Related Stack Overflow Blog post:

Domain Names: The Wrong Question

Apparently, the SE team is doing a complete 180° turn and going from this:

American audiences, generally, don’t trust series. They tend to believe that they want the best baking book, period, not whatever baking book comes in yellow. When they see a shelf full of yellow dummy books, they mostly say, “yeah, a bunch of second-best books.

Joel Spolsky, before NTI's commitment phase begun

...to this:

y'know, the dummies series .. it kinda works.

Joel Spolsky, "leaked" to the public by Jeff after NTI graduated

In a nutshell, the team now wants to focus on the engine brand, instead of the site brand.

As a result, nothingtoinstall.com is currently a redirect to webapps.stackexchange.com. The announcement blog post also has been changed accordingly.

If you have a stake in Stack Exchange betas, you probably better make yourself heard in this question on meta.nothingtoinstall.com meta.webapps.stackexchange.com.

  • I tried to keep the above text opinion-free. Don't reply to this, reply to the meta nti thread. – badp Oct 2 '10 at 16:41
  • 1
    Previously ( backup ), previously, [ previously.](joelonsoftware.com/items/2010/09/30.html) – badp Oct 2 '10 at 17:35
  • 3
    "reply to the meta nti thread" — how about those of us who don't want to create a WA account but find the discussion relevant? – kennytm Oct 2 '10 at 17:52
  • 2
    @Kenny It's not my choice to hold the discussion there. I feel it is network relevant however. – badp Oct 2 '10 at 18:00
  • @Kenny Because creating an account is so difficult. – TheLQ Oct 2 '10 at 19:08
  • 5
    @TheLQ: Yes it's easy, but what's the point for creating an account on a site that I won't use in the future? – kennytm Oct 2 '10 at 19:10
  • 2
    Since Food and Cooking has been mentioned multiple times as the one instance where a good name was proposed, I asked about what will happen if/when it graduates. – Pops Oct 3 '10 at 4:58
  • 1
    This will be great for Gaming, we are yet to come up with a good name – juan Oct 3 '10 at 15:34
  • 2
    Why isn't that being discussed here? – juan Oct 3 '10 at 16:04
  • Voting to close to prevent further answers here. My purpose was not to create a second place to post opinions to. – badp Oct 4 '10 at 5:53
  • 9
    Yeah, I must admit I don't understand why the discussion didn't happen here. It's the kind of 'weird' thing which makes for conspiracy theories about the team's real motivations. One feels mildly manipulated. – Benjol Oct 4 '10 at 6:02
  • This has also produced a Pro Webmasters question. – Pops Oct 4 '10 at 14:51
  • 3
    perhaps we're overthinking domain names because we want them to be too much. I think nti.com would be much better than nothingtoinstall.com. maybe it doesn't tell you anything about the site, but who cares. of course, this comment is about 85% off-topic in this thread, so i'll stop. – Kip Oct 4 '10 at 21:36

14 Answers 14


We did a 180° turn on the basis of coming up with a name for the site.

We have been basing our site names on the brutal, cut-throat availability of the .com domain market. So we end up settling on the “least worst” names based on the meager domain name options.

Instead, we want sites to come up with AWESOME NAMES—"awesome" being the functional part of the equation… without the crazy must-be-an-available-domain constraints.

You can read more in my latest blog post:

Domain Names: The Wrong Question

  • 6
    @Robert: this is not about name. It doesn't matter what the site is named, despite whatever some negative feedback about breaking up the network into smaller domain you might have got. It is about just "doing stuff" as one user put it, without consulting with the community or any announcement. This "awesome" bs just need to stop too. – SilentGhost Oct 5 '10 at 16:17
  • 2
    Um, did you even bother to read the blog post this answer is about? A huge percentage is about exactly the point your are railing against. – Robert Cartaino Oct 5 '10 at 18:07
  • 3
    Saying that you're consulting community is not the same as actually doing it. And yes, to answer your question, I did bother. If by the "point I'm railing against" you meant the amount of participation you're allowing community in the decision-making, there's preciously little about that in the blog post. The whole mess reeks of the Dilbert-land. – SilentGhost Oct 5 '10 at 18:39
  • 7
    I don't have a time machine. If you read that blog post and it didn't say to you "you're right; I dropped the ball", my writing failed miserably. – Robert Cartaino Oct 5 '10 at 19:01
  • 4
    @Robert - I've read your post, it's a good one and appreciated on a meta level, but despite your honesty it nonetheless appears a bit smoke and mirrors still, because the actual cause for the hasty reversal has not been disclosed yet: the naming issue aside (granted ...), you nebulously talk about circumstances didn’t give us much choice, negative feedback and letting problems grow exponentially - Which circumstances took away our freedom? Negative feedback by whom? Which problems could grow exponentially? Why don't you just tell us what exactly triggered this within hours? ... – Steffen Opel Oct 6 '10 at 0:18
  • 4
    ... I sincerely hope you not only want to represent the community to the company, not vice versa, but are in fact empowered and enabled to do so: please figure this whole issue out for us and fill the missing links, or, at the very least, provide more meta information on why it is handled like a trade secret ... I think it is more than obvious that the majority of users is perfectly willing and able to agree to disagree on a topic at some point, as long as arguments, influences, policies and processes are handled transparently (ideally in the first place ;) - Thanks a lot :) – Steffen Opel Oct 6 '10 at 0:18
  • 1
    (1) The Circumstances: The domain had just switched. It's exponentially harder to switch back when it becomes established. (2) Negative feedback: From pundits of top respected Internet sites; not my conversation so I cannot recall whom. (3) Timing: Friday afternoon meeting brings everyone's issues and opinions together. Discussion about the domain-splitting critique triggered the alarm. (4) Is Community Coordinator empowered to represent community: That's my job description and reason to be here. None of this is a trade secret; a blog post isn't a novella and it suited 99.9% of the audience. – Robert Cartaino Oct 6 '10 at 14:44
  • 3
    @Robert - thanks for following up. So after month of planning, apparently without consulting experts on the subject, the going public of NTI motivated someone from some top site criticizing something concerning domain-splitting, triggering an alarm in the team meeting and, despite the mess, you neither recall who nor what has been the argument? Wow, I don't know what to say anymore ... If none of this is a trade secret, you might just ask, don't you think? Well, maybe someday, someone will somehow shed more light on this still (e.g. somewhere) ... – Steffen Opel Oct 6 '10 at 16:12
  • Aaaaand there goes another 180-degree-turn; I just wish each of the announcements wasn't so reminiscent of "Oceania has always been at war with Eastasia." blog.stackoverflow.com/2012/06/welcome-to-the-arqade – Piskvor left the building Jun 27 '12 at 19:11

I've tried to refactor @Aarobot's answer so it is less angry and less personal-attack-ey, as I feel it has a lot of relevant points in it and people should be able to vote on them.

