Andrew: Coming up! Well, actually, here’s what’s not coming up. There’s no video in this interview but, frankly, between you and me, Mixergy was never supposed to be about video. It was supposed to be about ideas, about words. Audio only was the format I imagined. I only added video because it drew in more users to those ideas in the interviews. Well, in this interview, we don’t have any video. What we do have is the story of how two founders discovered an idea that led to a business that helps over 30 million people every month, including me. And how can you find a similar idea? I wanted to know in this interview. The other thing I wanted to know was what about getting users? How do they get the unfair advantage, as some people consider it, that helped them grow their audience in the beginning? And finally, where do you get your ideas for improving a business? All that and so much more, coming up.
Before we get started, I wanted to congratulate Mixergy fan, Carlos Ellorio [SP], who co-founded Art and Reed while listening to my interviews, the way you’re about to listen to this program. Can you imagine the pride he feels every time someone goes to Art and Reed.com and buys a custom-made suit or shirt? I just, five minutes ago, bought a suit and shirt there. Mixergy is made for doers of the world like Carlos and you.
And this whole she-bang here is sponsored by Grasshopper. Do you need a single phone number that comes with multiple extensions, so anyone in your company can be reached, no matter where they are? Well, go to Grasshopper.com already! It’s the complete virtual phone system that entrepreneurs love, and it’s sponsored by Walker Corporate Law. Do you need a lawyer who actually understands the startup world? How about that, instead of one that just figures that anything online is beyond him? Scott Edward Walker is a part of the startup community, which is why I recommend, when you need a lawyer, you go to WalkerCorporateLaw.com. Now let’s get started with the program.
Hey, everyone! My name is Andrew Warner. I’m the founder of Mixergy.com, home of the ambitious upstart. How does a blogger launch a collection of question and answer sites with over 30 million monthly visitors? Jeff Atwood is a developer who’s been blogging consistently about his word and his ideas on his site, Coding Horror, since 2004. Four years later, he announced StackOverflow.com, a site where programmers can help each other out. That grew into StackExchange, a collection of what is now 87 similar question and answer sites. And by the way, on a personal note, when I had a question about how to change a template on my website, I asked it on WordPress.StackExchange.com, and I got the perfect answer in under two hours. I invited him here to talk about how he built up such a great collection of sites and such a good business.
Welcome, Jeff.
Jeff: Yeah, thanks for having me.
Andrew: I shared my experience with it. Do you ever have a similar experience, where you went to your own website and got an answer to a question that you might not have been able to solve otherwise?
Jeff: Well, yes. I mean, I think most people do startups, I think, are really scratching their own itch, and I think when I created StackOverflow, it was certainly about, I wanted a site I could go to that behaved a certain way. I mean, you make a laundry list of what’s wrong with a lot of sites on the internet, starting with really basic things, like they’re too slow. Like you, (?) only takes three to five seconds to even figure out what’s on the page, like for you to even get content on the page to look at, and other basic stuff. Is it encrusted with ads? Is it encrusted with just noise? Not even ads, but just visual noise. I can’t even tell what I’m looking at. I have to spend a lot of brain power parsing, like what’s this page? What’s this? What am I even looking at? Dumb stuff you shouldn’t have to think about, and this happens all the time.
When you click a link, it’s like you frequently get smacked in the face with this mackerel of a page that you have to spend all that time making sense of. Beyond that, there was also a sort of evil, when the original idea, when I talked to Charles Bolski. He had the original germ of the idea, which was, “Hey, there’s this site, Experts Exchange, that does Q and A, and it comes up a lot when you search for stuff as a programmer.” And that’s good, except when you went there, it was like arriving at a used car sales lot. You have the salesman in the loud jacket with the smelly cigar comes over and immediately starts trying to wheedle a deal out of you, right? You feel very slimy just for having gone there. It was a very (?) experience.
Andrew: Specifically, what they did was… Actually, you tell me. What did you see when you were looking at it back then?
Jeff: Well they just had to do two things. One, Google won’t allow you to not put the results on the page that you show the user because of you’re indexing one page but showing a different page that’s bate-and-switch, right? It’s you’re showing Google one thing but you’re showing the user another. What they did was they would put the answer way down at the bottom of the page but at the top of the page they would show question and then the answer would look sort of hidden. There was a curtain there. They said, “Oh we’ll pull away this curtain if you pay us $9.95 a month.” But if you knew the trick which was if you arrive through Google you could scroll all the way down at the bottom and get the answer. This site was universally hated. But the thing that was frustrating about it is they had really good currency in Google. They were getting a lot of really good high ranking results. They had good information. They were just presenting it in a way that was very slimy.
Andrew: And there answers were strong. I mean they were user driven. Were you a member at the time?
Jeff: No. I hadn’t known about that. It’s funny you mention that because nobody ahs ever asked me that and it is true actually I did a long time ago. I’m talking 2001. I’d gone on the Experts Exchange. I think this is before they sort of turned the corner. The other thing that Experts Exchange did, it was kind of a violation of the community’s trust, was they took a model of people contributing content and sort of stuck it behind a pay wall that they benefit from. It’s they’re making money off the backs of their users in a way that’s really not super fair. But I did actually go on in 2001 and answer some questions. I think it was about a hidden share in Windows NT which you can do with dollar sign.
Andrew: (?).
Jeff: (?) keep it. D dollar sign. And I was like, oh this is how you do it. And I did answer on Experts Exchange. I actually have a membership. I don’t know what the credentials are. I’ve actually posted and answered there, but the site of all it survived the first internet crash and went through a bunch of sort of reinventions of itself before it arrived at it’s current model.
Andrew: Jeff, one of the things that I’m so fascinated by and one of the reasons I even created this site is to figure out where great ideas come from. Was it just a matter of saying this site exists out there, Expert Exchange. If we stripped away the pay model and added advertising instead, soft ads instead, we’d have a winning business. Or was there more to the process of figuring out where the right business was?
Jeff: That was pretty much it. I think the way I summarize it is, I would summarize it slightly differently. We basically stripped away the evil. That’s really how I view it and this site was loathed. one of the great things about having if you’re going to start a business, in my onion based on my experience, I guess all statements kind of implicitly predicated by that, but if you can come up with an enemy, a thing that you’re fighting that people generally don’t like, A it helps you explain what your mission is. You explain, look we’re like Experts Exchange without all the evil and they immediately get it. A light bulb goes off over their head. Like, oh yes definitely. You don’t have to spend all this time explaining what do you mean by this, what do you mean by this. It’s this bad thing but not bad. So definitely that was indeed the model. But still execution, that was just the idea right? The idea is to do this thing without the evil but execution is everything. that’s true I think of every business. We’re really serious about executions well, right?
Andrew: I’m sorry. Let’s stay with just the idea for a little bit longer so that I can understand it. Joel came to you with the idea. He suggested it. Did you instantly say I get it and I want to be a part of this?
