Stack Overflow Podcast - Episode #70

Summary

Stack Overflow Podcast – Episode #70 (April 19, 2011)

Hosts: Joel Spolsky & Jeff Atwood

---

Episode Overview

In this vibrant on-the-road episode, Joel and Jeff dive into the whirlwind of recent Stack Overflow developments. Broadcasting from Austin, Texas during Dev Days, they unpack the launch of new products (like Kiln and Stack Overflow Careers), discuss the rapidly diversifying Stack Exchange ecosystem (including unexpected breakout sites), and debate the perennial conundrum of reliable Wi-Fi at tech conferences. The tone is witty, candid, and hands-on, as both hosts share behind-the-scenes insights into product evolution, community challenges, and the real-life realities of running developer-focused platforms.

---

Key Discussion Points and Insights

1. Austin as a Tech Hub & Dev Days

• Why Austin? Joel highlights Austin's rich tech history and culture, home to giants like Dell and IBM, dubbing the direct Austin-San Jose flight the “Nerd Bird.”

  “Austin's a big tech city... This has sort of become a pretty major tech center.” – Joel Spolsky (01:27)

• Dev Days Recap: The hosts recall the hectic lead-up and successes at recent developer events, made more complex by launching multiple new products at the same time.

  “It was kind of hysteria leading up to that because there are basically four new products that all had to be ready for dev days.” – Joel Spolsky (02:54)

---

2. New Products from Fog Creek/Stack Overflow

• Kiln:

  • A hosted Mercurial-based version control and code review product.
  • The logo, chosen via community contest, features a “crazy dodo bird,” echoing FogBUGZ’s branding.

    “It's a crazy dodo bird. He's sort of insane. Yeah.” – Joel (04:00)

• Fog Creek Training Series:

  • Corporate training videos in beta; official release planned for Jan 1.

    “There's a website for that… and that's also sort of in beta in the sense that... they're still in post production.” – Joel (03:05)

• Stack Overflow Careers:

  • Introduced as a better developer recruitment system, leveraging Stack Overflow’s existing talent.
  • CVs (not just resumes) can be created, with public/private options and “vanity URLs.”

    “CV is the more European term for resume. But there's a reason we didn't use the word resume. …It’s more open ended.” – Jeff (27:23, 27:46)

---

3. Stack Exchange Explosion & Community Diversity

• Rapid Site Growth:

  • Open beta allows communities of all interests. Notable sites include:
    • Epic Advice (epicadvice.com): A World of Warcraft Q&A site, growing faster than expected.

      “The number one is called Epic Advice... it's a World of Warcraft Stack Exchange thingamajiggy.” – Joel (06:11)

    • Answers.OnStartups.com: For entrepreneur/startup questions.

      “...a ridiculously successful Stack Exchange where people are asking questions about startups.” – Joel (06:54)

    • Moms4Mom: Parenting Q&A; feedback from non-geek users is reshaping usability concerns.

• Accessibility & Usability:

  • OpenID integration has improved; most new users now just click the Google button, making login seamless for broader audiences, including non-programmers.

    “Google came in really late to the OpenID party, but… it's like 56% of all users use Google. Basically you just click the giant Google button and you're done.” – Jeff (09:35)

---

4. Competition, Community-Building, and Site Proliferation

• Competition Among Stack Exchange Sites:

  • Hosts discuss overlapping themes (e.g., multiple SQL sites), and how only active, well-populated communities will thrive.

    “It's not about having the topic. ...You somehow have the wherewithal to get a group of people to all pile into a website and start answering questions.” – Joel (12:28)

• Community Doesn't Build Itself:

  • The “platform doesn't magically create community”; driving sustained participation is still hard work.

    “Stack Exchange does not magically create community. That's still hard work.” – Jeff (16:45)

• Niche/Small Support Sites:

  • Even a site run by a single knowledgeable tech support agent can succeed for focused product Q&A.

    “One really knowledgeable one... People can ask questions and get an answer within 24 hours... that's already more valuable than using phpb.” – Joel (20:19)

---

5. Stack Overflow Careers & CVs: A New Approach to Tech Hiring

• Motivation & Philosophy:

  • Traditional resumes fail to capture developer abilities. Stack Overflow’s new model leverages community contributions, profiles, and reputation to connect developers with “companies that care.”

    “You can tell a lot more about a person than you can from a traditional resume.” – Joel (24:38)

• How It Works:

  • Programmers create public/private CVs (resumes), with vanity URLs.
  • Option to be discoverable by employers, flipping the script– companies pursue programmers.

    “Smart companies should be pursuing good programmers and not really the other way around.” – Jeff (30:11)

• Why Charge Developers?

  • $29/year fee for job-seeking CVs is meant to reduce noise (less spam, more serious candidates); employers pay much more on their side.

    “We have to promise [employers] is that these are real candidates... not just kind of entertaining offers... they're actually kind of serious.” – Joel (39:59) “One of the reasons we're charging money is honestly just to have a barrier to say you're either serious or you're not serious.” – Jeff (41:19)

• Refund Policy & Fairness:

  • Generous 90-day refund policy, to ensure no one feels trapped or unsatisfied.

    “If you decide, you know what, I don't like this. No questions asked, instant refund, no problem whatsoever...” – Jeff (45:59)

• Recruiters:

  • Only allow “retained search” recruiters (who work exclusively for one employer and disclose the company); not interested in contingency recruiters or resume spam.

    “We're not interested in this kind of thing happening... we're going to require [recruiters] to reveal the name of the company...” – Joel (38:54)

• Privacy Considerations:

  • Employers you listed as past/current will not see your profile by default; granular public/private controls for all CV content.

    “Everything's private by default 100% of the time. You can opt in to non privacy or making it public.” – Jeff (53:04)

---

6. Open Source, Scale, and Product Philosophy

• Stack Exchange as a PHPBB Alternative:
  • The goal: replace phpBB as the default Q&A/discussion forum; open access and customization foster innovation.

    “I hate PHPV so much that I just want to destroy it.” – Jeff (12:30)

• Scale and Small Sites:
  • Low-traffic “support” sites can thrive with just a handful of experts, using Stack Exchange as a dynamic knowledge base.

    “We're posting our own questions and answers in there... for Google to find and it can be edited and it's sort of better than a wiki in so many ways.” – Joel (18:27)

• Product Evolution:
  • Iterative approach– launch quickly, react to feedback, and course-correct as needed.

    “We're being poster children here for the benefits of shipping early and then adjusting course rapidly.” – Joel (56:03)

---

7. Wi-Fi Nightmares at Tech Conferences

• Tech Conference Wi-Fi:
  • Recounting the infamous Boston Dev Days DHCP failure and the universal struggle of making Wi-Fi work for hundreds of geeks.

    “Suddenly the 254th person or something tried to log on and the DHCP server fell down because it just didn't have a large enough pool of IP addresses.” – Joel (59:18)

• The Stack Overflow Network as a Knowledge Repository:
  • Joel points to a canonical Server Fault answer on providing decent Wi-Fi at conferences (Question #72767), demonstrating the community’s depth and resourcefulness.

    “It is a friggin awesome answer. ...the canonical place... That's the great thing about the Stack Overflow Server Fault model.” – Joel (61:13)

---

Notable Quotes & Memorable Moments

• On Austin’s Tech Scene:
  “The flight... from Austin to San Jose is called the Nerd Bird.” – Joel (01:38)
• On Logo Contests:
  “10 minutes before the end… I was like, 'Shut this down. We're not using any of those... over my dead body.'” – Joel (04:25)
• On Community:
  “Stack Exchange does not magically create community. That’s still hard work.” – Jeff (16:45)
• On Careers/Recruiting:
  “If you have a list of people on Stack Overflow... you can go and look at their profiles... you can tell a lot more about a person than you can from a traditional resume.” – Joel (24:38)
• On Privacy:
  “Everything’s private by default 100% of the time. You can opt in to non privacy or making it public.” – Jeff (53:04)
• On Refunds and Customer Service:
  “Much rather have a happy person not paying you than an angry person who has paid you.” – Jeff (46:54)
• On Product Iteration:
  “There’s only so much thinking... and coding you can do. You gotta ship it and then see what happens and then adjust rapidly.” – Jeff (56:45)
• On Conference Wi-Fi:
  “You get to one of these venues and you find a person that knows nothing about computers and they are in charge of the Wi-Fi…” – Joel (59:25)

---

Timed Highlights (Timestamps in MM:SS)

• 01:27 – Austin’s status as a tech hub
• 02:20 – Dev Days product rollouts
• 03:02 – Overview of Kiln and Training Series launches
• 03:55 – Kiln logo contest adventures
• 05:37 – Stack Exchange open beta and site diversity
• 06:11 – Epic Advice (World of Warcraft) surges in popularity
• 09:35 – Google OpenID dominates Stack Overflow logins
• 16:45 – “Stack Exchange does not magically create community”
• 20:19 – Support sites can work with a single expert
• 24:38 – Resumes versus Stack Overflow reputations
• 27:23 – Launch of Stack Overflow Careers CVs
• 30:11 – Smart companies should pursue programmers
• 39:59 – Why developers are charged for CVs
• 45:59 – 90-day, no-questions-asked refund policy
• 59:18 – Wi-Fi failures at Boston Dev Days
• 61:13 – Server Fault becomes the expert resource on conference Wi-Fi

---

Episode Tone & Style

Conversational, candid, and laced with infectious geek pride, Joel and Jeff blend humor with pragmatism. Their playful ribbing (especially around logo design, support forums, and customer policies) makes the episode engaging, while their honest reflections on building products and communities will resonate with builders everywhere.