Since it's more a recycling of someone else's answer I'm marking this Community Wiki.

The Community Member: Why are they doing all of this?

All of us here - all of us on the SE betas and Area 51 and even Stack Overflow and the trilogy sites to an extent - we're like volunteers.

While Stack Exchange is composed of autonomous communities, the actual relationship of community members with the team is similar to that of volunteers; we do our thing because (a) we get a hearty pat on the back, (b) we get the occasional perks, and (c) we enjoy the contribution we're making.

But try this: describe the concept of what Stack Exchange community members do to a non-community member. Their response may surprise you if it's not "Tell me more", nor "That sounds too technical" but instead "Why are they doing all of this?".

Community Members want to be part of a knowledgeable community

Each of us wants to be part of a knowledgeable community, but the only other venues for that are clunky discussion forums. The success of Stack Overflow (and all related sites) however had less to do with a new kind of Q&A site, and much more to do with a new kind of on-line community.

Now we're all - community members and team - working to create other similar communities. Community members were originally under the impression that they would get to contribute to and choose community identities, as this is what almost every healthy, functioning community does. An identity is central to any community, and its name is fundamentally tied to its identity.

It should be noted that, to anyone outside programming circles, "Stack Exchange" might as well be 102-47th Street. Individual sites become not so much communities as apartments in a nameless high-rise building - and that high-rise building has a strong identity build around a community of programmers.

How is a new site (Cooking and Whatever for example) going to establish an identity and a reputation as a separate community not standing on the shoulders of Stack Overflow if it is branded to seem like it is? Does it stand a chance to be perceived as "The place where cooks and whateverers ask questions" as opposed to "The place where StackOverflowers ask cooking questions"?

Contributing to communities has become discouraging

Many of us out there are trying to contribute to building communities which will stand of their own, but the actions of the team have at times been discouraging:

→ Note: Can someone add links/citations to these?

Now, communities are being denied the privilege of a name, which in turn hampers their identity.

A big part of the problem is the following:

We don't care about Stack Exchange
We care about our individual communities.

The average community member won't care if every single proposal other than the ones they're actively involved in fails. None of us care about somebody else's community.

We don't care about the Stack Exchange brand. We don't care how many or how few sites there are, and while the success of many of those sites could impact the success of our communities, we are nonetheless indifferent to them.

We are donating ample amounts of our personal time to these proposals/betas for one reason and one reason only: So that we can be part of a community that we're proud of.

What we don't want is to be a nameless, faceless portal page that's part of some conglomerate which only people in a specific industry have heard of. We'd rather at least try to make sites and communities which can stand on their own.

It seems to be that at least on WA, the community has spoken quite loudly in agreement with this principle.

If the team wants these communities to continue volunteering their time for Stack Exchange, then the communities need to feel more respected.

While it's the team's company and the team doesn't have to do anything we say, it seems like there are a variety of obstacles discouraging us from forming communities. There are the bizarre voting rules on Area 51, which people don't "get"; there are extra-long beta periods which give the impression that that the team doesn't trust the community to keep the axles greased; there's the fact that when and how communities get their own names is ambiguous, almost suspicious, and there's an uneasy feeling of contempt and elitism which the community members sometimes get from the team. Perhaps above all else, if the team wants to crowdsource, they need to try a little more to demonstrate that they don't feel contempt towards the crowd.

"Community based" businesses should never give such a strong 180º turn without any explanation. This is a trust breaker. The subsequent turmoil should be a strong hint that the community management process has failed and should be improved at least by being more open.

If the team believes this is the right choice, and genuinely want support, then show us; if the team's passion for the sites and the communities leads them to believe this is best, there is no doubt they can show us, because people respond to shared passion. The fact that we are here at all proves this to be true.

If the team wants complete control over the entire process - maybe its the only thing which will work - then they should say so. Creating whatever sites they want, with whatever names and branding they want, and seeing how many people join and contribute is what was done with Stack Overflow, and it worked.

But right now there's a bunch of people (volunteers?) who feel like they have been lied to about their control over their communities, and that can be a serious risk to the Stack Exchange business.

  • @jjnguy - thank you for adding a link/citation. – Richard JP Le Guen Oct 4 '10 at 16:50
  • you are welcome. That was the only good one I could find late last night. – jjnguy Oct 4 '10 at 17:03

As usually, the right way to go should be somewhere in the middle.

As I see it, the only problem is that the communities weren't able to come up with good, clear-enough and not-very-disputed names - exception being the cooking site, with Seasoned Advice.

The (proposed) solution from Jeff: "all sites will stick with their topic.stackexchange.com names until such time as they reach traffic levels roughly around what Server Fault currently has".

I'd object to that, saying that probably Server Fault wouldn't have gotten to the current traffic level without their own domain, which made it easier for people to recommend it, promote it and generally talk about it in various places.

What I propose is a middle-ground solution: new sites shouldn't get distinct domains until their community finds a good enough name - one that is highly voted and not contested, with the SO team (as payers of hosting bills & development costs) having a veto vote.

So, for the cooking site - everyone loves Seasoned Advice - they should get it at the end of their beta (in 4 days).

For the web apps site - the team vetoed, they have to come up with something better.

For the math site - the top voted name has only 5 votes, way too little, maybe the math community would rather stay at math.stackexchange.com, or will come up with a good name in 2 months / 2 years.