Jeff: Pretty much because I’m more of an execution kind of guy. I’m not a great idea person. I don’t sit around and dream up crazy ideas all day. I’m Mr. Execution basically. It was great for someone to hand me an idea I love on a silver platter because I knew I wanted to do something with community, because going back to my background I was coming from a blog that was insanely popular. A blog that was so popular it was starting to overshadow my daily work at the job I had. Where going to my job started to seem very quaint. I could go to my job, I could influence 15 to 20 people a day, or I could write a blog entry that 40,000 or 50,000 people read, respond to, and react to. It’s starts to seem a little crazy right? The difference of effort versus reward. I had this ball of energy. I was like, I have this ball of energy. This is a great responsibility. I want to take this ball of energy and I want to do something with it, something awesome. That benefits the community, that benefits the world, that I think is cool, that I think the world needs. I was just looking for a place to point that ball of energy and I called Joel. I was calling a bunch of people. Well, I was emailing a bunch of people on the internet asking them, I don’t know what to do. I have this blog. It’s gotten really popular. I want to do something. I don’t know what it is. Joel had a great idea to his credit. I mean I think Joel is much more of an idea person than I am for sure. But it was great to have that idea handed to me on a silver platter really.
Andrew: I see. I remember listening, years ago, to an interview with him. He did it with Gregory Galant. Where he basically said . . . I think you called him up, saying that you can make a living off your blog and you were looking for ideas for what to do next. Is that right?
Jeff: Yeah, that’s definitely it.
Andrew: So, you just called up people and you said, ‘What do I do next?” What kind of feedback did you get for that?
Jeff: Well, Joel and I kept the idea to ourselves, a little bit, until we figured out what we were doing. Like how we were going to do it and what was going to work. ‘Cuz, Joel is the type of guy that he really likes to work with people, ideally, in the same room. His big thing is, ‘let’s sit down to lunch together. So, for him to work with me, remotely, we had to wrap our heads around how that was going to work and all that stuff. And Joel viewed it as an experiment. And I think rightfully so. Because we didn’t know if it was going to work. Joel didn’t know if I was full of crap. Maybe I don’t know how to do anything [laughs] . . . [??] . . .
Andrew: How did he even know you were the execution guy? I’ve looked back now on so many of your old blogs, Blogga entries. And you seemed right, but, when you work on a project, how can you tell if someone is really that execution person, who made happen what you made happen? How did he know it?
Jeff: I think it’s a leap of faith. It’s like we knew each other. I had met Joel once in person. He had done a FogBugz world tour in the Emeryville [??], I’d met him. So, we had met once. And I think he was just trusting, essentially, previous reputation or my trail of artifacts that I’d left on the web led him to believe that this is a reasonable risk to take to work with me.
And I felt the same way about Joel. Joel has written legendary blog entries and he started a great company. We viewed it as, we’re taking some risks, we’re working remotely, but we’ve grown to know these people online. And that to me, I love that. I love the idea that you can grow to know somebody by their work online. It’s fascinating. It’s never the total picture of a person, of course. I mean, that goes without saying, right. There’s a lot of nuance and people are complicated, as they say.
But it’s fantastic to follow someone’s work for a long time, then get the opportunity to meet them or work with them and find out that their work was representative of the type of person that they are. I think . . . [??] . . .
Andrew: One thing I remember Adam of Hipmunk telling me is that if you’re scratching your own itch, try to get the idea out as quickly as possible. So that you can get feedback from other people who’ll make your product less about you and more about the world. Did you guys do any of that?
Jeff: We did. My feelings on that are, you have to have . . . you’ll spend a lot of time explaining what it is you’re doing. Even though we had a great mission, expert exchange without the evil. It was relatively easy to explain. The devil is in the details. There’s all these details, but how’s it going to work? What’s going to happen? And until you have enough to show people, and it’s actually representative enough, it’s not a total shell with thousands of missing functions. You’re going to spend a lot of time explaining the way things are supposed to be. And it’s very, very tiring.
That’s what I noticed with Stack Overflow because we announced we’re doing a Q & A site for programmers. And we had the podcast, where Joel and I had these calls every week. Or we would just talk and Joel said, “Hey, let’s turn this into a podcast.”
And I thought, great. We’re taking work that we would do anyway. Joel and I had to talk every week and turning it into some artifact that the community can provide feedback on. We can maybe provide some level of insight to the community, stuff they didn’t know or stuff we figured out, or that we were struggling to figure out. And that, sort of, virtuous cycle, surged early, where we tried to be as public as we could about what we were doing.
But, not . . . We wanted to build enough that people could look at it and show don’t tell. Rather than me telling you what Stack Overflow was, which we just spent a lot of time doing. I said, “Hey, just go look at it.” Sign up for the Beta. Bam, you can see what it is and then you can provide deep levels of feedback. So, that’s my only caveat. You have to have enough, so that you’re not constantly explaining, “Oh, eventually it’s going to work like this.”
Andrew: I see. And how long did it take you from that initial conversation to the first version of that you can show people?
Jeff: Oh, gosh, I’d have to dig through . . . I think, I kind of started . . . It was four years, almost four years, ‘cuz I did end up leaving Stack Exchange. I still have a strong connection. I have stock and we’re all very close friends and all that good stuff. But, I don’t do anything day to day with Stack Exchange anymore. And that was almost exactly four years, because it was March 2008 to March 2012. Because the babies, the twin baby girls, were born in February. So, that was kind of an inflection point in my life. In which it made sense to go in a different direction.
So, we started in March, the planning, and then . . . One of the things I learned early on was this idea that I can do everything myself was kind of impossible. Not because I don’t get things done but because I had no buddy. You know, the buddy system, of just having someone to ping and go back and forth was so critical. I had no idea how foolish it was to think, I’ll just write a bunch of code myself. Because, man, that did not work for me. Not even a little.
Andrew: You didn’t have that at Stack Overflow?
Jeff: Well, almost immediately, I sort of recruited a friend of mine who happened to be available who is a brilliant programmer, Jared Dickson, and he turned out to be available and luckily it worked out. And also, I was able recruit another brilliant programmer after that, Jeff Delgas, who I had known from sort of my previous lives and jobs and stuff and these are just great programmers that I always loved working with. Super smart guys, friends of mine so I was just fortunate to be able to bring them on and then we had the sort of two way and three way interactions going Gosh, that sped up like dramatically because just having someone to talk to in Ping daily.
Because Joel was really an advisor. I would talk to him once a week and it was great to talk to Joel and Joel has great insights about the way things work, of communities and so forth but for day to day sort of and how do we get this implemented, how should we do this? What tools should we use, not having a coding buddy was really crippling to me.
Andrew: Do you have an example of something of an insight that you got because you were working with coding buddies that you didn’t have before?