---

Final Thoughts

This episode captures a pivotal moment in Stack Overflow’s evolution, brimming with transparency about mistakes, product changes, and community lessons learned. The hosts’ commitment to openness, user empowerment, and continual iteration shines throughout. Whether you’re a developer, a product manager, or just fascinated by how online communities work, the candid discussions in this episode offer a wealth of insight, practical advice, and a few good laughs.

---

For more details and links to discussed resources, visit blog.stackoverflow.com.

Transcript

[Phil Windley]

Up Next, on episode 70 of Stack Overflow, Joel and Jeff discuss dev days, the diversity of Stack Exchange sites, the debut of CVs and careers on Stack Overflow, and the viability of Wi Fi at tech conferences from IT conversations. Hi, this is Phil Windley. Today I'm excited to bring you another great program from Stack Overflow with Joel Spolsky and Jeff ATWOOD Here on it conversations.

[Conversations Network Announcer]

The Conversations Network is a 501c3 nonprofit, and we need your help. For a tax deductible donation of as little as $5 per month, you can support this channel and the rest of the Conversations Network. So please visit conversationsnetwork.org to become a member and help us continue to bring our programs to the world for free. Our audio files are delivered by Limelight Networks, the high performance content delivery network for digital media.

[Phil Windley]

And now, here's Stack Overflow.

[Joel Spolsky]

We are doing a very bizarre podcast this week. I am on the road in Austin, Texas, at the Grand Super Extra Hyatt Hotel.

[Jeff Atwood]

It's sufficiently luxurious for your needs.

[Joel Spolsky]

It's got a nice atrium.

[Jeff Atwood]

Good, that's important. So, you know, I'm interested why you picked Austin as a destination.

[Joel Spolsky]

Well, Austin's a big tech city. It's got. You got Dell, Compaq, were both created over here. Compaq was Compaq Austin, or just Texas in general. But Dell, it's famous for. And IBM has long had a lot of labs and stuff. You got the University of Texas. So this has sort of become a pretty major tech center. In fact, the flight. The flight that. The nonstop flight from Austin to San Jose is called the Nerd Bird.

[Jeff Atwood]

That's funny. A little bit of trivia. The Nerd Bird. Nice. Cool. I didn't know that. I mean, I knew Austin had some interesting things going on there, but I didn't realize it was such a tech hub.

[Joel Spolsky]

Yeah, it's sort of the tech hub of Texas. Tech hub Texas. And so it's, you know, it's not like your Silicon Valley kind of tech hub, but it's, you know, a good third or fourth in the country. Or fifth. Right? Definitely the top 10.

[Jeff Atwood]

Yeah. So tomorrow, that's the Austin Dev days. How was the first dev days? Did you want to summarize it for people?

[Joel Spolsky]

Yeah, the first dev days was awesome. For those of you that weren't there, there's a lot of reviews you can find on the Interwebs, but we had a bunch of, let's see, Speakers. I can't remember. It's all just a blur. It's A blur of audio, visual, adjusting mic levels, getting displaying things and putting things on the screen and running up and giving little talks. It was kind of hysteria leading up to that because there are basically four new products that all had to be ready for dev days.

[Jeff Atwood]

That's true. We might as well just list them here.

[Joel Spolsky]

Might as well just list them here. So Fog Creek is a new product that Fog Creek is coming, coming out with, which is now in beta. It's called Kiln. It's a version control and code review system. So it's going to be a hosted version of Mercurial for those of you familiar with the Mercurial version control system. So that's one. Kiln was the second thing. Second thing was a training series for those of you that want to get corporate training videos brought to you by Fog Creek Software. And there's a website for that training.fogcreek.com sort of in house corporate training stuff. And that's also sort of in beta in the sense that, you know, we put up a website and we're taking orders, but the final. They're still in post production, so they're not going to be ready until January 1st. The third thing is stack Overflow Careers, which we should talk about extensively because that's extensively made. Your new Stack Overflow thing.

[Phil Windley]

Right.

[Jeff Atwood]

Well, I'm actually looking at Kiln because you guys had a logo contest. I actually hadn't seen the outcome of that.

[Joel Spolsky]

Yeah, we got some kind of like the logo. It's a crazy dodo bird. He's sort of insane. Yeah.

[Jeff Atwood]

Well, I like the. It's sort of similar to the Fog bugs. Kiwi.

[Joel Spolsky]

That's right. It's some kind of a bird that can't fly. And.

[Jeff Atwood]

Well, those logo contests are really scary in the beginning because you get the craziest button. You're like, this is. This is not gonna end well. And then it's true.

[Joel Spolsky]

Sometimes I actually, I actually remember, but it did.

[Jeff Atwood]

It came out good.

[Joel Spolsky]

Yeah. Only just. I mean, that was really the only one that was. It was acceptable. And I do remember like, like 10 minutes before the end of the logo contest telling people, shut this down. We're not using any of those goes over my dead body.

[Jeff Atwood]

Well, that's the problem. Well, that's the side effect. The shotgun approach. The shotgun approach has a very, very low hit rate. I mean, I remember being satisfied with like 5% of the entries and we had hundreds of entries.

[Joel Spolsky]

Yeah.

[Jeff Atwood]

And I'm talking about just satisfied to the point that I would even consider them.

[Joel Spolsky]

Right. Of Course.

[Jeff Atwood]

So that's just the nature of the beast with that kind of thing. But yeah, I like the logo that you have, so that's good.

[Joel Spolsky]

Yeah, he's our crazy new dodo bird, which will be the Kin logo Kill logo. Stack Overflow, there's another thing. Oh, Stack Exchange, which was officially launched and Stack Exchange we've been talking about here a lot, which is the version of Stack Overflow where you make your own site out of Stack Overflow source code basically. And that one had been in kind of a closed beta where you signed up and we basically we were letting people in as fast as we could, but we kind of limited and effective. With the first day of Stack Overflow dev days, it's now kind of an open sign up for the beta and pretty much everybody that signs up gets in within a day or two, I think, if not immediately.

[Jeff Atwood]

Good. Yeah. Stack Exchange, I mean my main. I've been pretty happy with Stack Exchange except for the colors thing that I talked about. And I'm seeing improvement on the colors front, which is good.

[Joel Spolsky]

Yeah, slight improvement.

[Jeff Atwood]

I'm happy.

[Joel Spolsky]

And you know what?

[Jeff Atwood]

It's definitely better.

[Joel Spolsky]

There's a lot of sites showing up on Stack Exchange that are getting a lot of traffic.

[Jeff Atwood]

Good.

[Joel Spolsky]

Yeah. Do you know what the number one is?

[Jeff Atwood]

No.

[Joel Spolsky]

You want to guess?

[Jeff Atwood]

I don't know. I can't. My mind isn't blank. Tell me.

[Joel Spolsky]

Okay. Actually, hold on, let me just see if I can bring this up here.

[Jeff Atwood]

You don't know.

[Joel Spolsky]

The number one is called Epic Advice and it'[e-mail] it's a world of Warcraft Stack Exchange thingamajiggy.

[Jeff Atwood]

Oh, that's hilarious. So that's actually doing the best.

[Joel Spolsky]

It probably is doing as well as Superuser and that's in a matter of two days.

[Jeff Atwood]

Wow.

[Joel Spolsky]

Hold on.

[Jeff Atwood]

That's amazing.

[Joel Spolsky]

Yeah. Okay. Not quite as well as Superuser, but. But it's. But it's rapidly closing in on Superuser. Wow.

[Jeff Atwood]

It's still really good. So that's great.

[Joel Spolsky]

So that's epic. Epicadvice.com about World War I. Never would have thought doing it has colors you may not like, but actually they've done a little.

[Jeff Atwood]

I saw that one. That one does look good. I like that.

[Joel Spolsky]

It's tweaking. They basically wanted to have that geeky amber on black look like the old fashioned monitors that were a little bit cool than the green on black monitors.

[Jeff Atwood]

Right.

[Joel Spolsky]

And it's looking pretty decent color wise. Let's see, what are some of the other huge Stack exchanges. Okay, so our friends Jason Cohen and Dharma Shaw, who are two successful entrepreneurs. You know, I would say by any measure, successful tech startup entrepreneurs have jointly created answers. OnStartups.com See, OnStartups.com is Dharmesh's blog where he blogs about startups and answers on. Onstartups.com is also like a ridiculously successful stack exchange where people are asking questions about startups. Things like, how do you keep yourself motivated scaling up from one web server plus one DB server. How do you start and use a board of advisors?

[Jeff Atwood]

We should be reading this.

[Joel Spolsky]

I know. What are the topics?

[Jeff Atwood]

We have no idea how to do any of that stuff.

[Joel Spolsky]

That's a good one. What are the top 10 startup friendly states in the USA? I'm gonna go with the trouble. The weird thing is all the startups are in Massachusetts and California and those are probably the two least favorable attacks. Perspective, I think.

[Jeff Atwood]

Right? Yeah, that one. That one is good. I was very excited to see that one. I was actually kind of surprised you didn't want to run that one in some capacity. I know you have a million things to do.

[Joel Spolsky]

Yeah, I was going to start one in my spare time, but whatever, as long as there's one there, that's. I'll go and answer questions and ask questions and whatnot. Okay.

[Jeff Atwood]

So, Joel, you're officially backing this one then.

[Joel Spolsky]

I like the. On startups. Com. That's. That's the one I.