  • 9
    I'd mention that I think the way the Nothing to Install name was reverted was plain wrong. It should have been discussed on their meta before, it wouldn't have been such a biggie if the name was left for a few more days. But the way it was done, it unnecessarily annoyed a lot of the site's users. – Dan Dumitru Oct 4 '10 at 18:51
  • 1
    I like the entirety of this opinion very much. +1 – Pekka Oct 4 '10 at 19:15
  • 6
    To be honest, I don't even like SeasonedAdvice.com all that much. But it's 10000 times better than *anything* dot stack exchange dot com. Heck, I'd vote for cookingoverflow.com if it would get us out of the dot-SE ghetto. And I'll wager that if those were the stakes (agree on a name or don't get a name), voting would go up... So I do like this idea. – Shog9 Oct 5 '10 at 0:17
  • it's impossible to define "good enough", though. There will always be some noisy member of the community who insists a certain name is "good enough" and "much better" than the simple topic.se.com alternative. And communities themselves may be biased as well; their desire to have a 'special' name may override practical concerns like, "gee, this name is special, but not very clear or useful to the broader internet." – Jeff Atwood Oct 5 '10 at 0:40
  • 3
    The SO team doesn't usually have a problem ignoring noisy members of the community - why would this be any different? Similarly, the ability to veto a name and state "suggestion X isn't clear enough, try again" seems enough control to come up with a practical name. – Peter Boughton Oct 5 '10 at 1:05
  • 1
    The more often I read this answer, the more I feel this is really the best and most constructive way to go. – Pekka Oct 5 '10 at 9:10
  • 2
    @Jeff those are weak counter-arguments. Letting people vote (with high thresholds, and the clear message "don't vote for the least worst name, vote for the one you WANT!, or vote for having none at all!") will easily overpower noisy members. And if the community bias is that much of a problem - I'm not sure it really is, because those communities are supposed to attract professionals who get inside jokes and all that, correct? - then let the whole of the SE community vote on domain names. That will equalize whatever community-related "bias" there is. – Pekka Oct 5 '10 at 9:56
  • @pekka only if you think it's "constructive" for me to executive veto every name so far (yes, even seasoned advice) on the grounds that none of them are "good enough", that is, better than the simple vanilla topic.stackexchange.com. Personally, I'd rather do it based on a sane, objective metric like traffic -- in combination with the "highly voted" factor. – Jeff Atwood Oct 5 '10 at 9:58
  • 1
    @Jeff I see what you mean. But I don't understand using traffic as a metric for this. It looks totally random: How will a certain traffic level provide for a good domain name to pop up? Quite the contrary, the more established your site becomes, the more difficult it is going to be to give the baby a new name. What I could live with is a threshold of active community members voting "yes" for a specific domain name. The idea behind the rule being that if the top x number of people who shape the community all want a certain domain name, they are entitled to get it. – Pekka Oct 5 '10 at 10:08
  • @pekka as I see it, domain naming is a privilege not a right. Once a city gets to a certain size, they have more privileges within the network. Same way rep works on Stack Overflow -- get to 2k and you can do things normal users cannot. Traffic is like rep in that sense. – Jeff Atwood Oct 5 '10 at 10:11
  • 6
    @Jeff fair enough. We'll have to agree to disagree here: I think things like server resources are privileges that can be extended as the city grows; but a domain name should be introduced as soon as possible if a good one is found regardless of traffic levels. I'll add one more suggestion in that vein as an answer here, even though the decision seems to already stand. – Pekka Oct 5 '10 at 10:31
  • 1
    @Jeff I would prefer you to veto names until the community comes up with one that will satisfy you than pretend you have an objective metric for determining when the community can override your veto. Let's be honest here, you'll never be prepared to let the community override you and recreate nothingtoinstall.com, so why bother pretending? Tell them to keep trying until the answer is good enough for you. You've decided you have a veto. No-one pretends that you don't, so take responsibility for your own actions and don't pretend there's an objective metric. – Richard Gadsden Oct 5 '10 at 19:35
  • @richard if a site has generated a lot of weight (traffic) on the strength of its Q&A, from my perspective it has earned the right to have the power to make decisions like this. Otherwise, it's like the junior programmer who's written 10 lines of code in his life giving you advice on your code... – Jeff Atwood Oct 5 '10 at 20:39

I think this is a weak and shortsighted policy shift, and that it will utterly fail to do the things being claimed for it. I cannot see any individual site of any merit gaining any significant benefit from being branded as an SE ghetto. It is possible that SE as a whole might gain something, although that's far from certain -- if the brand comes to be associated with a constellation of scrappy and inadequate subsites that wouldn't make it on their own, well you won't be able to say you didn't set yourself up for it.

What is certain is that this sudden lurch is going to piss off quite a few of the people you're asking to pioneer these sites, and it's hard to see how that can be anything other than a Bad Idea.

Yes, sometimes it's necessary to do things that not everyone likes. You can't please all the people all the time. But trust and goodwill are the essence of online capital. They are things that SO has plenty of right now, but it doesn't take much to dilute them. A few arbitrary and wilful acts, a bit of perceived unfairness, and a bunch of petulant "it's my party, I know best, and if you don't like it you can sod off" comments from on high can turn things sour pretty quick.

Suddenly, what seemed a trustworthy and reliable set up well worth investing time and effort into, instead looks like unsafe territory where the rules can change at the drop of a hat. Worse, it looks like the earlier trustworthiness was always a lie. That's really a terrible message to send out.

Obviously, this particular case may come to nothing, just a trivial storm in a teacup, soon forgotten. A few of us whiny geeks with too much sense of entitlement will sulk for a little while but will ultimately be unable to stay away from the One True Q&A Paradigm.

But in the long run, doing this sort of thing is not good for your reputation. And rep counts for quite a lot around here, I think.