Jeff: Oh, Gosh, all the time. Jared immediately went to j query which was of course, before j query was the standard on the web for interacting with [??]. And he’s like…
Andrew: Did I just lose you?
Jeff: …like this j query thing. It was like cool. I trust your judgment. That sounds really cool. Are you hearing me?
Andrew: Yeah, I hear you now. OK. Yes. I got that. I see. What about going back just a little bit further back where it was just you and Joel talking about Joel Spolsky, I should give his last name of course for the transcribers. It was just you and Joel talking and then you put it out into the world.
What’s one piece of feedback that you got or one direction that you had your eyes open to that that you wouldn’t have been aware of otherwise?
Jeff: Well, it is true that it’s good to get, once you have enough that you’re not constantly explaining what it is you get to show people, it’s definitely true that say you give three free months of time. You can spend 3 months quietly coding away in your room or your tower or whatever. Or you can spend three months sort of reading user feedback on actually trying to use whatever software it is that you are creating.
And I think in my mind there is really no question that that 3 months is much more effective if you’re actually reading what the people are saying based on trying to use your software. Because otherwise you’re just sort of imagining, you’re predicting what your users will do when they see your software.
And those predictions, even good predictions like the ones Shawn and I had don’t always turn out to be true. Like one thing we found immediately that was just a huge oversight was we had this idea of Q and A as a strict Q and A system. There’s a question and there are answers and that’s it. There’s no other discussion that should go on. But we realized there’s a lot of clarifications you need. Like “Oh, I don’t understand this piece of your question. Can you elaborate?” or something about your answer but what about this, “Can you provide a source?” “Where did you read this?”
So in the absence of that, people were constantly posting answers that were talking to other answers. And this made the system very, very noisy which was the opposite of the goal and we realized well, we screwed up. Like we didn’t provide one level of meta discussion, which is now what you know as comments on Stack Overflow. And that happened during the beta very, very rapidly because we saw people constantly talking to each other in answers. We realized they’re trying to tell us that the system is broken in a way that we didn’t understand until they started using it.
Andrew: I think I had that same experience with the example that I gave at the top of the interview, where I said, I think I said ‘I’m looking for a theme that would only apply to independent original blog posts and I think someone in the comments said ‘Hey, wait. I think what you mean is template that only works for blog posts. And then that kind of clarification helped both me understand what the problem is and also other people answer it. I can see the value of that.
Jeff: And if you think about the way Stack Exchange works, comments are very important but I think you brought up a great point which is comments ultimately most of the time should imply edits to the post. So comments aren’t really meant to live on forever nor are you meant to have these long back and forth conversations. It’s mostly about oh, let’s make the post better. It’s like a little post-it note you put on the post, we can make it better by doing x and then you edit the post and that comment doesn’t really need to live there anymore.
And it’s a really edit driven system versus the conversational system [??]. It’s like five pages of people figuring out what it is you’re talking about and then on page 4 you would post “Oh, here’s my rephrased version of the question that actually makes sense now”. But the problem is that’s now on page four, not at the top where it should be. So you see what we’re getting out of the system.
Andrew: I’m wondering what I was changing because you know what, Jeff? As an outsider, I would think, hey, you know what. It’s so easy to have a great idea that basically takes someone else’s idea and reduces, you call it the evil but maybe I would just say change the business model on it. And there’s nothing else you need to think through. Any questions you have you go back to this site that you didn’t like and you see how they solved that one thing which, and then we’ll bring it on to our site minus the revenue model. So what else wasn’t that simple?
Jeff: Well, I think it’s also…
Andrew: I know, of course, that it’s not like that.
Jeff: Well, let me also clarify that, we were also building a bit of Frankenstein monster, in that John and I were trying to take not just, you know copy [??] but subtract the evil but also say, hey, lets incorporate in our Frankenstein monster all the pieces that we like, all the body parts that we know are awesome, from other sites like from Digg and Redditt we loved voting. We knew voting really worked to motivate people, it really worked to escalate good content [??]. From blogs I knew that ownership was this big-big concept, when somebody comes and says, “You own the question, you’re the question owner, you have special privileges, you look at the answers and tell us which one you like the best” right, that’s another dimension beyond voting. [??], I like this answer the best as the question owner and as the owner, you know, you’re tending your guard, you’re like, “Oh this is my question, I want it to be a really nice question that has really good answers”
So, ownership was really important. The whole concept of Gamification which, that word did not even exist when we started building Stack Overflow, but you have a reputation system of, how much do we trust you as a user and trust comes from other people voting on your content, In other words I vouch for this person, their content is good. If a lot of people do that, then hey, we’re going to trust you to do stuff in the system. The whole concept of, you know, obviously Wikis and editing where you go to a five year old forum post and it’s like this tombstone that nobody’s ever updated, it’s really-really ancient. If you’re lucky, somebody may have added something at the bottom of the updates but we love the aspect of Wikipedia where, everybody is like, “Hey, it’s cool to edit other people’s content, even if I didn’t write it” you know, there is no, this is mine, nobody touch it mentality.
If you look at what programmers do, you don’t want to work with other programmers that are like, “Don’t you ever touch my code” do you? Because that’s not the way programming is supposed to work, you’re supposed to collaboratively build things, you make it awesome, I add more awesome, I fix a bug that you accidentally introduced and you know, it’s all cool, right? You don’t get offended, you don’t get pissed off that someone touches your stuff and we wanted that mentality to go through very much like Wikipedia, on the site. So, its kind of combining all these things together to build this Frankenstein monster of awesome stuff.
Andrew: It wasn’t a monster though because it clearly all worked well together but a lot of people, myself included, when we do what you’re talking about, which is look around and see and pick the best of Digg and the best of that site and the best of that and that thing that we love from years ago that no one is doing anymore and we combine it all, it does become a monster that the audience doesn’t even understand, let alone how to interact with. When you did that, how do you make it all still work well together and feel like its a part of the site that makes sense and not become this monster that no one gets and doesn’t do anything?
Jeff: I think there was a laser like focus on the Q and A page, because if you look at most pages that are generated on Stack Overflow, Stack Overflow was not, I don’t want to trivialize my work or anyone else’s work but it’s really not, at least if you count the number of pages, its not a complicated site, you have the home page, you have lists of questions, couple different variants of that, you have the FAQ, when you have, essentially the main unit of working is the question page right, like I am on a page, I am a user, I type something [??], Oh, I can’t do X, I wish i could figure out X ? enter, and ideally you end up on a stack exchange question page and this is the money page in the system right? and we spend a lot of time thinking about, how can we make this page really-really clear and really simple, like even if you don’t understand all the nuances , all you have this reputation system and you, you know, badges and you have, you know, certain rules about how things are sorted and it’s all very-very complicated but you don’t need to care about any of that right?