[Jeff Atwood]

Okay, cool. That's awesome.

[Joel Spolsky]

That's great. I'm going to be betting on, but, you know, anything can come up. I mean, there's definitely. They're going to be competitive stack exchanges and they're going to be ones that sort of compete against Stack Overflow itself. That was something we were supposed to talk about, right? Yeah, let's talk.

[Jeff Atwood]

It is, it is. But let's come back to it because I do want to mention one other Stack Exchange site that I have been exposed to accidentally, and that is Moms for Mom.

[Joel Spolsky]

Ah, yes, Moms for Mom.

[Jeff Atwood]

Because I was actually getting usability feedback from my wife Betsy based on her usage of Moms for Mom. And that was hilarious to me because the thing that's cool about Stack Exchange and the thing I'm excited to see it sort of push boundaries on is we had a very specific tech audience in mind when we built this thing. I mean, it was essentially programmers and we kind of spider down into IT guys and sysadmins and just general pro users. But these are people who love computers and they're down with lots of geeky things. So what we're finding out with sites particularly like Moms for Mom and I suppose epic advice as well, although those would be gamers, which would kind of be pro users. So I think Moms for Mom is the best example, in my opinion, of just sort of the average user how they're going to be impacted by. Does this crazy thing on my screen even make sense to me? So it was amusing to me to get feedback and I got the little lecture from Betsy about OpenID. She's like, you know, yeah, the OpenID, I don't know. She's like, I don't know if I get that. So I have to like, I'm going to sit down with her and go over it and see what is going wrong. But for most people, yeah, it's basically Google at this point, honestly. I mean, I did some stats on OpenID and it's already. Google came in really late to the OpenID party, but it already on like stack Overflow. It's like 56% of all users use Google. Basically you just click the giant Google button and you're done.

[Joel Spolsky]

As long as you have a Gmail account or a Google account, which most people do.

[Jeff Atwood]

Which most people do, then just click.

[Joel Spolsky]

That button and just stop worrying about it. Yeah, yeah.

[Jeff Atwood]

And Yahoo has done a pretty good job. They've really improved their support to the point that it's pretty easy to use. I mean, the kind of emerging standard is you essentially click the button and it directs you to that provider and then you get redirected back. There's no typing in of URLs anymore, not really. You just click the giant button and that seems to be usable enough for most people. You know, click the giant button, type in your email address, your email name and password, and then you're done.

[Joel Spolsky]

So I think there's a couple of things we could do to make even openid a little bit easier is explain it to people a little bit more gently. But I'm pretty sure that in order to support the customers of Stack Exchange, we're going to have to add a native logon facility so that you can make an account on the site if the site creator wants you to be able to do that.

[Jeff Atwood]

Well, I support that because I thought for Stack Exchange that made sense. I mean, we had a specific audience in mind, so I have no problem with that going on in parallel. I think that's actually kind of necessary for what you guys are building. So let's come back to your original question, which was we had talked a little bit about what if Stack Overflow decided to start a site like Moms for Moms. We would never do this, of course, because that's not even remotely our audience. But we had talked about us competing with Stack Exchange sites, which we don't think is going to be a huge problem. But then what we hadn't considered was the inverse of that, which is what about Stack Exchange sites that kind of compete with us in a limited way? Like there's I think one called SQL Team.

[Joel Spolsky]

Right. There's actually a couple of SQL ones right now. There's a SQL SQL. It looks like there's a SQL Server Central, Stack Exchange, which I would guess that Redgate Software, which owns SQL Server Central, sign up for one. Yeah. And those are specifically about SQL Server questions. Well, I feel like we're now, we're now. The truth is we are now competing against those. If somebody has a question to ask, they could go to either of those sites. SQL Server. It's not fair for me, not to mention the actual URL. I think it's just SQLServiceCentral. Stack exchange.com?

[Jeff Atwood]

Well, don't we have like a list I'm going to post. There's a list somebody put together on Meta. Like a list of all the Stack. There was like 80 of them.

[Joel Spolsky]

There are quite a few.

[Jeff Atwood]

There's so many.

[Joel Spolsky]

We just haven't had time to actually do a directory. And plus our assumption was it's not like you're like, I would like, oh, look at that. Stackexchangesites.com who the heck is this?

[Jeff Atwood]

Right? So there's a lot of these sites.

[Joel Spolsky]

There's a lot of Stack Exchange directory sites. Oh boy.

[Jeff Atwood]

This is exciting too because one of my goals in doing this, and this is why we had the whole open source conversation that was not so pleasant, was I just want to get it out there. I want to get rid of PHPVP as the standard thing. I hate PHPV so much that I just want to destroy it to the extent that we're seeing tons of people adopt Stack Exchange. That's awesome. I get really excited about that.

[Joel Spolsky]

Here's one of the SQL server ones. It's called Ask SQL Team. It's asking SQLTeam.com that is one of, I think two SQL server oriented websites. Now the bottom line is that we did not create Stack Exchange in order that there be exactly one website on every topic. And then we could somehow make money by charging people or something to be the website on the topic. What we've discovered what we discovered right away with Moms for Mom with the questions about startups. What was it? Answers on startups.com on startups, yes. And obviously with Epic Advice about World of Warcraft is that if you have the ability to somehow get a community to come into a site, your site's going to be successful. That's the differentiating factor. It's not whether you come up with the interesting topic for people to ask questions and answer about. It's do you somehow have the wherewithal to get a group of people to all pile into a website and start answering questions? Because what will make people come back to a Stack Exchange is that when they type questions, they get answers to them. And so there's two ways to provide that. One is you can either answer the questions yourself. Staying up late at night, which we do ourselves on the Fogbugs Stack Exchange, a tiny website which we use for tech support purposes. You know, we actually have tech support people that check that and our programmers check that. And so you do get answers there, or you bring in a large enough community like at Epic Advice, where there are people that just love answering questions about World of Warcraft and have a lot of fun doing that. So you got to somehow get the community there. It's not about having the topic. And There may be 300 SQL server sites, but you know, one of them eventually is going to get the mass of people. And I personally hope that that stuff just stays in Stack Overflow. But then there may be stuff that's just so specific about SQL Server that it never really shows up in Stack Overflow. Or there may be people that see themselves more as SQL Server admins than programmers in general, and maybe they'll have more fun on the SQL Server sites.

[Jeff Atwood]

You know what's funny about Moms for Mom is that John Skeet was actually on the top user page, which prompted people to wonder. It's like John Skeet's everywhere. It's like he's going to be dominating every Stack Exchange site.

[Joel Spolsky]

Yeah, but he can't. He's never.

[Jeff Atwood]

1, 2, 3.

[Joel Spolsky]

He can't keep up. He's not going to keep up. Let's look at. Is he even on Epic advice? Does he. Does he even play World of Warcraft? John Skeet? I don't think so.

[Jeff Atwood]

Well, you know, the funny thing about the World of Warcraft metaphor is that there are some class of alts. Like people who play the game will have other characters that they start just for the purpose of having other characters. And I Kind of use some of these sites, like say the sequel sites as like alts. Like you could actually have, you know, you could experiment and just have another character that you run in the SQL domain and you just check that because you're so into it. Right. Like there's not enough on Stack Overflow because how many questions are about, you know, hardcore procs or. I mean there's not that many. There's a lot, but it's not like every day. Whereas you could go on the SQL site, the SQL Stack Exchange site, and then just like every day you'll get five or six new really good hardcore SQL questions.

[Joel Spolsky]

Yeah, yeah, that may be the case that. Well, it all depends on whether they get the audience. Now in the case of SQL Server Central, I know they definitely have an audience, you know, because they're the largest. If they're definitely one of the largest, if not the largest SQL Server specific discussion forums on the web and they definitely know that they want to move to a Stack Exchange like model.

[Jeff Atwood]

Right.

[Joel Spolsky]

So, yeah, so our answer is may a thousand flowers bloom. But it's really not about like our software is great, but it's not going to magically create a community if you can't get some bodies on the ground in there answering questions.

[Jeff Atwood]

The money quote there is Stack Exchange does not magically create community. That's still hard work. Right, right. I mean we're still, and let me use us as a specific example, we're still growing. Super user and server fault. I mean those are growing, but slowly growing.

[Joel Spolsky]

Hey, did you see on whatever Quantcast, we're like 850 or something. Largest site.

[Jeff Atwood]

Yeah, but most of that is Stack Overflow though.

[Joel Spolsky]

Yeah.

[Jeff Atwood]

Stack Overflow is still growing in a surprising clip, which, I mean it's already quite large.

[Joel Spolsky]

Yeah. Here we are on Stack Overflow Network. According to Quantcast, as of this very second, 958th largest site.

[Jeff Atwood]

Yeah, yeah, I think we cracked the.

[Joel Spolsky]

Top thousand websites, which is insane.

[Jeff Atwood]

Yeah, that's cool. That's very, very good. I mean certain things that we're doing. And that implies another weirdness with the models. Like if you don't get to a certain scale with your Stack Exchange site, does it even work?

[Joel Spolsky]

Right?

[Jeff Atwood]

You know, because I mean we're building for scale. I mean we're optimizing for public wide access stuff.

