those folks who are answering a variety of questions across multiple skill sets, not spending all their time in the same set of tags.

Now, it is true that the top 40 tags shift over time; for example Android went from #36 to #27 just in the last month. Rest assured, if you earn the Generalist badge, like any other badge in our system — barring exceptional circumstances — it won’t be revoked.

(thanks to balpha and chaos for their assistance in determining how this badge should work.)

Our API private beta is coming to an end, which means it’s time for the API public beta to start.

We’ve set up a dedicated site to support the public beta at …

stackapps.com

It’s called Stack Apps because, well, that’s what it is — a place for applications that run on our “Stack”. You can either find existing apps that are already out there, or learn how to write your own apps.

We’re calling it the Stack Exchange API because our API isn’t tied to Stack Overflow — it’s designed to work on all current Trilogy sites as well as all future Stack Exchange websites we launch, too.

What can you do on Stack Apps?

• Find applications, wrappers, and libraries that use our API — or list your own
• Browse the FAQs to learn how the API works
• Provide feedback on and vote for the applications listed here
• Get an API key
• Ask questions about how the API works
• Tell us about bugs or problems with the API
• Suggest improvements to the API

Fair warning, though, this is still a beta, albeit a public and more stable beta.

1. Version 1 is read only. Coming up with a solid API is hard enough without adding writing and authentication to the mix. For the initial release, it’s a read-only API. We’ll take on the much more challenging problem of writing (and authentication) in v2.
2. The API may change during the public beta. While we expect far, far less breakage than we had during the private beta, the intent of this public beta is to keep improving the API, so there may be changes. We want the API polished up for a formal “locked in” V1.0 release about 2 months from now.
3. If you build to our API, we will support you. We’ll be on Stack Apps daily helping out in any way we can, and listening to all your feedback. If you’re contributing your valuable time building an app on our API, the least we can do is provide a stable platform for you to build on. We plan to have a solid 1.0 API that is reliable and supported for a very long time. That’s a promise.

If you’re interested in applications that run on all current and future Stack-engined based sites, please participate in the public Stack Exchange API beta. Visit Stack Apps, see what you think, and give us your feedback. Help us create an API that doesn’t suck!

We’re bringing back a feature we had to put on ice way back when we launched Server Fault*: automatic account association. Due largely to the vagaries of how Google implemented OpenID, we reluctantly made account association a manual process.

But not any more. As OpenID has matured, we’ve figured out a way to make it happen. So as of today, whenever you log in to any site in our network, which is currently…

• Stack Overflow
• Server Fault
• Super User
• Meta Stack Overflow

… we will automatically find any related accounts you have on any site in our network, and associate them with your account. The only thing you need to do for this to work, is log in using the same OpenID on each site.

But wait! There’s more!

At new user login time, in addition to association, we also:

1. Grant you a +100 reputation bonus on the current site, as long as you have at least +200 reputation on any site in our network. So you’re not treated like a newbie on every new site we launch.
2. Copies your profile from the network site where you have the highest reputation, to the current site.
3. Locate any of your questions that were migrated to this site and make you own them again. It was impossible for you to own them prior to this point, because you didn’t even have an account on the current site!

None of the above is truly new, but it is now fully automatic at login time. And we’re not taking away manual control, either — you can still associate and de-associate manually via the accounts tab on your user page, as before.

As you might imagine, this sort of “identity that follows me from website to website” becomes more and more important as we launch more sites in our network — as planned with Stack Exchange 2.0.

* _{no, I haven’t forgotten it’s almost the one year anniversary of Server Fault going public. We have something very special planned, so stay tuned…}