All you need to care about is, Oh look, here’s a page, the best answer is at the top and at the bottom, there’s a forum, you don’t even have to register, we don’t require you to register to answer at all, you can just start typing and press submit on any stack exchange site, so we make it, we reduce friction on the thing that we want, which is smart users doing drive by searches and finding this and say, “hey, I happen to know the answer to this question” which happened to me all the time on the Coding horror blog, I would write a blog entry, I don’t write as many [??] blog entries as I used to because the scope is so much bigger of the blog, but I would write about little narrow technical issues that I was having and somebody would just drive by a year later and have the exact right answer. I was like oh my god that’s amazing, right. How does that happen that’s so magical and it works, I think in large part because you don’t have to register and because the topics are so narrow, it’s not like we’re asking, is Brittney Spears cool?
It’s not like a general purpose thing, this is like a narrow technical thing that only five people really ever cared about in their entire lives. So there is not really a lot of incentive to spam that because its ridiculous right, who would? and you get stuff like for example John Carmack of Doom and Quake, answer the question on super user about, he’s doing a lot of [??] stuff and somebody asked some technical question about like, refresh rate and video cards and stuff like that and John Carmack happened by and typed in an answer, right and he din’t have to register and that’s exactly why we designed the system the way we designed it. John Carmack wouldn’t hit a log in wall and say “Hey, you know what? The hell with this. I’m John Carmack and I don’t have time to log in.”
Andrew: I see. So it was just the focus on what you wanted people to see if they came in as complete strangers off of Google. The question essentially at the way they might have asked it in Google and an answer that floated to the top and if that answer changed because the world changed, then the next right answer would flow to the top and you said “What do we need to get to that?. If it’s voting that changes over time, perfect. So we’ll allow voting so that when there’s a new version of Windows that changes the answer to a previous question, a new answer could float above that older answer. If there’s a new person who’s flying to this site and has a better answer, we want to get them to end.”
Basically look, look how tough it is. I’m even having a hard time explaining this but it seem that what I’ve got is you just wanted that clear page that people see when they come from Google and you kept asking yourself what do we need to do to get to that. Do I have that right?
Jeff: You do but let’s go further. When I had algorithms class, the one really useful class I had in computer science was algorithms. And we had a really good teacher which always helps. And one of the things that he brought up that I thought was a really cool way thinking of this was when you’re designing an algorithm the first think you ought to do is think about the God algorithm. If you were God, how would you solve this problem? And that is called the 0 of 1 were you just immediately know the right sort of things. Say you want to sort something and you do the minimum number of operations to sort the list.
So what’s the God page on stack exchange? You type the question into Google and you end up on a page that has nothing on it except exactly the answer you wanted. I mean, literally nothing. It’s blank except for ‘Jeff, do this’, right?
Like right now I’m having a problem with my Nexxus 7 tablet where whenever I try to buy something on the Google Play sort says “An error has occurred” which is spectacularly unhelpful. So the God page for me would be I type in “Google Play error try again ” and it says “Jeff, do this on your tablet”. click here, click here, click here. Bam, fixed, right?
That’s the God page. It has nothing else on the page. And then how close can you get to that? You got the God algorithm which is obviously unattainable. And then you try to do as much as you can to get there. And everything that you see on a Stack Exchange question page is there in service of that goal. Sometimes, it’s there to help the other users answer the question. Sometimes, it’s there to organize the questions. Sometimes it’s there because we need to drive people to the Governmental site, the metas behind the scenes for they can help us organize the questions and if they care about the stuff, figure out why things work the way they do on the site.
There’s election systems. You can be elected a moderator. But I think that’s the way I’ve always thought of it. It’s start with the God algorithm and then just see how close you can get to that.
And the [??] is not bad. It’s question at the top and then immediately under it, a really, really good answer. And the real power of Stack Exchange is because we’re very strict about this to the point that people will call us jerks all the time because we’re so strict about the way things work and how things operate on the site. You get really good answers filtered really fast.
So it’s very, very common when you go to Stock Exchange what do you guarantee? It’ll be fast because we’re fit and speed is family value. We always have [??] for speed. It will load very, very rapidly so even if it’s not correct, you’ll be able to move on very rapidly to the next link on the page, which you can not say about half of the internet. Most of the time when I click on links it’s like “Oh, no. Five seconds of waiting to even figure out what’s going to show”. So it’ll be fast. It’ll be relatively clean, relatively to the rest of the internet. We don’t do a lot of noisy advertising. There’s not a lot of gunk on the page. And it’ll have a clear question and a reasonable answer very near so you don’t have to scroll. You don’t have to scroll.
Think of this in really basic terms, like “Oh, no. I have to scroll, scroll, scroll. I have to go to page two. I have to go to page 3. I have to go to page four.” This is work for me as a user. I don’t want to do that stuff. Just give me the God page. Just put the answer on a page where I can see it. We get very, very close to that if you think that through. And that was, I think always the designing guiding ethic that we use to designing the question page.
Andrew: I remember reading Howard Schultz’s book about Starbucks, the first one. His name is just not coming to me but there was a story about how he was sitting on a Starbucks trying to ponder the question of whether he should allow non-regular milk because non-regular milk might just ruin the whole atmosphere and ruin the authenticity of its coffe. And of course, today you walk in and you can see soy milk and 2% milk and this milk. So we know which direction he went in but it didn’t just happen. He had to really sit and ask himself what are we about? What’s our goal here? How does this milk fit in with our goal?
Did you have questions like that? Was there one big question like that that made you came down to the heart of who you were and what this business was about, early on?
Jeff: Well, there’s always been a lot of pushback because on thing that people love is discussion. People love to just hang out and discuss stuff. And there’s nothing wrong with that per se. Is it wrong to have discussions? But when you adopt the mental of we owe a Q and A system that really does change the focus of what you’re looking at. So everything that happens has to serve two goals. Did you get a better question and did you give better answers? And the odd thing about discussion, as much as people like it, is that it degenerates rapidly. Like discussion is kind of a wild energy that can be very, very hard to harness so our way of looking at discussion is we’re going to suppress it. We don’t really want discussion on our site and this pisses a lot of people off because they want to engage in discussion and it’s very, very entertaining to do so.
And the other thing you look at is what’s the goal of our system? Well, the goal of Q and A system is ultimately kind of learning. You’re trying to learn stuff about the world, how the world works your selfish problem can get solved and also leaving this trail for other people to find. Other people will find your question and they will also have an illuminating answer so that helps them tet things done and this is how you stand on the shoulder of giants and this is how you build things in the world.