[Joel Spolsky]

The one way you can avoid that is, I mean you can make a very small site and have it work. A good example would be fogbugs.stackexchange.com just about fogbugs and that's basically going to replace what was a discussion group that was very low traffic because there aren't that many people that have Fogbug specific questions that they can't get an answer for or they just call us or something. But it was there, and especially during the beta it was reasonably active. But I mean, we're talking 10, 20 people going in there every day. Not like, you know, not like a large site. But fogbugs. StackExchange.com is a support forum for Fogbugs works really well because first of all, we have tech support people and our programmers are peeking in there all the time making sure people get answers. And secondly, we're posting our own questions and answers in there. We're using it as a knowledge base. So if we suddenly have we noticed that we've gotten three or four emails asking us how to do X, Y or Z, we'll post the question ourselves and the answer on our Stack Exchange, just so it's there for Google to find and it can be edited and it's sort of better than a wiki in so many ways.

[Jeff Atwood]

Okay, so I have two related questions. So one, what would be the. Okay, let's start with the first question. Does, does this work as a support form in your opinion? Like, would you buy a Stack Exchange site for support for your product?

[Joel Spolsky]

Oh, heck yes. I'm going to shut down all the other support forms that we're, that we're using, which are all actually Fogbogs discussion groups, and we're going to shut those down and use Stack Exchange.

[Jeff Atwood]

I was honestly surprised because for a long time we used User Voice, which, you know, it's built as a support E type forum.

[Joel Spolsky]

Yeah, User Voice has a bit of an emphasis not so much on providing support so much as getting feedback on future versions, which is a little bit different. It's more like not, how do I solve this emergency problem I'm having right now? It's more like tell us what you want from version 2.0.

[Jeff Atwood]

I was actually surprised, I mean, because I went in very skeptical of Stack Overflow working as a discussion feedback site. But I have been shocked how well it actually works.

[Joel Spolsky]

You did add one feature for that, right? You added the special tag.

[Jeff Atwood]

Yeah, I added two things. One is moderator specific tags so that moderators can have certain tags that they own and also required tags. So every time you start a question you must have one of these tags. The support is still kind of janky for that, honestly, because that was a meta specific thing. So I didn't do it right, probably should go back and do it right. Because if other Stack Exchange sites want to do that, the support for that needs to be more robust. So the second question I would have for you is what's the minimum number of knowledgeable people you'd have to have for the site to actually be worthwhile? Like, even assuming some amount of just random anonymous Internet traffic, how many users do you think it would take to make a site worthwhile? They were all really good users.

[Joel Spolsky]

One really knowledgeable one.

[Jeff Atwood]

One, yeah. Really?

[Joel Spolsky]

Yeah. I mean, for a while, the Fogs knowledge Exchange, it was just rich because it was his job to go in there and make sure that people were getting answers. And so people, you know, he'd go in there at least once a day or once or twice a day. And that meant that you got an answer in 24 hours, and that's about the best you could, you know. And then you're going to have a site where people can ask questions and get an answer within 24 hours, unless their question is already there, in which case they get an answer immediately. And that's already more valuable than using phpb, where you got to find that search box, which is in a different place every time. And you got to search on Google and you got to make an account if you want to search, and all kinds of wacky things that some of those sites make you do.

[Jeff Atwood]

Well, let me give you an example.

[Joel Spolsky]

That's a functional site. I mean, it's not a tremendous site, but it's functional. And actually, if you look at, like if you look at the homepage of the FogBook site right now. Let's see. One thing I do to tell how active a site is, is you can sort of look at what questions are getting their most views, or you can look down at the bottom and see what's at the bottom of the first page. You know, how many days in the past is that? So we've got a. On the Fog site, we've got two weeks worth of questions on the active page, which means it takes two weeks to fill up the active page. That's about the rate at which that site is filling up. That's not a very active site. And yet I think if you ask people that are using it whether they're getting value out of it, I think they would say, yeah, I'm getting my questions answered quickly.

[Jeff Atwood]

Well, and just to give you an example of why I like this, and even between you and I, Joel, this careers thing, Joel is really driving a lot of the features Because I feel like he has a lot more knowledge about it than we do. So we have a lot of internal communication about what's going on in careers. And I really sometimes have even been urged encouraging Joel to don't email me the answer, like post it on Meta so I can just read it there like everybody else. Because the more we can get out public, really the better it is. I mean, basically it's just escaping that email silo as much as you can. I mean, there are obviously things that are still private that we have to email each other about, just stupid things that nobody would care about. But I totally agree. I mean, if you can put your stuff, just get away from that email silo of emailing customers and get to public information other people can find, it's a much better long term investment in whatever it is your company is doing or you're doing.

[Joel Spolsky]

Exactly. Yeah. Yeah, yep.

[Jeff Atwood]

So let's, let's transition to the careers thing. That was, yeah, yeah, that was, that was the big announcement for the Stack Overflow team because we've been working on that for, you know, the last six to eight weeks. Actually quite a bit, quite a bit longer than that. And I was, I was, I was really excited, actually, finally get it out there so people could look at it, touch it and play with it. Now, right off the bat, I think we, I don't want to call it a misstep, but I think there was a little bit of a miscommunication, a misunderstanding about how we should have been doing this with regards to public versus private. Yeah, we've already made, we're going to make an adjustment on that. So maybe, Joel, you could talk about that.

[Joel Spolsky]

Well, so the original idea. Well, let's just talk about the basic idea. The basic idea is that we got these great developers on Stack Overflow. I mean, they really are. And when I look at people posting on Stack Overflow and I think about how I, as the manager at Fog Creek, need to hire great developers, I think, boy, I wish I knew which of these people were on the job market and if they would even remotely consider, you know, a job. And I just sort of wish I had a way to push a button and get a list of people on Stack Overflow who are, you know, let's say, knowledgeable about, you know, a few technologies that I consider to be valuable, maybe Lisp and C and Visual Basic on the market, willing to consider a job in New York City. So that's exactly what we wanted to make. We wanted to make a way that employers could Basically search for people who were looking for jobs on Stack Overflow and kind of see them on Stack Overflow. So instead of just looking at their resume, I mean, resumes really suck as a way to communicate or prove that you're a good developer. If you've ever had to screen resumes, if you've ever had a pile of resumes to go through with programmers, you just sit there saying, I can't even begin to guess who's good here. Right. There might have been two people who have been working at Microsoft for 10 years, and they both went to Ivy League schools and they both got good grades. And, you know, one of them is just 100 times better programmer than the other one. And you can't. You're not going to know by looking at their resume.

[Jeff Atwood]

And you still look at resumes? Pretty much. Yeah.

[Joel Spolsky]

Well, we have to. Yeah, and I. And I do all the time. And we have a scoring system that we use to try to isolate the ones that are worth interviewing.

[Jeff Atwood]

Oh, tell me what it is so I can optimize for it.

[Joel Spolsky]

Yeah, you sure can. It's. It's. Go to Amazon.com search for Joel Spolsky, and look for my latest book, Smart and Get Things Done. It's chapter three.

[Jeff Atwood]

You tricked me into buying. You tricked me into buying your book. That's so uncool.

[Joel Spolsky]

Okay, well, I'll tell you what it is. We look for evidence. Some of this stuff, it's hard to find on resumes, and sometimes we don't see evidence of it on the resume, even if it might exist. But we look for things like passion for computers. So I want to see that you play with computers in your spare time that you were into. You were programming TRS 80s when you were a little kid. I want to see that you just love computers because that's often a good sign. And there's a bunch of other stuff like that that we'll look at on resumes. We'll look at how good you are communicating. How clear were you at communicating in a cover letter? We'll look and see if you're writing a cover letter that's specific to Fog Creek, as opposed to a cover letter that you've copied out of a book that doesn't say anything about Fog Creek. And the reason for that is that the one copied out of a. All else being equal, it's more likely that you're applying to 100 jobs. And if you're applying to 100 jobs, it's less likely that you'll work for us. If we hire you because you're applying to 100 jobs. And anyway, why are you applying to 100 jobs? You don't have any confidence in your abilities. So maybe that's because you're a loser. So anyway, so that's. It's not, it's not guaranteed, right? Like there's no proof that just because you sent me a generic cover letter that you're not going to be a great developer. But I got nothing else to go on. Was that, was it a generic cover letter or was it a specific cover letter saying, you know, I love Joel's software, I want to move to New York because, you know, that's the awesomest city in the world, I want to work for you. And that second one, we're going to interview you first, you know, all else being equal, just because there's a higher chance that it's going to work out. But that's a, that's a very, very weak, you know, indicator. It doesn't prove anything. There might be somebody that just, you know, couldn't be bothered to write a custom letter that day or whatever. And so, so the old system of resumes, just trying to browse through resumes really stinks. But if you have a list of people on Stack Overflow that you know are looking for jobs in New York City and you can go and look at their profiles on Stack Overflow, see the questions that they're asking, the questions that they're answering, see if they're good at communicating, see how many points they have, see if they're making an effort to help the community, see if they can explain things, see if they appear to have in depth knowledge, see if the code that they post is clean and elegant, smart and intelligent, you can tell a lot more about a person than you can from a traditional resume. And so it seems like a great system for somebody who's looking to hire the great programmers. And it's a great system for the programmers too, because you want to be where the good employers know to look. So that's what Careers is all about. It's all about matching the great programmers with the great employers. That was really high level.

[Jeff Atwood]

That's okay. I mean, that's fine.

[Joel Spolsky]

Maybe you should explain that, given that now Jeff will tell you what it actually does.

[Jeff Atwood]

So what it does is you can go to careers.StackOverflow.com, it's currently in beta, so there's a very low, low introductory rate and you can actually create what we're calling a cv. Now, CV is the more European Term for resume. But there's a reason we didn't use the word resume.

[Joel Spolsky]

Yeah.