  • one thing is certain: when you are asking the wrong questions, you'll always get the wrong answers. It's clear now we've been asking a terribly wrong question w/r/t naming. – Jeff Atwood Oct 5 '10 at 2:18
  • 3
    @Jeff The question is only wrong because it turns out you're not willing to entertain the answers. In the SO scheme of things, you're suddenly that guy with "0% accept rate" by his name. You designed that scheme, FFS -- you know what it means. – walkytalky Oct 5 '10 at 8:05
  • no, the question is actually wrong. See blog.stackoverflow.com/2010/10/domain-names-the-wrong-question – Jeff Atwood Oct 5 '10 at 8:12
  • 3
    @Jeff The distinction between domain name and community identity is a good one, though nowhere near as clear cut as "we've been asking a terribly wrong question". But you'll have a lot of trust to earn back on the latter front after this display of ill-tempered autocracy on the former. I hope you manage it. – walkytalky Oct 5 '10 at 9:39
  • 3
    (As an aside, you could start by getting over your personal strop on Aarobot's answer. "Call my wife"? Really, Jeff? Is that honestly the image you want to project for your benevolent dictatorship?) – walkytalky Oct 5 '10 at 9:40
  • 1
    @walkytalky I would cite "garbage in, garbage out" on that particular post. Also, there's little you could do to make me angrier than huffily presume that "I'm about the money." And as for ill tempered, well, there's a reason my power animal is the Wumpus. codinghorror.com/blog/2006/04/your-personal-brand.html You've got three arrows... use them wisely. – Jeff Atwood Oct 5 '10 at 9:51
  • 4
    @Jeff A fair enough thing to be annoyed by, but rather peripheral to the text -- as we can see from LeguRi's remix. All I'm saying is that, to my eyes, you've turned the post from being all about Aarobot's intemperance to being about yours, which seems like questionable PR. But I'm sure you're incredibly tired of this whole subject, so I'll stop banging on. – walkytalky Oct 5 '10 at 11:52
  • @walky it wasn't peripheral to the original version at all; the whole point of that was to be an angry rant that makes other people angry. You can see for yourself what that accomplishes. (I agree the revised version is dramatically better.) – Jeff Atwood Oct 5 '10 at 20:41

I thought about reposting my WA answer here, but it's already getting enough attention, so I'm going to post something new instead that's been brewing in my addled brain since last night.

All of us here - all of us on the SE betas and Area 51 and even Stack Overflow and the trilogy sites to an extent - we're volunteers.

Oh sure, the team paints pretty pictures about us being autonomous communities and I myself have written about some of the underlying social phenomena. But underlying all of those things, our actual relationship with the team is that of volunteers; unpaid workers who do our thing because (a) we get a hearty pat on the back, (b) we get the occasional perks, and (c) we sometimes, sorta, enjoy the work we're doing.

A few weeks into the Cooking beta, I mentioned the concept to my mom and what we were trying to do. Her response surprised me. It wasn't "Tell me more", but it wasn't "That sounds too technical" either, it was something totally out of left field. She asked me, Why are you doing all of this?

And it was a bit of a struggle to come up with an answer. What it all came down to was that I wanted to be part of a knowledgeable community but that the only other venues for that are clunky discussion forums, and that if you want something done right in this world then you've got to step up and do it yourself.

The Community aspect was, by and large, the crux of my argument back then, and still is today. We actually have a few chefs there, people with real kitchen experience, people who know a lot more than I do and are willing to share on occasion - unlike all of the forums where I'd mostly be jabbering with ordinary Joes in the same boat that I'm in. I'm there answering questions not for my personal reputation but for the reputation of the community - that we get your questions answered fast. And now I'm there moderating not because I like to tell other people what to do, but because I desperately want the quality of questions to stay high so that we can establish ourselves as a reputable community.

A community's identity is central to that community and the name is, fundamentally, the identity. We learned this crap on the street in high school; you're either a member of the jocks, nerds, goths, emos, ravers, punks, hipsters, metalheads, trendies or preppies. If you're not part of something then you are just not cool. Worse than that, you're essentially nobody.

Originally it was stated that we would get to choose our identity, which is what almost every coherent community does. They might name themselves or they might adopt a popular name chosen by outsiders but either way, that name defines who they are.

What is our identity going to be now? Nothing. To anyone outside programming circles, "Stack Exchange" might as well be 102-47th Street. Individual communities are apartments in a nameless high-rise building. That's not really a community at all, it's a weekly tupperware party.

And this brings me back to the point about being volunteers. Many of us out there are doing our damnedest to help build real communities, but it seems that the team keeps kicking up clouds of sand in our faces.

First, the Area 51 rules were changed and all of the votes were scrapped.

Then the beta period was extended from 60 days to 90 days, so we've all been forced to watch activity dwindle while we prepare to promote ourselves (which is nigh on impossible while the site is still in beta).

Then proposals like Software Engineering started getting questioned because the system failed to produce quality definitions, which was blamed as a failure of the community itself.

Then legitimate proposals like Compiler Design started getting canned on the grounds that they would be a "drain" on existing sites - even though the evidence clearly shows that no such activity is currently taking place on any existing sites.

Now, communities are being denied the right to privilege of a name, and thus an identity, guaranteeing that we'll look like losers and nobodies if we ever try to get ourselves noticed by bloggers or even friends.

Well, team, here's a newflash for you:

We don't give a f*ck about Stack Exchange.

We care about our individual communities.

I don't care if the Webmasters, Photography, Stats, Mathematics, Home Improvement, GIS, and Ubuntu proposals all fail. I don't care if every single proposal other than the ones I'm actively involved in fails. None of us care about somebody else's community.

We don't care about the Stack Exchange brand. We don't care how many or how few sites there are. We don't care how much money you make from clicks and ads, unless you plan to give us a piece of the pie. We don't care what your SEOs and VCs say. We don't care that you want Stack Exchange to be the next Wikipedia.

We. Just. Don't. Care.

We are donating ample amounts of our personal time to these proposals/betas for one reason and one reason only: So that we can be part of a community that we're proud of.

What we don't want is to be a nameless, faceless portal page that's part of some conglomerate that's supposedly very big and important but most people outside a specific industry have never heard of. We'd rather at least attempt to make the sites stand on their own rather than have to suckle at the teat of a franchiser. And it seems to be that at least on WA, the community has spoken quite loudly in agreement with this principle.

What it all comes down to is this:

If you want these communities to continue volunteering their time for your business venture, then you are going to have to start treating those communities with a little respect.