But entertainment, which a lot of people love, gets in the way of that a lot. And although a certain amount of entertainment is first tolerated, like you can be amusing in an answer, you can be amusing a question, but if the entire purpose of the question is to amuse and all the answers are not so much about learning but about how entertained were you, this is a very different thing than what we were set out to build. And we spend a lot of time explaining this to people because they love it and they’re like I’m kind of learning. And we have to say to them ‘Maybe a little but only in the way of like in the way you randomly read the encyclopedia. You know, you just randomly sat down and turned to page 12 of the encyclopedia and are going to learn some random fact that may or may not be [??] to me. It’s not really directed learning. Just kind of a random walk mixed in with a whole bunch of LOLs.
Andrew: Jason Calacanis created a similar site at Mahalo. He turned it into a question and answer site and I remember him once tweeting out a question that he asked on his own site. It was ‘What do you think of the new Simpsons’ intro?’ and then he wanted people to go back and forth and engage in conversation. Obviously increases page views, it gets people familiarized with the software. It gets more activity on the site, which maybe brings in new people who will ask more serious questions and I could see the excitement of doing a question like that.
Did you guys ever consider that and have to stop yourself from taking on more traffic? And, if you did, how do you lay down the law like that?
Jeff: Well, it took us a while to figure this stuff out. And we didn’t know all this stuff when we started. Certainly, if you look at the early days of stack Overflow, there’s a lot of questions that were tolerated that now would be closed down instantly. What’s your favorite food as a programmer?
Andrew: And you used to allow that because?
Jeff: Well, we don’t actually allow it. Some of them were grandfathered in for historical reasons like we put a notice on the question that says ‘Look, this is from an earlier era when we hadn’t figured all this stuff out. You can look at this. It’s here because it’s linked to and it has some learning value.’ That one doesn’t actually. Most of the stuff we kind of allow, the criteria we use are essentially what’s on the learning. Can you really honestly learn stuff that makes you better at your job from this question in a reasonably directed way?
Andrew: When you did allow that, Jeff, why did you allow it? Were you trying to figure out where you stood? Was it that you didn’t have enough traffic and you needed to accept things? When you did allow it, what was the reason behind it? I want to understand the thought process that helped you guys evolve so that when we find ourselves in a similar situation, we can call on what we remember you did?
Jeff: Well, I think it’s listening to your community. I advise occasionally start thoughts that are crazy enough to ask for advice and I keep coming back to they’ll ask me stuff like, I don’t want to name names but they’ll say “Shouldn’t we do x or y?” and I’m like “Why are you asking me? Like, why don’t you ask your community what you should be doing?” Because they’re going to tell you, right. These are the people that are invested in what you’re doing and are already there and hanging out. Kind of like being pulled into the gravitational well of your effort naturally. You didn’t have to make them come, they just came there.
And if you’re not leveraging and interacting with your community in a real, honest way to the point that you’ll say no to them, we can’t do that, we can’t do that, I don’t mean like the stupid giant backlog. “Oh, sure. We’ll get to your suggestion” but in honesty, man. Like I really believe in this, honestly. I tried to explain this is what we are, this is what we do. We’re a car. We can’t necessarily be a truck but I’m all for making the car more awesome. Let’s add power windows. Let’s add curve fillers for doing a lot of parking.
Andrew: And so it was your community that said ‘Guys. We don’t want to know what other developers eat. We just want to focus on important things’?
Jeff: Pretty much and also, interestingly enough, we built in a reputation system that incentivizes getting reputation for post. But think about what it means. Say you have 100 reputation for posting a hilarious programmer cartoon compared with this other person who has 100 reputation for posting a really great answer to a Ruby question that gives a really clever technique. Think about that. Think about what’s happening to the reputation system. The cartoons are devaluing your reputation system, because people are getting a reputation for, they didn’t create that cartoon. I mean it wasn’t even really work, was it? It was just a cartoon that someone saw that related to programming, whereas this other person, this is a technical answer about programming, right? Where do you want reputation to come from? And people that have reputation that are invested in the system see this happening, and they’re like, “Wait a minute, this person shouldn’t have the reputation that they have. It’s not the same as the reputation I have.” It’s not fair. The system is not fair, and that’s what they’re telling you, and that’s the implicit message.
Andrew: How did they tell you this in the early days? What was the means that they used to communicate this stuff to you?
Jeff: Well, people always ask, “What’s your biggest mistake?”, and I love that question now because I have a really good answer for it. On Stack Overflow I had a very fight-clubesque way of viewing Stack Overflow, which was, “we’re here to talk about programming. Do not talk about Stack Overflow, OK? Because I don’t care. We’re here to talk about programming.” But, people wanted to do it. It kept coming up, it coming up, and every time I told people not to do it, they just couldn’t help themselves. They had to talk about like, “What if an upvote was worth 5?”, or technical details about the system, “Why are we accepting this question?” Over time, the community eventually said, “You know what? Jeff doesn’t like this. We’re just going to set up a place to discuss Stack Overflow offsite in PHBBB,” which is one of the traditional forum softwares. That was like a dagger in my heart, because I hate PHBBB.
One of the reasons we created this was partially a reaction to the terribleness of forum software. I said, “OK, you’re right. You’ve forced my hand. You’re absolutely right.” We need a place for the community to discuss the place. We need a Meta. We need a discussion site. At that point I set off, and we created a branch at the engine; well, it’s the same engine, actually, and not a branch. It was kind of like on Wikipedia where you have the talk page. When you go to an article on Wikipedia, there’s a place you can go to, to talk about that article, but it’s behind the scenes. It’s sort of behind the curtain, and Meta is that. It’s the place where we go to discuss, you know, why do we have all these questions about programmer food? Does that even make sense here? It’s great because this is fundamentally a thing created of and by the community.
Now, if we were selling a product that we created, then I think the dynamic would be different, but you’ve got to realize, who is doing the work in the system? It’s not us. I’m not in there typing in programming questions all day long. I mean, I have some on Stack Overflow, but it’s the community. The community is doing all the work, so why shouldn’t they have a say in what happens, a really powerful say in what happens on the site, and why shouldn’t we be listening to them? Because if they’re not happy, how do we even have a site?
Andrew: You know, it occurred me that I didn’t ask you one of the basic questions that I ask all entrepreneurs, which is how did you get your users? How do you get so many people who are willing to come on and answer difficult questions?
Jeff: Well, we cheated. This is like the answer to how to be successful at digital music distribution? Step one is to be Radiohead, right?
Andrew: [laughter]
Jeff: And then everything works for you. Joel and I cheated mildly because we had these big audiences of programmers that respected us and listened to us for better or worse, and we said, “Hey, we’re creating this thing. Stack Overflow. Please take a look at it.” You point the rays of all these eyes at this thing. You do get a big head start, but I still maintain that it doesn’t matter. If what you’ve created isn’t fundamentally working, eventually that just wears off. That’s not a permanent advantage. You can’t make everything instantly successful by just having a big following. I think we had a big following, and that helps kick start it, but fundamentally it was a good idea. It was executed well, and it was a sane reaction to some of the evil that was happening on the web.