[Jeff Atwood]

Because one, I feel like resume is kind of a word that has negative connotations to me now.

[Joel Spolsky]

Yeah.

[Jeff Atwood]

And I think CV just, just, it's more open ended as well. That's probably what attracted me to, is that CV is not about I want to necessarily go out and get a job tomorrow because I'm desperate. Right. CV is more like I've done some really cool stuff and I would like to share that cool stuff with you or anyone else who's interested. Okay. So to me that's, that's why we chose that word. Because people have asked, they're like, oh, it's so European.

[Joel Spolsky]

Yeah.

[Jeff Atwood]

But I just think it's more open ended and it makes it more friendly. You don't feel like, okay, I'm immediately on the job hunt the minute I put a resume out. Because that's not really the intent. And that's one of the major adjustments we're making is that since originally the view was that, okay, the CVs will all be private, but now we're going to make it so that you can put a public CV up at no charge. There's no charge or anything for this and you can actually share that with the world and you can actually have what's called a vanity URL. So it'll be careers.StackOverflow.com whatever. So there's a certain amount of land rush involved. So you kind of, if you're interested in this. We haven't done the public thing, but we're thinking by the end of the, by the 23rd, it will definitely be available.

[Joel Spolsky]

That's pretty soon.

[Jeff Atwood]

Yeah, it's still pretty soon. So the intent is that as a programmer, if you want to, this is completely optional. So let me be upfront about this. Like, if you like Stack Overflow and you don't care about any of this career stuff we're doing, you don't have to even look at it. It's not going to bother you. It's not going to be surfaced on the site in any meaningful way. You can just ignore it and continue as usual with Stack Overflow. But if you're into a listing your cv, like sharing some of the cool stuff you've done with people in a more detailed way than just the free text box that we have on your Stack Overflow profile or B, you actually are sort of in the thinking about getting a different job or wanting a job as a student. Then go tocareers stackoverflow.com you know, fill out your CV and then you can do two things with it. You can make it public, optionally, so that it's visible to Google and the world through your vanity URL that you can set up yourself, that that's free, cost nothing. And then B, probably the more interesting thing for people who are actively looking for a job, you can file it, which makes it searchable by employers. So when you do this, you're flipping a bit that says, hey, employers, I am actively looking for a job and I'm interested in you proposing me work that would be interesting to me. And the thing that I think is interesting here and sort of the angle we're taking is that smart companies should be pursuing good programmers and not really the other way around.

[Joel Spolsky]

Absolutely.

[Jeff Atwood]

Okay, so we're kind of flipping the equation a little bit and saying if you're a good programmer and you have a great cv, a nice track record of stuff that you've done, you have a track record on stack overflow of being helpful to your peers and actually writing in a clear and concise way, and some really cool posts on stack overflow that that should be enough for employers to look at you and say, this person would be a great fit for my company. And we're trying to attract the types of companies that appreciate passionate programmers and aren't just like resume hunters, Right? They're not just keyword scanners. They're actually looking for people who give a crap about our craft, you know, because they realize that, hey, our company is built on like your company. Joel, for example, obviously is an obvious example where, you know, it's all about having the best programmers. And what's your famous, what's your formula? The best programmers, the best software, the best environment. Yeah, you're really interested in having the best programmers because you realize it's a strategic advantage. And for a lot of companies, honestly, I don't think they care. They know that if they don't have the best programmers, it probably doesn't really matter.

[Joel Spolsky]

It does. They just don't think it does.

[Jeff Atwood]

Well, yeah, they don't think it does.

[Joel Spolsky]

How much it matters to get the best program, how much better off life would be.

[Jeff Atwood]

And we're not interested in those companies. Like, we're not really. We don't care if those types of companies don't use the site because we don't really care want them to. But for the type that do appreciate, you know, hitting the high notes, that whole essay, they're going to come to stack overflow careers and actually clue in and actually drill down into your cv, drill down into your stack overflow profile and sort of get a sense of what you're about and hopefully they'll be approaching you. It's almost like a dating site is the way I like to look at it. You're trying to make a love connection. Right. You're passionate about programming. This company is passionate about passionate programmers. Right. So you're going to make a love connection, they're going to approach you via email and hopefully there'll be some interest at some point. And it just really focuses and narrows down the list from like Career and Creole Builder and Dice and Monster where it's just kind of like a sea of sort of sameness and just random people putting random resumes out into the world and just. It's just a statistics game. Right?

[Joel Spolsky]

Right.

[Jeff Atwood]

We're trying to tweak the odds, like cheat the house, like incredibly so that you're already dealing with passionate programmers who are pretty good and you're attracting the type of companies who understand what this stuff is all about. So your hit rate is hopefully much, much higher. Yeah, that's really the attraction.

[Joel Spolsky]

Somehow when I think about how the job market works right now for great programmers, what happens is that in a couple of big places, like let's say that you're in New York, people tend to know who the big employers are who are good places to work. So in New York, people know there's about 1500 Google people. Google has great lunch. They don't have the greatest working conditions. It's really just a big room full of desks. But they have pretty good, pretty plush circumstances as far as programmers go. And it's certainly a great place to be a developer and everybody knows that. So the question is, what if you're in Austin like I am right now, do you know what those companies are? Or what if it's a smaller company of like five, six people that are just passionate, you've never heard of them and they're doing great with their startup and they want to hire a couple more people and it would be a fantastic place to work but nobody's heard of them yet. And so similarly, what if you're that great programmer in sort of a second, second tier market or somebody or you're not that well known yet. This is kind of the way of matching up the great people that are. It's not like, you know, Brian Kernighan going and getting a job at Google as he did. It's not about, you know, like the number one main Programmer getting a job at the number one. It's all about sort of the second tier, the. The not so well known names of great programmers getting jobs at the not so well known great companies.

[Jeff Atwood]

Sure. Because there are a lot of companies out there, like small. I mean, because Fog Creek is a small company. I mean, Stack Overflow is a small company.

[Joel Spolsky]

Sure.

[Jeff Atwood]

You know, you may not know all the companies out there.

[Joel Spolsky]

Yeah. That you could be working on Stack Overflow right now if you lived in New York and applied for a job at Falk Creek.

[Jeff Atwood]

Now, one question, Joel, that came up that I specifically want to address was about recruiters. There's some subtlety to the recruiter story because my vision of this was that we want the companies whenever possible, we want to cut up the middleman and connect programmers with the companies. Right.

[Joel Spolsky]

And that's what programmers want to. Because the good programmers know that they have a choice of where to work. And that means that they want to hear about this company they're gonna be working at. And they don't even know if they want to respond or interview until they've gone to the company. They've heard the name of the company, they've gone to the company's website, they've looked at what they make and what they sell, and, you know, they've kind of got an idea that if this is a company that they would be willing to spend a few years at, you know, whether they can get behind the mission. And so a few years ago, when we did the. When we first launched the job board, the Joel on software job board, jobs.jolonsoftware.com, which is now also the Stack Overflow job board, jobs.StackOverflow.com, i asked around and I asked what people wanted, and they said the number one thing is we don't want to see any stinking recruiters on there. And the way you can tell the recruiters on these stupid job boards is they never tell you the name of the company where you'd be working. And the reason they never tell you the name of the company where you'd be working is because they don't want anybody to be like, oh, craft is hiring. Craft sounds good. And go to the CRAFT website and go to the job section and apply for a job. Because these recruiters are working on a commission basis and they get paid if and only if the job comes. The candidate is found by them. They're basically working on a, what they call a contingency basis. And so if you've ever gone onto Monster or Hot Jobs or looked at the classified ads and you've noticed that 90% of the ads that you see there have some random sketchy number you can call. They don't tell you what the company is or what they make. They just list a whole bunch of things that you have to do. You have to have five years of Ajax experience and 23 years of Ruby on Rails experience and so on and so forth. And. And a lot of these classified job listing type sites are just full of those. It's 100% that, certainly Craigslist. What you're looking at is these contingency recruiters just trying to get as many resumes as they can and they're going to throw every single one of them at these companies in hopes that one of them sticks and one of them actually gets hired. So this is not our market, this is not what we're interested in. We're not interested in this kind of thing happening on careers.Stack.StackOverflow.com there's another kind of recruiter out there, by the way, that also goes by the name recruiter but does a very different thing. And that's called. Oh, remind me what that's called. I just wrote about it. Contingency based Meta stackoverflow. We're just talking about this today. The contingency recruiter and the.

[Jeff Atwood]

Retained search recruiter.

[Joel Spolsky]

Search recruiter. Thank you. I just found it. Wow. Okay. So the retained search recruiter is somebody where you've got a company, let's say that they make guitars and they need a great programmer, but they're a guitar company. They know nothing about programmers and they don't know how to hire a great programmer. So they will hire a retained search company and they will say, I'll give you 30,000 bucks, get me a programmer and I will work. And the retained search recruiter always works on an exclusive basis. So they're the only person trying to fill this job. There is nobody back at the company trying to fill this job and there are no other recruiters working at other companies trying to fill this job. And because of this exclusivity, they can feel free to say, Joe Guitars is looking for a great programmer and they can use the name of the employer because they are working solely on behalf of that employer. And for all intents and purposes, that's almost like they are an employee, you know, temporarily. Of the guitar company. Of the employer. Now these guys are fine people to work with. When it's on a retained search like that. They have a very ethic they need to get this job filled by somebody who's qualified in order to maintain their reputation in the business. They're not going to get any more work if they send you garbage. And so they will really only send over good resumes or people that they have personally vetted. They will generally do all the interviewing themselves to make sure that the person is qualified before they send them over to the company. So these people will tend to show up using careers stackoverflow.com but the first kind of recruiter, the contingency recruiter, the kinds that are basically just resume spam factories where it's all about sourcing as many resumes as they can find and sending as many of them on as possible in hopes that one of them gets hired and they get to keep their contingency commission. Those people are not going to want to use Stack Overflow, those contingency recruiters, because they won't want to reveal the name of the company and we're going to require them to reveal the name of the company where the, where the candidate would be working.