Yes, I get it, it's your company and you don't have to do everything we say. But you throw up obstacles and insult us at every turn. The bizarre voting rules on Area 51 are fine, it's just that most people don't "get" them. Beta periods need to be extra-long because you don't trust us to keep the axles greased. Fledgling subcommunities must abandon hope and join or die. Communities will only get their own names when they're making you enough money, and even then they might not because most of their ideas for names suck (oh, I forgot, Joel Spolsky came up with NothingToInstall - so it must be the community's fault for voting it up so high). Oh, and anyone who disagrees is clearly not worth listening to, so we're just going to revert the one existing domain anyway regardless.

If you want to crowdsource, you need to start showing a little less contempt and elitism toward the crowds.

If you genuinely want our support, then convince us - with evidence, not a bunch of flippant remarks and bad analogies and vague references to The Google, whose proprietary algorithms Google outsiders know precious little about and could turn their metaphorical back on you at any time.

If, by contrast, you wanted complete dictatorial control over the entire process then you should have just said so. I don't see why you even bother with the whole Area 51 ritual at this point. Just create whatever sites you want, with whatever names and branding you want, and see how many people are willing to join and contribute. That's what you did with Stack Overflow and it basically worked.

But don't dupe a bunch of innocent people into thinking that they have even a modicum of say in their community's destiny when you're so eager to pull the rug out from under them within a matter of hours.

It's insulting. It's hostile. Frankly, it's bad business.

Don't take your community for granted. It takes months, even years to build trust, but only minutes to destroy it.

  • 3
    "I don't care if every single proposal other than the ones I'm actively involved in fails. None of us care about somebody else's community." That's a bit nearsighted. If we didn't care about the network, why would we have bothered to create anything after Stack Overflow? Why would we be having this conversation? Quite simply, it is my job to care about the whole network, and ensure that everyone gets good answers to their questions. – Jeff Atwood Oct 3 '10 at 18:02
  • 6
    also, the idea that "this is about money" is ridiculous and frankly offensive. At no point have I ever done anything for money. Read more here: codinghorror.com/blog/2010/06/the-vast-and-endless-sea.html – Jeff Atwood Oct 3 '10 at 18:06
  • 6
    @Jeff: What's nearsighted is starving individual communities of essential tools in order to promote a branded network. What's offensive is declaring that domain names are "vanity" items for sites that "reach a certain traffic level" (i.e. are profitable or at least monetizable enough). What's especially offensive is telling us that as the owner of a company, you don't care about the money. Of course you care. The issue is balancing that with customer and "employee" satisfaction, which is a requirement for long-term success, and these days I haven't been seeing a whole lot of that. – Aarobot Oct 3 '10 at 20:22
  • 2
    I don't even look at my "pay stubs". I forgot to pay the house mortgage last month. I have literally no idea how much money is in my bank account right now (other than 'enough'). When I say I don't care about the money, I mean it. If you still don't believe me, email me (team@stackoverflow.com) and I'll let you call my wife direct on her cell phone to back this up -- since, you don't believe it coming from me, obviously. This is a real offer. I'm locking this until you (or anyone else) takes me up on that offer; I'm deeply offended by this claim that "it's about the money". – Jeff Atwood Oct 4 '10 at 4:28

I'm not gonna reply on NTI, because I don't give a rat's ass about NTI. If The Powers That Be actually care about our feedback, they can discuss it on Meta - otherwise, I'm gonna assume that this brain-damage is limited to NTI.

...As for the whole "SE as For Dummies" idea, I've only two words to say to that right now: developer.com

Ok, so comments appear to be disappearing here again... Just in case someone gets trigger-happy, the "developer.com" comment references a company with a long, long history of devouring developer sites, destroying their individual identities, and then using their content to bring in ad revenue while their former communities scatter to the four winds. Spend a bit of time poking around there, if you're bored - if you listen carefully, you might hear the cocooned torso of CodeGuru softly crying, "kill me... kill me..."