That set up a really nice dynamic of explaining it and also like good-guy- versus-bad-guy, which the world doesn’t really like that. Expertexchange is still there, although you never see them in search results. I do want to point that out we did kind of actually vanquish them in terms of like you just don’t see Expertexchange in search results anymore. It’s a big pie. I never said, “We’re going to destroy Expertexchange.” They’re going to continue to have a business doing whatever they do for a long time. The pie is enormous. I never viewed this as a zero sum game where we must win at the expense of others. I think the pie is enormous and getting larger. It’s just carving a section of the new pie is the kind of the way I look at it.
Andrew: At one point you guys were going to allow anyone to create their own version of StackOverflow. I created one for my site. Others did it, too. There was Startups.com where you could go and ask questions about startups, etc. I asked Joel on the day that you guys closed it down to come on and talk about why it didn’t work out, and you trusted me enough to come on and talk about openly why it didn’t work out. I don’t want to spend too much time on it, but can you just tell me from your perspective where the idea came from and why you think it didn’t live up to expectations?
Jeff: Well, I was never really a fan of that model because launching a community… Essentially, it was OK. Cut us a check, give us a credit card and then bam! You have a stack and community, but money is many ways is like the least important aspect of community.
One example I can give is many years ago when I started taking in advertising on my blog, it was actually starting to be a lot of money. I was like, wow, this is a lot of money. I had a full-time job, so I was like, what am I going to do with all of this money? I should give some of this back to the community that made me what I am, like made the blog really popular. So, I said, “I’m going to donate $5,000 to an OpenSource.net project because I thought that was the right thing to do.
First, it was just hard to figure out where to give it to. That was hard. And then, the project I gave it to couldn’t really figure out what to do with the money I gave them anyway. I just got an email. He was a super nice guy. It was a [?] turn Wiki. They eventually shut down the project because it just didn’t work out for whatever reason. So, this is the complexity of money. Money doesn’t really tell you anything about anything, like it doesn’t tell you if you have a good idea for community. It doesn’t tell you if you have people that will care about your community. It doesn’t guarantee that people will understand that this is a forum, like we have certain rules about the way that stack exchange works, that are fairly strict. It’s very high signal, low noise and that takes discipline. Are you willing to enforce the discipline it takes to achieve a system like this?
None of that had anything to do with whether you can give us a credit card to launch a site. To his credit, Joel figured this out because that didn’t really work and created some sites that were kind of a little bit embarrassing. Like, one of the ways Joel and I look at this is when new sites go online, are you proud to be associated with the content that’s being generated here or are you embarrassed? Too many were falling into the embarrassed category on stack exchange 1.0. I don’t want to name names, but generally it’s hard to build community, man. It’s a hard problem.
Andrew: But there are so many people who have V bulletin sites still to this day that really should be using a question and answer format because that’s the main reason why people go over, but they build communities on that software. There are people who are building communities in some way on WordPress. Why could they do it on Stack Exchange?
Jeff: Well, it’s not exactly the same thing. There is a subtle difference, but Q and A is pretty directed, and not all communities can survive on Q and A alone. Let me give you an example of one that definitely can that we just launched into final design, Mathematica.exchange. Now Mathematica was technically on topic at Stack Overflow because it’s a programming tool for sure. It’s very math oriented, but the community was so strong. No, no, no. We need our own place. We need our own place to go. We need our own place to ask questions and stuff.
For a technical topic, like Mathematica, Q and A is awesome because it’s just the facts, ma’am. Give me your questions, here’s the answers, but compare that with something like Legos. If you have a community of people that love Legos, can they express everything that they need to survive as a community in terms of questions and answers? I would say no, almost resoundingly no because if you do Lego, one of the common things as an example, there’s tons of things you’re going to want to do is you’re going to go build something awesome in Lego. And you’re going to want to share that with other people on the site. So, there should be a thread titled “Share your awesome Lego projects” and all you do is post a picture. “Look, I built this thing. Here’s some detail about it”, and other people are going to go, “Holy crap, that’s awesome.”
I’ve seen threads like this on many forums that were amazing. People created stuff that was just incredible because people aren’t going to share stuff unless they think it’s pretty good anyway because they don’t want rejection. But threads like that are completely forbidden on Stack Exchange. You just can’t do it. There’s no question there. You can’t say, “Look at this thing I created”, but for a lot of communities that, amongst many other things, is what they need to survive as a community. Q and A alone is not enough, and I said for a long time that I was OK saying, “Our engine is good at certain topics.” We’re now figuring out what that is. The community is telling us, let’s try this, let’s try this. Some of those don’t work, and that is by design. We’re not going to adjust the mission to make it discussiony. We’re just going to say, “Hey, that’s not a topic that works on our engine.”
The good news is there is a huge set of topics that I think would work brilliantly on our engine, and I don’t think there’s any finer engine in the world for technical Q and A, in the world. I legitimately think this is a fantastic mouse trap we’ve built for that particular use case, and it cannot be defeated, basically. It’s a nearly perfect system for that kind of topic, but the weakness of such a system is – that’s a subset of altogether different topics.
Andrew: You guys raised, according to Crunch Base $18 million. Now, actually is that number right?
Jeff: I don’t know. I don’t like talking about that because I can never remember what I’m allowed to say, so I just don’t say anything.
Andrew: So, let me ask you this about…
Jeff: I’ll tell you this it’s not totally off base. It’s not crazy.
Andrew: Here’s something that you might be able to answer. Well, on the day that you announced that you were teaming up with Joel Spolsky to launch this, you said about him, “He is very much living the dream. He founded a company with the express intent of not cashing out with VC money.” Now, once you raise this much money, you’re basically saying we have to cash out. How did that make you feel, the discussion about raising money?
Jeff: Well, VC is really about the money, obviously, because VC is money, but a lot of very rich people are also very philosophical about what they’re doing. Meaning, they’re not about, “I want to maximize all my return,” they said, like Joel and I do, that “hey, we want cool stuff that makes the world better to exist in the world,” right?
Andrew: Mm-hmm.
Jeff: And that’s their priority number one. Now, if they can make money while doing that, that’s awesome and that’s also a goal. Those are the type of venture capitalist that you want to be in bed with; not the evil, let’s get blood from a stone, how do we turn maximum profit, how do we extract the maximum amount of income, and all the bad ways of looking at the problem. So, to me, VC, the more we looked at it we realized there were some really smart VCs that are really about using their money to create a better world, period. That’s sort of the way I look at stack overflow and Stack Exchange and Stack Overflow. It was never about the money; it was about, “I want this to exist for selfish reasons. I want this to exist because I think the world needs it, and I think it makes the world better, and I think glorification of math, physics, programming, Mathematica is a net good for the world.” People who love mathematics, creating awesome information about Mathematica, makes the world a better place. It makes my internet better for me, so I want more of that, please, you know.