[Jeff Atwood]

See now this is a good example, Joel, of why you are the pm, I guess you would say, of this thing we're building because I didn't know any of that at all.

[Joel Spolsky]

Yeah, well, I had to.

[Jeff Atwood]

Joel is really helping us drive this forward because Joel is much more intimately involved in hiring on a day to day basis. I mean, I don't really hire for Stack over.

[Joel Spolsky]

Right.

[Jeff Atwood]

I sort of handpicked people that I'd work with. Which is totally cheating.

[Joel Spolsky]

Yeah, it's not, but Joel does.

[Jeff Atwood]

It's something you do every day. You've written books about it.

[Joel Spolsky]

Yeah, so you're. And by the way, we never use any kind of recruiters. We just don't want to.

[Jeff Atwood]

Well, I would be shocked if you had actually. So if you had to summarize. So one question that came up on Meta was like, okay, so why should I use the CVs functionality to look for work? Like what's, what's the competitive advantage? Like, if you would summarize, like why would someone pay us to list their cv? Yeah, well actually, let's just come back. So why are we actually charging the programmers?

[Joel Spolsky]

Oh yeah, why do we charge the programmers for this? And believe me, I'm not charging them because there is some controversy around this.

[Jeff Atwood]

Obviously it's because Joel hates programmers. You heard it here first, that's why.

[Joel Spolsky]

No, the reason is because in order to attract the great recruiters, we have to promise them something. What? We have to promise them Is that when I give you a list of resumes, those are real candidates. They're not just kicking the tires to see, you know, how much they would be worth in the open market. They're not just kind of entertaining offers, they're actually kind of serious. They actually think that they can get a job, which is important because you don't want people kind of reaching for a job that they can't get there. There will be people who are like, gosh, programming sounds like a great job. And they'll submit. I mean, I get resumes. Every time we put something, even on our website, we get resumes from people that have never programmed money out of an ATM machine. Well, that is sort of beyond their ability to type those digits into an atm. So we are, I mean literally resumes from people that worked at Dunkin Donuts and sometimes very serious cover letters saying, you know, I've been working as a salesperson in real estate, I'd really like to get into computers. I think I could be a great programmer. I saw that there are all these 21 day days books, I don't really say that, but utterly and completely impossible for them to ever get this job. And they're just applying because what does it cost them to email a resume? You know, it's free, it's expanding, it's basically spam. If you, if they get a 1%, if they have a 1% chance of that great hundred thousand dollar programming job, well that's better than what they're doing now. Especially because they can apply for a thousand of those jobs just by spamming their resume.

[Jeff Atwood]

Yeah, yeah. The whole one thing I immediately got, and I was hoping people would get, is that one of the reasons we're charging money is honestly just to have a barrier to say you're either serious or you're not serious. Because gosh, I mean, I just, I remember my personal experience going to back in the day of Dice and Career Builder and it's just, you feel like you're in a wasteland. It's like embarrassing to even be in this bad neighborhood of like just random people applying for random jobs they're not even qualified for. You know, there's so much noise in the system that we want to have a system where hopefully there's a lot less noise. And this is, I mean that's a key aspect of this. Right?

[Joel Spolsky]

Yeah.

[Jeff Atwood]

You're serious.

[Joel Spolsky]

So the great thing is, I think on stack overflow, I don't know how this is going to pan out, but my estimate is there's Going to be like a thousand resumes on there. That's not going to be a big gigantic. These are all the programmers in the world. These are a thousand pretty good people. You can find out if they're good by looking at their profile. If you're a recruiter. Not a recruiter, sorry, if you're a hiring manager and you want to hire one person in Austin and you go and you look and among these thousand people, there are 12 in Austin, that doesn't seem like a lot, but six of them are really good. All of them are looking for jobs. And you're going to have your choice of those six great people. And that's good enough to fill one position. You know, we're not going to be, we're not going to be staffing Microsoft and Google and Oracle off the staging Exchange. Right. But, but we are gonna be a great place to find one person. And the question is, would you rather pay a little bit of money and be one of the six great people in Austin looking for a job or have it be free and have 200 unqualified people in there with the six good people? Eventually that's just gonna, the good employers are just gonna give up on, on us if we have a bunch of junk resumes in there.

[Jeff Atwood]

And just to be clear too, this isn't necessarily. I mean, this is really about improving the lives of programmers, which to me is the core goal of stack overflow. Right. In some small way, searching for information just became a little bit easier because our system is just a tiny bit better than the other crappy systems that are out there.

[Joel Spolsky]

We're helping people be awesome programmers.

[Jeff Atwood]

Well, exactly. And this is really the same goal of careers is this. This doesn't come out of we need money. This comes out of the current system kind of sucks.

[Joel Spolsky]

Yeah.

[Jeff Atwood]

And as programmers, we build things and one of the things we're going to build is, hey, we think we can build a slightly better mousetrap. And you know, if it helps one programmer get out of, you know, corporate drudgery job into like a cool programming job where they actually value, you know, code, then I think that's totally worth it. I mean, that to me really is the goal. That's. It's still part of our core mission, which is in a tiny way improving the lives of programmers.

[Joel Spolsky]

Yeah.

[Jeff Atwood]

So I just want to be clear about that because anytime we introduce money, you know, things get weird. And I mean, that was honestly one of my reservations with this system was.

[Joel Spolsky]

Well, also, don't get me wrong, the employers are Paying most of the cost of operating the system.

[Jeff Atwood]

Oh yeah, we should mention that because I mean there's no information on the employer side.

[Joel Spolsky]

Yeah, we haven't really said that. But they're going to be paying, I think probably the minimum for an employer would be $500 to get access to a search.

[Jeff Atwood]

And that's just like one day or whatever.

[Joel Spolsky]

Look at it, it's like one search. Yeah. So believe me, the employers are going to pay and it's going to be worth it to them because I don't know if you, if most people listening know this, but if you hire a recruiter or just basically a budget for a typical company to fill a typical job is 1/2 to 1 year salary. So if the job is $100,000 job, which is a pretty typical programmer salary, say in Silicon Valley, for an experienced programmer, 100,000 is about average. So they would have a budget of 50 to 75,000 to fill that job.

[Jeff Atwood]

Holy crap. So their budget is half of a year's salary.

[Joel Spolsky]

Wow. That's why. Yep, that's why they're paying 350 for that classified add on stack overflow. And they're probably running 10 on other sites that are similar, charge about the same amount. If they use the contingency recruiters that's going to those, they're going to take at least 3 months salary, if not 6 months for a successfully placed, successfully filled position. That's basically just what they estimated the cost of filling that position. And then don't forget that when they hire somebody new, if there's some kind of turnover, that person shows up and needs training. Just it takes time for them to come up to speed. So you're paying their salary while they're getting their computer set up. And when all that is added together, filling a position is like a 1 salary year pretty much in lost money. So there's actually kind of a lot of money there on the hiring. The employer who doing the hiring, they expect to spend money to get the good candidates to find the good candidates.

[Jeff Atwood]

Right. So yeah, just because we're encouraging the programmers doesn't mean we're not charging the employers.

[Joel Spolsky]

Yeah, we're charging the employers a lot more. The programmers is just the minimum that we think it would take to make them look respectable. And of course you and I have always had a philosophy of if somebody doesn't feel like they got value out of what we charge them for, we just return their money.

[Jeff Atwood]

And that's true. That is one thing we're adopting wholeheartedly and without Reservation on the stack overflow side is when we charge you money for any reason that it, well, not forever, but in the first 90 days if you decide, you know what, I don't like this. No questions asked, instant refund, no problem whatsoever. It's like, send us a one line email and it's done. So yeah, there is that money back guarantee aspect to it. I always appreciated that. And I think you had put some stats together too of like. And actually I remember reading an article where somebody said they had tried to sell something online and people just weren't buying, but the minute they put in like a guarantee that all of a sudden it removed whatever barrier there was. So I think it just gives people confidence that the risk has just decreased a lot.

[Joel Spolsky]

From the Fog Creek perspective, with the software that we sell, we get about 2% of 1 to 2%, let's say 2% of the people take their money back.

[Jeff Atwood]

And wouldn't you much rather have like a happy person not paying you than an angry person who has paid you?

[Joel Spolsky]

Oh yeah.

[Jeff Atwood]

Doesn't that seem like just a horrible, horrible recipe for disaster? Like if there's something about our service that's pissing you off and it hasn't satisfied you, can you imagine if you kept their money?

[Joel Spolsky]

The trouble is, entrepreneurs and business people, a lot of times they start thinking about what's fair. And I just remember seeing, I don't remember who this was, so I'm not going to name names, but I just remember somebody in the old businesses software discussion forums who got into a fight with a customer over $5. And he's like, listen, first of all, you waited, blah, blah, blah, and then you used it. And then after you asked for the refund, you used it again. And then, you know, you don't get your $5 back. And he got into a long argument with a customer and then came onto business software discussion forum and said, okay guys, who's right here? I'm just confused because my customer is totally wrong. And it was true. The customer was not justified in asking for their $5 back. They just, they just weren't. It was entirely the customer's fault. And I said, are you kidding? You're having a fight with a customer over $5? Yes, it's $5. And he's like, gee, you're right. I'd be like, I am ever so sorry. It is my fault. Have $10 back and here's free service for the next six months. Right? Well, let me at least maybe they'll go say good things about you.