  • 2
    The cooks appear to be the exception to this rule since their name isn't a train wreck – random Oct 3 '10 at 1:37
  • 2
    @Pops: since when do I get to use punctuation to increase my word-count? – Shog9 Oct 3 '10 at 1:43
  • 4
    I don't understand your statement about developer.com at all. In all seriousness, could you rephrase it to be clearer, while avoiding unnecessary debate over how many words are in domain names? What does developer.com have to do with the branding of non-programming Stack Exchange sites? – Richard JP Le Guen Oct 3 '10 at 14:24
  • 1
    +1 @LeguRi. I'd never visited developer.com until a minute ago, but I'm not sure how it's relevant. Admittedly, I couldn't tell what the site's purpose/goal was, and my eye was totally lost as to what it was supposed to be looking at. – Pops Oct 3 '10 at 15:38
  • 1
    @LeguRi: You kinda have to have experienced the history to get the full effect, but here's a rough timeline: back in the late '90s, there were a bunch of little developer communities out on the 'Net. Then developer.com bought them all out, and rolled them into a big "network" with similar styles and branding, effectively making each one a subsection of the larger site, increasing over time, culminating in the reduction of each site's identity to a little heading under the internet.com banner. – Shog9 Oct 3 '10 at 15:57
  • 2
    @Pop: Your experience testifies to my point: developer.com is utterly irrelevant, the communities that were absorbed into it scattered. I'm sure they get plenty of Google traffic, since the entire purpose of the amalgam was to generate ad revenue for a series of parent companies... But any identity or respect they might once have held is long gone. They ARE the "For Dummies" of developer websites. – Shog9 Oct 3 '10 at 16:00
  • 1
    @shog9 we're shooting for more of an O'Reilly type model at this point. Animals on the cover is your branding. – Jeff Atwood Oct 3 '10 at 17:02
  • 1
    @Jeff: I can honestly say that I have no idea what that would mean applied to SE. – Shog9 Oct 3 '10 at 19:58
  • 7
    @Jeff: Why do you keep hammering the O'Reilly metaphor? O'Reilly programming books are like Stack Overflow tags - treatises on highly-specialized subjects within the specific umbrella of programming, marketed toward programmers, who all know the O'Reilly name, and even the O'Reilly books have fairly distinctive branding, they just have a few common elements. SE is aiming at a worldwide audience of everyone on every topic, the majority of whom have never heard of Stack Exchange. The model you're actually following now is a weird hybrid of Yahoo Answers and about.com. – Aarobot Oct 3 '10 at 20:30
  • 7
    @Jeff: That's not the O'Reilly model anymore. The animals on the cover are for practical tech books. As they've expanded, they've added designs. They did that because, for instance, Nutshell designs wouldn't have worked in the cooking section of a store. When they published a book that would fit okay in the culinary arts section, they rebranded. They even rebranded for less practical tech books. Stackexchange sticks out in the wrong ways outside the tech space. In other words: naming is hard. – freiheit Oct 3 '10 at 21:37
  • 2
    @Jeff But the domain name is the first part of the site the users see, just like the cover (or spine) of a book is the first part of a book the customer sees. The custom designs you mention are more like the design of the interior of the book: something you don't see until you're already somewhat interested in that particular site or book. – freiheit Oct 3 '10 at 23:15
  • 1
    @frei I would say it may be the first part they see, but it isn't the first part the vast majority of people process. The first thing they process is logos and style. People simply remember a url (when they remember it and not just google for it) like they would a street address - a necessary piece of information to get them where they want to go. Think of some major brands you know. In your head, do you see the url to their website or the branding on that website? – Zypher Oct 4 '10 at 0:01
  • 1
    @Zypher: If we want to gain publicity by convincing bloggers to mention our sites (which many of us believe [and the team originally seemed to believe] the majority of initial support will come from), then the title/domain name is indeed the very first thing that potential users are going to see. Your assertion seems to assume the same thing that the team is assuming which is that 90% of the traffic will come from Google, just like a mature site - except that hardly any traffic is coming in to SE from Google right now because there just isn't that much to index. – Aarobot Oct 4 '10 at 1:57
  • 3
    [ Cooking for Geeks ](oreilly.com/catalog/9780596805883) is an O'Reilly book!? I've been wanting to read that for months, and I had no idea it was published by them. – Pops Oct 4 '10 at 2:57
  • 2
    @freiheit - I would like to agree with you, but find your arguments weak; in today's day of shortened URLs I feel like the domain name is less and less relevant to end users. Additionally, I think when comparing books and web sites, the domain name is less like the cover of a book and more like its library call number; it makes finding it easier and tells you a bit about the subject matter, but doesn't really affect people's interest in the book/site past that. – Richard JP Le Guen Oct 4 '10 at 4:20

While I can see both sides of the argument here - communication with the community was botched badly on this one, but it's not like the management doesn't have good points - I think (and @rchern has already pointed out over at meta.webapps) there is one additional problem in setting "ServerFault-like traffic levels" as the precondition for a site to get its own domain name.

Renaming a site will become much, much harder once it is successful.

The bigger and better known a SE site grows, the more difficult will it become to re-brand it one day, once it is on the required traffic level.

I can't see a site that takes off, and becomes a success in its field, being able to leave the .stackexchange.com umbrella once it is big. The pains would be so huge: Losing Google ranking. Losing brand recognition. Losing audience that isn't happy with the renaming. And so on, and so on. If anything, the pains will be bigger compared to having a domain name now, not smaller. Nobody will be willing to take the risk.

I really think a site staying inside the SE umbrella is very likely to stay inside it for good. An individual domain name once the site has reached this and that level is a really dangerous promise to make.

Am I wrong? Is this too pessimistic? Any thoughts?

  • 2
    I've been saying this since this whole mess started! – Rebecca Chernoff Oct 4 '10 at 19:32
  • @rchern I'm sorry if I repeated something you have already said from the start. In my defense I have to say that the discussion has expanded so much (and across two sites) it has become really difficult to keep track of :) – Pekka Oct 4 '10 at 19:33
  • 1
    @Pekka, I didn't mean it like that. It is something I've said elsewhere, but I absolutely agree. I don't buy the team (or the community for that matter) being willing to switch the name of an established site like this. – Rebecca Chernoff Oct 4 '10 at 19:43
  • @rchern yup. This is bound to lead to future clashes, when a community has reached Server Fault levels, but won't get a domain name for these reasons. This is a dangerous promise to make – Pekka Oct 4 '10 at 19:45
  • Losing Google ranking. I think that Jeff mentioned somewhere that this was a non-issue since there will be a 301 redirect in place, but other than that I agree. – Tim Stone Oct 4 '10 at 19:47
  • 1
    @Tim, losing google ranking is what started the mess. It wasn't about naming. It was about google search results. There was a 301 in place for the 24 hours NTI was actually NTI. – Rebecca Chernoff Oct 4 '10 at 19:49
  • @rchern: Just so I'm clear, are you referring to the bit about ranking in the Meta NTI/WA question, or something else? – Tim Stone Oct 4 '10 at 19:59
  • 2
    @Tim, the former. When it was initially presented before being posted to MNTI, google rank was the main issue and naming was the secondary issue. – Rebecca Chernoff Oct 4 '10 at 20:32
  • @rchern: I had a response to that, but in light of the fact that I think I might have imagined reading comments/I'm not confident at all about my perception on how the ranking process works, I'm just going to strike my original comment. :P – Tim Stone Oct 4 '10 at 20:58

I have some concerns, and maybe it only affects our indecisive gaming clan, but I'm (at least hoping) that we're not the only ones who suffer from this problem. So I'm posting here as a response.

How does our logo fit into all of this? For us, the logo is rather entwined with a future domain name - we really can't make a generic logo that will fit whatever name we plan to be blessed with. Web Apps has their shiny "power in the clouds", Cooking can use a chef's hat, and in both scenarios it is emblematic of the realm without conflicting with the majority of good names.

But with the exception of one domain name proposal that really sounds better used for a billiards SE, none of our current logos match any of our domain names. It all looks haphazard when matched mindlessly - a moving pong paddle for Pause for Help? A frantic button mash for Stage Six? A generic controller for Gaming Owls?