Andrew: You know what, inherently, though, they do want an exist, but it seems like that’s OK, because you got an exist. I’ve never raised money for a company, so tell me. I don’t know enough about vesting from the point of view of an entrepreneur who has to vest, but did you vest after four years of being at the company?
Jeff: Well, again, I can’t really talk about all that stuff, because A) I can’t remember what I can and can’t talk about and, B) I don’t want to step on any weirdnesses, but I will say that when you leave, as I did, new options open up, and it seems to be just the discretion of the board, pretty much, as to what they can and can’t do, from what I can tell.
Andrew: Without being specific about yours, what kinds of options open up?
Jeff: Again, I’m real tentative about being specific about anything, but essentially anything short of complete madness, right? Because, you’re leaving and you’re not going to be back, and you’re not a part of the power structure of the company anymore. I’m still a shareholder, but I have no say.
Andrew: Oh, I see. I thought you meant that any kind of offer could be made available to you, and then you ended up with some great offer. You’re saying that it’s out of your hands. You can’t decide anymore what happens with your . . .
Jeff: Well, once you leave there are sort of exit options.
Andrew: So, did you cash out a little bit so that you could take some space and get a cushion?
Jeff: Again, I can’t be specific.
Andrew: You can’t say that.
Jeff: I can’t be specific about anything that happened, but I did completely leave the company, and that opens up options you don’t normally have.
Andrew: I see, OK. All right, I get it. I’m trying to come up with a question that will call out what you learned. How are you different? What do you know now about running a business that you didn’t know before? What’s the most important thing, I should say?
Jeff: The most important thing is your team, without a doubt.
Andrew: Your team?
Jeff: Absolutely.
Andrew: You’ve never run a company like this before. How did you figure out how to find the right team, how to run a team, how to say things that were awkward to people? How did you become a good manager?
Jeff: I think part of it is that I was always kind of a good communicator, because I have my blog, obviously, and I had written and people liked the way I was writing. I think communicating and writing is such a powerful skill because it’s such a great way to sort of get your thoughts down and actually understand the way you think and communicate that to other people in a way that they can understand. Already, I had the advantage of being just a pretty good communicator, and I’m sort of a good evangelist. I like to motivate people. I like to be on a mission from God, basically. That is a strength that I have, and I find that, in this case, I got lucky because I had friends that were available, but I do think you can attract people to your cause because it’s inherently a good, noble and just cause, and it’s the kind of thing that needs to exist in the world, and a certain subset of people are going to be really into that, and say, “You know what? That’s a great mission. I want to be on this mission with you.” That’s the way I look at this. You’re essentially attracting people by virtue of the mission being a good, noble and interesting mission. Now, I don’t know how that works when you run, say, a company that does mail filtering or something that’s not necessarily that inspiring. But for me, that’s how it works.
That’s how I attract people, and that’s certainly. Now I’m not going to tell you that scales, because it probably doesn’t, actually. And certainly, as the company got to be, Stack Exchange is now, I think, 50 people, there was a huge set of people in New York doing sales and development that didn’t know me very well. And it kind of does break down a little bit, because I wasn’t there, and it was a distributive team, I’m not going to say that system scales, but I will say that you can build some pretty incredible things with a team of five or six people. Like very incredible things. I don’t think you necessarily need a ton of people to build amazing things anymore.
Andrew: Yeah, 50 isn’t even that many, considering how many people are on the site. I remember I took this weekend to learn how to, I took a triathlon weekend training program. When I was done with that weekend, I could immediately go back to Central Park and see all the people who just didn’t know how to use their gears properly, for example. The mistake just stood out in my mind and if they did, they’d be able to ride faster with a lot less effort, without spinning their feet over and over on the pedals.
Now that you’re out of Stack Exchange, do you see something like that, where you’re looking at other entrepreneurs and going, “Oh no, they just don’t get this one thing that would make such a big difference, or this other thing over here that wouldn’t change everything, but would still make their lives easier,”?
Jeff: I’m really hesitant to give blanket advice. I essentially try to communicate what I did that worked for me, but I have this thing called the Jared from Subway problem, where Jared from Subway lost 250 pounds by eating Subway subs, which is kind of crazy if you think about. So do you then go around and say, “You know what you should do? You want to lose some weight? Well, I’ve got a plan for you, buddy. You should go to Subway, buy a sub everyday.” It’s like, well, a lot of things. Different things work for different people. There’s so many ways to succeed and fail.
Andrew: So what worked for you that wouldn’t necessarily work for us? What did you get out of it, it would work for you?
Jeff: Well, I think doing things in public, I tried to summarize this in sort of How to Stop Sucking and be Awesome Instead. But certainly, a couple of things. One is do a lot of stuff in public, where you’re getting a lot of feedback all the time, because people will tell you, give you really useful information about what you’re doing, whether it’s good, bad, ugly, all that stuff. The sooner you can get to that feedback cycle and start turning the crank, is extremely powerful. If you can optimize for turning the crank on feedback, you’re going to, it’s going to be hard to fail, because you may not build Facebook, necessarily, but your odds of failing go way, way down. I think that’s very, very important. And then something I touched on too, just have a mission that not only you believe in, but you can communicate as sort of a net good to the world, like mission that matters, and when I say matters, I mean, how do you define that? One way to look at this is saying “What have you done to help someone else today?”
If you look at Stack Overflow and you look at Stack Exchange, it’s all about people helping other people solve problems. That’s what makes it so wonderful, right? It’s not because I’m awesome, or Stack Exchange is awesome, it’s because people are helping each other. This is a system that incentivizes people to help each other solve problems in a responsible way, not with the help vampires and all that stuff. But people that are willing to put forth the right amount of effort, because we have a disciplined system, will get really, really great help. That’s what you’re signing up for when you go to Stack Overflow or Stack Exchange. When you work at the company, that’s what you’re signing up for. It’s like you’re building tools that help people help each other. They help them do things for other people, so it’s not a selfish thing.
Andrew: One of the things that happens when you build in public and you crank the feedback mechanism is you start to identify your own personal weaknesses. Frankly, people start yelling all kinds of weaknesses at you that may not even be true, but you do start identifying your own shortcomings or your own challenges. What are some of yours that you’ve noticed now, after running Stack Exchange, after blogging in public?
Jeff: Well, I think caring about stuff is double-edged sword, right? Like Steve Jobs would cry all the time, which is crazy, for a grown man to burst into tears about some of the relatively silly stuff that went on. It’s not that big of a deal. But the flip side is that he really cared super, super deeply about that stuff, and that’s what made him really good at it. And I think that’s the downside, it’s like religion. You become evangelical to the point that you have jihads, right. “I believe this so deeply that I’m going to start attacking people that don’t believe the same thing that I believe.” That’s a very, very easy pitfall to fall into and also very poisonous and dangerous, all the downsides of religions without all the upside. So, believe deeply in your mission, but do not be totally dogmatic about it. And of course, having a sense of humor helps immensely. For me, when people attack and they’re like, “You’re system is crazy. It doesn’t make sense,” its about humorance. It’s like, “Fine.” You’ve just got to take it in stride.