[Jeff Atwood]

Well, exactly. Let me give you an example. Like Flickr. So there was a time when I was experimenting with image hosts for my blog, trying to figure out, like, how to deal with bandwidth and stuff and what were my options for hosting images on a blog. And I looked at Flickr and my cursory look at Flickr said, okay, this could work. This is plausible. So I paid them for two years or something. Whatever their default signup was, it was like 50 dol. So after doing that, I finally read some little clause in their terms of service says, like, you can't use this for commercial or professional things. And you also had to link back on every image technically, which not everybody does, by the way. So I realized, okay, I'm going to violate the terms of service if I do this. So I wanted to cancel. And, like, I had just paid for it, like, hours ago. And Flickr has a no refunds ever policy.

[Joel Spolsky]

Really?

[Jeff Atwood]

I have never been angrier at an online service. And for, I swear, probably still. Actually, I'm doing it now. I will tell anyone who will listen how much I hate Flickr now for as long as they will suck it up and listen, because they were totally jerks to me. That's just. I know.

[Joel Spolsky]

How much did it cost them to give you some free bandwidth? Zero.

[Jeff Atwood]

It made me so angry.

[Joel Spolsky]

That policy makes people angry and gains them. Zero.

[Jeff Atwood]

Yes, exactly. That's my point. So if you go the opposite direction, say, hey, you know what? No refunds ever.

[Joel Spolsky]

Yeah, people.

[Jeff Atwood]

You're going to turn people against you. They will actively go into the world and tell people how much you suck. Which, I mean, what amount of money do you want to put on that?

[Joel Spolsky]

Right? Yeah.

[Jeff Atwood]

So, yeah, anyway, enough about that, probably. Okay, so we're necessarily trying to take over the world with careers. As you pointed out. It's not meant to be an enormous.

[Joel Spolsky]

No, we want to take over, you know. You know what we want to do? We kind of want to make the high end. Like, right now, if you go to the high end of, like, recruiting CEOs for Fortune 500 companies, there's like a very plush, elite, expensive. Like, we're talking about people that can run a Fortune 500 company. There's kind of this really, really. They're these big multinational recruiters, headhunters that they know exactly to do these complicated searches. They provide a very, very high level of service for the top Fortune 500 CEOs. So we want to do some same thing for the top 500 programmers in the world, Right?

[Jeff Atwood]

Exactly. So it's a Bit of a niche service. I mean, I don't know how big it's going to get. We reserve the right to fail. This might be a totally failed experime, and we think so based on our current signup levels. It's actually doing pretty good. But it is absolutely our goal to make these love connections and we're going to do everything we can to make that happen for you. So if you do sign up and you do pay us the $29 a year, rest assured we are working tirelessly on your behalf to make this happen. Now, one final thing, Joel, that came up that I do want to cover is privacy. So if someone actually lists.

[Joel Spolsky]

That was one of the areas where we adjusted a little bit midstream.

[Jeff Atwood]

Yeah. We didn't actually have a good plan for this. So let's say, you know, you're actually working in a company, but you, at the same time that you're working in that company, you want to also sort of be on the hunt. Well, the catch 22 is there. Like, what if your employer actually signs up for Stack Overflow Careers and then sees that, oh, Joe is looking for a job? So one thing we're going to do, like, by default, is every employer that you list that you worked for. Now, bear in mind, this is string matching. So string matching is what it is. I don't know if a better way to do this, but if any of those employers sign up for Stack Overflow Careers, employer side, we're going to screen you from the search results entirely if you've ever worked for that company. Actually. Is it just the current job, Joel, or all of them?

[Joel Spolsky]

All of them. I think, actually just to be safe.

[Jeff Atwood]

If you've ever worked for that company, any point in your history, you will not show up in their search results by default.

[Joel Spolsky]

Yeah.

[Jeff Atwood]

And on top of that, and you.

[Joel Spolsky]

Can change that if you want.

[Jeff Atwood]

Yeah. There's also a filter you can set up. It's basically a regular expression. We'll give you like a tester and everything. And you can set up a string. Say anybody matching the string cannot see me on the, on the employer side. Now, this obviously doesn't, doesn't apply to the public side. And another cool thing about the public CVs is that you can have a public CV just for fun. Just having a public CV doesn't necessarily mean, hey, I'm looking for a job.

[Joel Spolsky]

Right?

[Jeff Atwood]

Right.

[Joel Spolsky]

We pretty much. I think it would be great if, like, literally every working programmer maintained a public cv just so that, you know, if somebody sees your name or Sees you on stack overload and wonders who you are, sees you posting about something, wonders what your experience is, wonders about your credibility, wants to know if they should invite you to speak at a conference and you know, that kind of stuff. You've got a public CV page just to basically announce hey, here's who I am, here's where I've worked, here's what I've done, here's my education.

[Jeff Atwood]

Right, and we're going to offer super granular visibility on the public side so you can have a very detailed CV and just hand pick this, this, this, this. Only these five things will show up on the public.

[Joel Spolsky]

Ah, it's like Facebook.

[Jeff Atwood]

Yeah, well we kind of have to because honestly for me, like I don't really want my CV to be public like at all. Like I just, it's not something I'm really interested in. Plus well I already have my blog. I mean I'm a bad example of this anyway. Anything that I want to public would have been, I would have blogged about already. But I think there are some programmers like me who are just into privacy in terms of their work experience.

[Joel Spolsky]

Oh sure, absolutely. Yeah.

[Jeff Atwood]

So I mean this is totally opt in. Like when you create a cv, there are manual steps to make it a publicly visible to Google with your vanity URL or B file it with employers for to indicate that you're actually looking for a job. None of that is automatic. So everything's private by default 100% of the time. You can opt in to non privacy or making it public.

[Joel Spolsky]

While we're talking about privacy. Actually I should mention that one thing I hear a lot from neurotic programmers is oh, I don't want to have anything, I don't want to appear on the web. And the truth is you're probably on the web. I could search Google for you and I would probably find some article that appeared in your hometown newspaper when you were a kid about how you got in trouble with the police or whatever. This is gonna be some little thing that if I Google view I find and the only way to get those things out of Google and I get email from people saying oh, I posted something embarrassing on your discussion group four years ago and now it's coming up with my name. Could you please take it down please? It's coming up at number one in Google hit for me. And what I constantly tell people is the only way to get rid of negative stuff about you on the web is to just basically wash it out with tons of positive stuff. It's like put Some serious shit on the web. Get your identity on the web, explaining who you are in a way that you control in a positive light so that it dominates it and kicks that article about you breaking your neighbor's windows down to page 23 of the Google results.

[Jeff Atwood]

Right. You want to control your online presence and that means actually publishing information into it.

[Joel Spolsky]

Yeah. And that means actually saying, you know, at the very least, here's my side of the story. You know, I was a programmer here and I did this and these are my qualifications. Sometimes, you know, if I get a resume from somebody or even if I just hear from somebody, you know, I've been looking for speakers for Stack Overflow dev days. A bunch of people emailed me and said, hey, I'd like to speak in Toronto or wherever, Seattle. I'm an expert on such and such. The very first thing I'm going to do is going to punch their name into Google. And sometimes it shows up. I find out very quickly who they are. I see people talking about previous speeches that they did and reviewing them, or I see, you know, where they work. I see that they've written a bunch of books on Amazon. I see what they're doing. Public, public Persona is very quickly, sometimes I don't see any of that. And, you know, all I see is just some, like a few accidental things that just happen to be, you know, what's left and shows up on Google for that person. And that's, you know, for many people, that's rarely the image that they want to have of themselves as the number one hit on Google.

[Jeff Atwood]

Yeah. Much more excited about the public thing. I wish we had actually thought about this earlier because it does change some of the design stuff we have to do. Right, but what are you going to do? And I think it does fit better with the model of Stack Overflow, the people that we let test the career stuff in beta. One of the bits of feedback I got was that, well, this isn't really like Stack Overflow because A, I have to pay money for it eventually, which is optional now, and B, there's no public. There's nothing here that's public. And since everything on Stack Overflow you do is public by default, there was an expectation that the career stuff would in some way be public as well. So now that we're allowing CVs to be public, that I think better fits people's mental model of what they're expecting us to do.

[Joel Spolsky]

Yeah, and I think that's. I think we're being poster children here for the benefits of Shipping early and then adjusting course rapidly. You know, the product that we launched on day one is the first beta that we launched exactly a week ago today was pretty close. It was off by about 10%. We didn't have that public by default. And what was the other thing? We didn't have protection, you know, if you didn't want your resume showing up to previous people. So there's a little bit of course correction there, but I would say it was about 10% change in degree. And it's great to get that feedback quickly and to be able to implement that quickly. And by the time this thing goes out of beta, it'll be awesome, right?

[Jeff Atwood]

Yeah. There's a certain amount of exploration you can only do by shipping a product. There's only so much thinking and private punificating and coding you can do. And to solve the problem, there's just a lot of people that have a lot of different expectations and just you got to ship it and then see what happens and then adjust rapidly. And that's really what we do in Stack Overflow. I mean, we try to. We deploy every night. I think I've mentioned that.