In all cases, we need to solidify one in order to truly solidify the other. But now we can't really solidify the domain name, and I can live with a period of time as "Gaming Stack Exchange". But we're going to live with a logo during this period, and moreso than our generic name that unique logo is going to be an identity to us. And people are going to be attached to this logo, which down the line is going to shoehorn us in our plans for making a unique domain name.

We still need to come up with a core theme that we can use to develop our logo and domain name together, so that they do not conflict with neither each other nor what at least enough of us want to see when we visit our home. So for us, this change doesn't truly affect our need to continue hunting for a name. How do we come up with a unifying logo if we don't get a nice, matching name? What if we come up with a good logo, but there really isn't a good matching name that can go along with it, especially considering the problems with purchasing domain names in the first place?


If so, why not taking Stack Overflow, Super User and Server Fault to programming.se, helpdesk.se and admins.se, respectively?

Seriously, I think the private domain will be a great additional benchmark for the popularity of the site. If it won't survive with offering Google good content, it should die instead of being artificially kept alive by the power of Stack Exchange. You have made a great job with cleaning Area 51, go on!


Birthing out a community leaves stretch marks the likes of which all the cocoa butter in the world isn't going to smooth over nicely. Easier instead to build sites of what will become an aggressively generic network.

Despite sharing the same DNA as its parents, it's far less hassle naming a newborn with a generational suffix than giving it a distinct name and permission for personality.

Stack Exchange isn't about building communities, it's about building search engine traffic to topics that ads can sell well on. Nothing to do with philanthropy and all to do with making sure search results are spammed out with keyword.stackexchange.com

Don't worry though. This is only limited to Nothing To Install. Until the next topic graduates to ditching the nappies.

But by then, dismounting from the tramampoline will appear smoother.

  • 5
    it is about building communities; when 90% of our traffic comes from Google, how do you think new community members are finding us? All those hand-printed flyers I put on campus? – Jeff Atwood Oct 3 '10 at 3:49
  • 2
    Well... That's how I got here. The flyer said there would be pizza! – Shog9 Oct 3 '10 at 3:59
  • 1
    @Shog9 Is there not any pizza? Because I was told there would be pizza. I'm a classic guy, pepperoni and cheese for me! – Jared Harley Oct 3 '10 at 4:25
  • @Jared: The pizza is a lie! – Shog9 Oct 3 '10 at 4:27
  • @Shog9, since when is cake the same as pie? – Pops Oct 3 '10 at 4:56
  • 1
    @Pops: since cheesecake! – Shog9 Oct 3 '10 at 5:49
  • @Jared, +1 for pepperoni and cheese pizza, by the way. Someday I will have to look into the pepperoni lobby's influence on the pizza industry. I mean, how often do you hear about pepperoni being used outside the context of pizza? – Pops Oct 3 '10 at 15:35
  • 1
    @Jeff Okay, 90% of your traffic comes from Google. But how much of that traffic is coming to stay and create quality content? – Pekka Oct 5 '10 at 9:14
  • @Pekka: Actually, Jeff said to me that that's not important. I can see his angle, of course; I just think it applies best to communities much larger than any 3-month-old SE would ever be. – Aarobot Oct 5 '10 at 14:45

The team's current position on the issue:

All sites will stick with their topic.stackexchange.com names until such time as they reach traffic levels roughly around what Server Fault currently has (~38k visitors/day).


So, feel free to pick a name, and if it's good we'll try to use it in..

  • site redirects (so seasonedadvice.com will redirect to cooking.stackexchange.com)
  • site visuals (graphic logo tagline)
  • site tagline text

Jeff Atwood

  • 1
    The fact that this policy, which seems good, is generating so much emotion is very surprising to me. – jjnguy Oct 4 '10 at 15:20
  • 5
    @jjnguy: i don't think it's a policy per se, but rather manner in which it was done. The policy that arrives on the hills of decision is not a policy at all, but rather a fixing-a-mess measure. – SilentGhost Oct 4 '10 at 15:22
  • 4
    And to top it off it's posted on meta.cooking ffs. – SilentGhost Oct 4 '10 at 15:29
  • 6
    @jjnguy There would have been zero problems if this had been done this way from the start, or the change would have been introduced in an actual discussion along the lines of "the community naming process isn't working out. What do we do? We are thinking about sticking with stackexchange.com" – Pekka Oct 4 '10 at 16:45
  • 1
    @Pekka, While I understand that it was not handled as gracefully as possible, Jeff is responsible for the success of his company. If he (and Joel) feel that they need to make an executive decision, then they must do that. The success or failure of this endeavor is their responsibility. They need to do what they think is right. And, since we value the success of our communities just as much as Jeff does, we need to trust their judgement. – jjnguy Oct 4 '10 at 17:01
  • 6
    @jjnguy I totally agree that the executive decision is always theirs to make. No question. But this is not a business where a product is sold, and customers can either take it or leave it. Here, a core of relatively few people spends a lot of time - voluntarily - building the basis for the business's success. You need to give these people at least the courtesy of informing them ahead, and offering a coherent explanation on how the decision was reached. This failed magnificently - not for the first time. Which is why I can't help but feel that much of the anger is somewhat justified. – Pekka Oct 4 '10 at 17:08
  • I do hope this policy will be blogged because having policy info spread about the network is a bit confusing. As with the Area 51 duplicate policy, the Stack Overflow Internet Services team appear to be victims of their own success. Some parts of the process haven't worked out so well, and it's understandably difficult to change course fast enough to avoid upsetting people. Like naming things, managing a community of human beings also appears to be very difficult. – moberley Oct 4 '10 at 17:15
  • 5
    @jjnguy: Communities are not Agile. You can't "refactor" the entire system again and again on a whim. As Pekka says, if this policy had been enshrined from the start then it would have been different. For one thing, I would not have put in as much effort as I did - most sites would probably get treated like perpetual betas instead of trying to iron out all the major wrinkles within 90 days. The point is, we were promised that SE2 would be like SE1 but better - free, managed, open to everyone. This has turned out to be false. I wonder how the Math Overflow community feels about this. – Aarobot Oct 4 '10 at 18:30
  • 38k? The SE 2.0 site with highest traffic (Gaming) currently just has 7.6k, and WA now just has 1.3k, I wonder if it is possible to make this far in a year. – kennytm Oct 4 '10 at 20:36
  • @Kenny It sure isn't going to happen overnight (yes, I have a few fixes and extra stuff for this waiting to be deployed) – badp Oct 4 '10 at 22:23

As I think having a domain name is important for the development of a site and more than just a vanity privilege, here is a full suggestion for a voting process that will ensure quality.