Andrew: And you can brush it off right now, are do you still want it?
Jeff: That’s right. For the most part, yes. But like anybody else, somebody is wrong on the internet addiction. It’s like we’re “Oh, this person is wrong on the internet. Now I have to go tell them all the ways in which they are wrong in great detail.” So I think that’s a downfall. That’s sort of the downside of being very evangelical about the mission is you sort of can fall into those patterns and you have to be aware of that and moderate that a little bit. That’s the flip side of being super excited about something.
Andrew: Is there a question that you wish I would have asked, that I should have asked you throughout this interview?
Jeff: No. I don’t think so. I think you had great questions, actually.
Andrew: All right. Oh, the name of the book. It’s “Pour your heart into it” by Howard Schultz. I know you are a big reader. Anything that you recommend?
Jeff: Well, all my books are simplistic.
Andrew: Isn’t that the reason why you started this site?
Jeff: Kind of. It was that programmers actually don’t read books anymore. But there’s a certain type of book, you have to put a caveat around that. Programmers don’t read very narrow, technical books anymore because it’s faster to get answers online. It’s actually more efficient than it is to, say, go to a book, look up the table of contents, find out the section that you need, all that stuff.
But philosophical books about why we program, how we program, like “Code Complete”, “Don’t make me Think”. Those are still fantastic books to have on your book shelf. It’s the fat “How to Program Python in 24 hours” I don’t think you need that book anymore. I think that’s what Stack Overflow does so I want to put a caveat around that.
But the one book I do want to recommend and it’s technically a book about parenting but it really isn’t is “How to talk to Kids so Kids will Listen” and I blogged this on CodingHorror as How to talk to Humans because when I read this book I realized that the way we talk to children, because children have no defense mechanisms. They don’t really understand the way they feel. They don’t understand anger. They don’t understand fear. They can’t control themselves, basically. There’s little balls of [??] and learning to deal with children and understanding all these things they go through and how to talk to them in a way that won’t sort of make them blow up, turns out to be really effective with adults as well.
Because a lot of adults haven’t full grown up. I’m sure I am in this category as well. You never really, fully grow up. Everyone’s still little bit of a kid inside and some of these techniques are amazing. And one of the simplest ones that you talk about a lot is you avoid negation. Because a lot of times with children they’re doing something and you want to say ‘Just stop doing that. Just don’t do that’.
But that doesn’t actually work very well and that tends to make them blow up but what works a little bit better is when you explain the reason why you want to do them. “Don’t run out on the street. If you run out on the street, I’m going to spank you” works less effectively than “Please don’t run out on the street because if you run out in the street a car may hit you and then I might be sad because you would be hurt and that would make me very sad”. That works a lot better, right? But as an adult you go quickly to the negation. And it’s kind of an anti-pattern. The book is full of really great examples like that of stuff that I was doing without even thinking about it with my own child and it kind of carries over into your relationship with adults. So I kind of recommend it for everyone, whether you have kids or not.
Andrew: You know? I read a little bit of it because A.J. Jacobs said that he used it to talk to his wife but the thing that stood out for me was when a kid says ‘I’m not cold. I don’t want a sweater’ and you say “Of course you’re cold. It’s snowing outside. You need your sweater.”
Jeff: Negation.
Andrew: Sorry?
Jeff : That’s negation, right off the bat.
Andrew: That’s negation and instead you’re supposed to listen and then what was it that she said, that the woman said that you’re supposed to respond and then express how they feel and then tell them how they feel.
Jeff: One of the craziest things about that is just identifying someone’s feelings, echoing the feeling back to them, lets them reach their own solution more effectively than handing them the solution in a silver platter.
Andrew: Can you do that in real time, when someone says something to you, can you not negate it if it’s crazy and just express it back to them to show that you understand it?
Jeff: I can with kids. I’m struggling. I’m still trying to adapt that to adults because I’m Mr. Efficiency. I have very little patience and this is a huge crippling character flaw with children because with children you need infinite patience. Ideally, more than infinite patience.
So I get impatient and I want to rush directly to the solution, which I kind of intuitively have sort of an idea. This is a really good solution, let’s just go to that immediately. I don’t want to have all the interviewing discussion like let’s do this, do this, do this. I view that as a waste of time. But in a democracy, the system where people need to figure this stuff out, you actually kind of need that level of discussion, even if you know the answer is actually the perfect answer, you can’t put it forth too rapidly. You kind of have to heard the community along and sort of let them reach the solution. I think it’s particularly frustrating for impatient people like myself.
Andrew: And finally, how are you going to figure out what you’re going to do next? Do you have a process?
Jeff: I’m already on the next thing.
Andrew: Did I miss that on your site? What’s the next thing?
Jeff: I haven’t talked about it because it’s secret.
Andrew: Oh, OK. So how did you come up with it?
Jeff: Let’s go back to the Podcast. Think about the way we started “Stack Overflow.” We identified something on the web that needed to change. We viewed it as a strong villain figure that everybody vilified and hated. If you think about the software that you use on the web and the websites that you use, there’s still a lot of software out there like this. It’s kind of a forgotten software category in my opinion. But it won’t be when we’re done, I think.
Andrew: So you identified software that people are neglecting and you said, “I think I could do it better.”
Jeff: Yeah, people are neglecting and also a lot of people hate – and for good reason. But, yeah, and it’s, again, selfish because I’m like “Wow! I’m tired of clicking on these links that end up on pages that suck. How do I fix this?”
Andrew: So did you already start showing it to people and getting feedback?
Jeff: A little bit. We’re still building it. As I talked about at the start of the show, you get tired of explaining it after certain points, it’s like “No, just go look at this thing.” So we’re building up enough of it so that I can stop explaining it and just show it to people and they’ll see what it is. But I can say it’s a fully open source thing meaning if you want it, you can just take it, run it, grab it, install it on Amazon EC2, install it on your PC at home, wherever. It’s fully 100 percent open source.
Andrew: I’m looking forward to seeing it. Until then we’ll all just go to codinghorror.com. Some of my favorite articles on there are the ones about writing. I guess it’s in the early days you were writing more about how to blog even if you think you’re a bad writer. Why you should keep on blogging. I love those. I eat those kinds of articles up!
Jeff: Yeah, well it’s practice. It’s like anything else. You’ve just got to do it to get better at it.
Andrew: Yeah. All right. Well, thank you again for doing this interview.
Jeff: Yeah, thanks very much, Andrew. It was great.
Andrew: Thank you all for being a part of it. Bye.