[Joel Spolsky]

Yeah.

[Jeff Atwood]

And every day I try to make Stack Overflow better in some tiny way. And that, to me, is how you build a product. You just slowly, over a period of years, polish the crap out of it.

[Joel Spolsky]

Right.

[Jeff Atwood]

So careers, hopefully will be the same way. Did we have any listener questions?

[Joel Spolsky]

You know, the trouble is, I don't know if we do because you're on the road. I'm on the road, and I don't have. I brought. I forgot. I'm missing one little piece of wire connection that I should have brought that would have allowed me to even play listening to your questions. So I don't even have the technical capability of playing them this week, but I'll have that fixed for next week. When we're doing a show in person, you and I will be together in San Fran and we'll put together a show live in person, and that way we'll be able to get some listener questions in. So if you do have listener questions, something you want to talk about, especially if you want to talk about the new stuff, Stack Exchange, careers.StackOfLow.com, then definitely do call the listener hotline and put those in there and we'll talk about it next week's show. Maybe we can take questions at dev days for the podcast or something.

[Jeff Atwood]

Oh, and give them our number to call in.

[Joel Spolsky]

Yes. Okay. So I guess, do we have a number overflow the Number to call the podcast hotline is 646-826-3879. Eventually I will have that memorized or you can email an MP3 or file out the [e-mail] and we'd love to get those questions in. So what you do is just record a little message up to up to 90 seconds telling us what you want us to talk about and we'll play it on a future episode. Meanwhile, anything else we want to announce real quick? We were mostly just jib jabbing about all our products this week.

[Jeff Atwood]

No, I don't think so. I mean, that's pretty much everything we've been working on.

[Joel Spolsky]

That's what we've been working on.

[Jeff Atwood]

Yeah.

[Joel Spolsky]

We're in the midst of stack overflow dev days. I'm sitting here in Austin. Tomorrow morning we'll have the second dev day. The Boston one I think went pretty well. There were a couple of little tiny things that we'll fix for the next time. Really interesting. Oh, you know, something really, really interesting happened. I do kind of want to talk. I want to bring up a stack overflow question. It's actually a server fault question, which I linked to from my homepage on Joel on Software, which is after the. But what happened in the Boston event is we had pretty good WI fi and suddenly the 254th person or something tried to log on and the DHCP server fell down because it just didn't have a large enough pool of IP addresses. And when I started to dig into this, I discovered that nobody has good WI fi at conferences. It never quite works. And there seems to be a disconnect between the people that run the conferences and the physical, the AV people and the venue themselves. You get to one of these venues and you find a person that knows nothing about computers and they are in charge of the WI fi. And you're trying to tell this person, I'm going to have 600 geeks here tomorrow, all with two devices each trying to get on the WI FI network. They got their iPhone and their laptop and they're trying to get on the network with both of them. And did you check the DHCP server? And they look at you and they smile and they say, we have excellent WI fi. We'll see what happens in Austin tomorrow. I was just talking to the guy last night, really nice guy, really awesome. He said, it's got enough capacity for 4,000 people. Thinking to myself, oh, huh, he's got a WI FI network that can handle 4,000 people. That is great. I am so Psyched to see that work tomorrow because at these conferences it never really works. So point being, I went on to Server Fault and wrote a little question. Why is Internet access and WI fi, why does it suck so badly at tech conferences? And it's server fault question 72767 and if you look at the number one answer there, started by who? I don't know, it's a community wiki by now, but it is a friggin awesome answer. And the number two answer is good and the number three answer is pretty decent and there's a lot of really, really, really good information there. And this particular question about how do you provide WI fi for a room of 400 geeks in a small area or 3,000 geeks or whatever, it's actually interesting. There is kind of an art to that and it's not really documented. There are a few people wandering around in the world that know how to get this to work and they work for some companies that you probably never heard of and you wouldn't think to hire Marriott Systems as an example of them. And, and they know the secret art to getting WI fi to work at a conference. And now you can get most of that information from question number 72767 on ServerFault. So that's my exciting ServerFault question of the day.

[Jeff Atwood]

That was a good one. I like that one.

[Joel Spolsky]

Anyway, if you know anything about WI FI in a large area, it's like a wiki now, right? So go into that question. The number one answer has been edited by a bunch of people. Go in there and tweak it, add a little bit more information, add links to other resources. I spoke to Ryan Carson who's running our conferences from carsonified today and he said, yeah, I just, I went into that question and took my number one guy and said, hey, I want you to read the answer to this question and make sure that we have this at all carsonified conferences from now on as that rapidly becomes the canonical place. And that's the great thing about the Stack overflow server fault model. I'm kind of hyper. I've been drinking a lot of caffeine today.

[Jeff Atwood]

Well, it's kind of a hard problem too. So it was an interesting question from that aspect as well. I mean, you kind of have to throw a lot of resources at it if you have like a giant room full of geeks and all of them are on WI Fi because someone else pointed out like everybody wifi limited.

[Joel Spolsky]

Yeah, yeah, yeah.

[Jeff Atwood]

I mean to make WI fi really good, you got to Throw money at it to some degree. Right. Part of it is I'm not sure.

[Joel Spolsky]

Well, here's. You can read the question, but some of the things which I thought were most interesting, number one, get the heavy users off of it. Because WI fi just has very, very limited bandwidth. There are only a few channels. And so the first thing you need to do is make sure the heavy users are off it. And to do that, the best way to do that is to provide fixed Internet, you know, on a landline, on a wire, in as many places as you could possibly can. You can have places, certain places where bloggers can sit and get fixed wired access. You can even have breakout rooms or break rooms or something that have it. Just so if somebody has a large file to transfer, they go there and use that because the WI fi is causing problems. So that's the first thing. Then there's a lot of details about using the right hardware. There's a lot of questions about, you know, how to get the bandwidth to the venue itself. There's just a whole bunch of interesting stuff to learn about this. It is definitely possible. There are conferences that are doing it. TechCrunch 50 was sort of famous for having spent outrageous amounts of money and they got it to work just by spending money. But if you look at technically what's actually going on there, it's not the money that's making it work. It's actually just the limited number of people that actually know the secret, that actually have the information for how to do it, because you can spend a lot of money. There was a Le Web in Europe was sort of famous because they allegedly spent like $100,000 or something. They gave Squizcom some gigantic amount of money to provide WI FI at the conference. And it did not work at all in any way, shape or form, even for the speakers. I mean, it was just a complete embarrassment. And that's because they didn't know what they were doing, but they knew that it was a big conference so they could charge a lot of money. And you know, there are a few people that do know kind of the secrets. They know what equipment to use, they know how to prepare the room. They'll spend a couple of weeks before checking things out at the venue. There's something interesting I saw about the teched conferences that Microsoft does in that Microsoft did in Australia, which is that when they went in there, I think weeks before the conference was scheduled to start, just to see if there was any, if there's going to be any interference with the WI FI in the conference venue. They found that the venue already had its own WI fi that was on lots and lots of different channels, like lots of different SSIDs that they were making available. Now the way SSIDs work is it's kind of a broadcast thing where every access point says hey, I am SSID blah. I am SSID blah. And they just keep broadcasting that several times a second. And they had so many of these at this venue that they were using up all available bandwidth just broadcaster, broadcaster what their SSIDs were. And so in a real industrial strength system you would only have one SSID for all the access points. And you can do that by using these things that they call thin access points. Where all the access points job is to do is to take the frame off the air and then transmit it wired to a central router which basically serves as an uber access point for the whole venue, even though the little transmitters and receivers are sprinkled all over the place. So the first thing they discovered is that all bandwidth was being used by these SSID broadcasts and they had to basically get the venue's crappy WI fi turned turned off completely before they even started, just to open up a little bit of bandwidth so that they could make a WI FI thing work with a minimum number of SSIDs. So there's really interesting stuff like that. Not that many people know it. They don't know how to debug for that. They don't know how to go to the venue and test for that stuff in advance. And the people that do know it are making it work. And hopefully that information will all start to show up in this article. 72767 on server fault. So anyway, do call in with your questions. The Stack Overflow podcast hotline. There's a transcript weekend key. I'm still doing the end of the show, right? That was all like just extra freebie stuff still. Yes, yes. There's show notes. The show notes list all the URLs that we've mentioned during the show and just like a quick summary kind of thing. And that's [e-mail] and in the show notes there's also a link to a transcript wiki, which is a wiki page by volunteers around the world who transcribe the contents of this show for the benefit of the hearing impaired and to make them searchable by search engines. And you know, just to get angry about something that Jeff or I said, usually what they transcribe is some stupider things that we say. It's usually me so, Anything else?

[Jeff Atwood]

Okay. Well, I guess that's it for this week.

[Joel Spolsky]

That's it for this week. We'll see you next week.

[Jeff Atwood]

See you next week.

[Phil Windley]

You've been listening to Stack Overflow with Jeff Atwood and Joel Spolsky.

[Conversations Network Announcer]

The Conversations Network is a 501c3 nonprofit and we need your help. For a tax deductible donation of as little as $5 per month, you can support this channel and the rest of the Conversations Network. So please visit conversationsnetwork.org to become a member and help us continue to bring our programs to the world for free. Our audio files are delivered by Limelight Networks, the high performance content delivery network for digital media. The post production audio engineer for this program was Joel Spolsky. Our website editor was Jeff Atwood. The series producer is Jeff Atwood.

[Phil Windley]

This is Phil Windley. I hope you'll join me next time for another great presentation from Stack Overflow here. On it, Conversations.