  • Let each community decide on the domain name in a one-week-long election process.

  • A site must reach a certain (but relatively low) activity threshold to get an election. I personally would prefer activity (questions, answers, or votes) over traffic to measure activity, but be that as it may.

  • Establish a separate format for this more in the style of how the moderator elections used to work; Don't use the Q&A format

  • Instead of voting suggestions up and down, have users commit just like the process of committing to a proposal works right now. Users can commit to a maximum of 1, 2 or 3 domain names, and they'll be publicly visible as committed to that name. They can even leave a short message as to why they support the domain name.

  • a fixed number of active users (50, 100, 500, whatever) in that community need to commit to a domain name for it to get chosen.

  • Add

    no domain name is good enough right now, vote again in 1 year

    as a voting/commitment option.

  • If no domain name is decided upon within the one week period, the site will not get a domain name at that time. The next election will be held in one year's time, provided the community still fulfills the necessary metrics by then to qualify for an election.

  • No executive Veto: If the people want a domain name, let them have it.

  • Optionally alert the whole of the SOFU community of elections on a specific site, and allow them to comment there (but not vote). This provides thousands of laypeople to speak out against too insider-y, freaky, area-specific, obscure proposals. The last word should be with the community the domain name will apply to, though.

It would have to be determined how to deal with already registered domain names whose buyability hasn't been determined yet. There would have to be a pre-vote in the Q&A format like it is right now, and any domains with a decent chance pre-registered or excluded if they turn out impossible to buy. These things would have to be worked out and wouldn't all be easy.

But this would be a fair shot at true democracy: It would give every site the opportunity of getting a domain name, but under very strict rules. And if a community, as a whole, so badly wants a domain name that it's even ready to settle for a crappy one, I think it needs to have it.


I think I am starting to come around to the idea of a "series" for StackExchange sites. Obviously, this needed to be discussed sooner, but my two cents anyway:

The sites share a common header across the top. (The little thing that says "StackExchange".) Why not have the rest of the sites as subdomains of stackexchange.com? I know this causes the sites to be more "branded" but isn't the style/theme enough to make a site unique? (Never mind the topic itself.)

  • 4
    At least stackexchange.com is becoming the new .info – random Oct 3 '10 at 1:26
  • @random: Yup. I like it. Wait until ".se" becomes a new TLD - no wait, it already is! – uɐɯsO uɐɥʇɐN Oct 3 '10 at 1:39
  • 1
    @George: (Rosinante is) way ahead of you – Pops Oct 3 '10 at 1:49
  • .info is considered a spammer's domain. – random Oct 3 '10 at 1:52
  • 5
    You're really talking about two different things. Domain names are essentially product names. It sort of matters who makes a product but it's not part of the name, just part of the packaging. Half of the people I speak to don't even know which company owns the Corvette, but of course it is branded with the Chevy name/logo. Similarly, you don't go out to buy the "RIM Blackberry", it's just a Blackberry. What the generic SE-branded sites look like are the generic apps included in Windows now; Mail, Messenger, Paint, Calculator - for those who don't really give a damn as long as it's free. – Aarobot Oct 3 '10 at 14:03
  • 2
    @aaro no, we're more like the good default apps shipped with OSX. – Jeff Atwood Oct 4 '10 at 4:52
  • 6
    @Jeff: OS X apps are a great analogy. Approximately 5% of the market will use them, and that 5% will be made up almost exclusively of people who already trust that brand. You can keep rolling out more and more default apps, but as long as you keep the OS X branding, your user base will not grow much beyond your existing market. On the other hand, if you want to do something really crazy and different and rope in a totally new market, you should consider ditching the "Mac" branding and going with a radical new name, like "iTunes" or "Final Cut." – Aarobot Oct 4 '10 at 19:10

I just posted a response to the article and wanted to share it here as well.

I think it's important for the success of the site to have as many users as possible who are interested in web applications (or whatever the subject is), and that's why Search Engine Optimization is important. Not for clicks and $, but for sharing knowledge: you want to have new users find the site.

The question of the domain name is very important for SEO (Search Engine Optimization) and therefore traffic.

I came to this conclusion after launching AskAboutProjects.com (SE on project management)… 12 months later, I think we have a great StackExchange site, but very little traffic from Google searches.

Sure it would be ideal to get a viral type of expansion, but realistically without thousands of readers/followers, there is little chance to get much traction.

On the other hand, choosing the right domain name does make a huge difference in the long run (by which I mean after few months already).

There is quite a lot of evidence that using an exact match of searched keyword in the domain name gives a huge SEO boost.

Evidence #1: ebook.com See this article: http://www.thedomains.com/2010/09/23/ebook-com-hits-the-auction-block-with-4-5-million-unique-visitors-a-year/

Evidence #2: a google search on “web applications” gives as SERP 1 wikipedia, and SERP 2 a site using ‘webapplications’ inside its domain name.

Update -> Evidence #3: Perfect example with EntrywayOrganizer.com with the story told in this pdf by Estibot This sites rank SERP 1 for the keyword search... "entryway organizer".

In conclusion my recommendation would be: if you want to make your life easier for traffic, grab the domain webapplications.net which is selling on Sedo.com for 1,088 USD (or the .info which is selling for half of that). It is a small investment which can greatly change the outcome down the line.

You might think this is expensive. It’s not… if you plan to use it for something good like a StackExchange site.

note that I do not own these domain names, and have no interest other than to share my little SEO experience.

Best of luck.

  • 1
    Sounds like you have a lot to contribute to Pro Webmasters... perhaps starting here? – Pops Oct 5 '10 at 13:10

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .