Here is another possible explanation for the correlation between reputation and CV upload:

Higher reputation --> more time on the web to answer questions --> likely unemployed --> more desire to upload CV.

That's a nice theory... I don't have any way to test it, but I've seen a LOT of StackOverflow CVs and have NEVER seen someone who was unemployed.

More likely that they're UNDERemployed, bored at a dead-end, stupid job, so they have time to answer questions.

That said, the typical high achiever on Stack Overflow is also a high achiever at work. I know for a fact that Jon Skeet's employer loves him despite the fact that he has the highest reputation on the site.

This makes it sound like the ideal trajectory for a Stack Overflow user is "Answers a lot of questions, gets a new job, answers almost no questions", which I think is a little problematic.

That said, there are a lot of working arrangements in the world, and "spends almost all time working on assigned tasks" is just one of them. I get quite a bit of Internet karma (numerical and otherwise) while physically present at the day job, in downtime between tasks. My employers and I have an understanding on this: they treat me like an adult and let me manage my own use of time, I keep earning all the gold stars they stick onto my productivity evaluations.

("Has time to X" is not a measurement of productivity. It is a poor proxy for productivity, which is probably being used because management is not competent at actually measuring productivity. If you actually measure productivity, and the numbers say that Engineer X is 40% more productive than the company average, and you think "That's funny, he seems to have time to update his blog and check Facebook in between reading on new Javascript techniques and, oh yeah, there might have been some coding involved", then you know what that tells you? That Engineer X is 40% more productive than the company average.)

The rapid-fire criticism/defense of every Joel-related article on HN makes me wish I had access to such helpful critics.

I'm currently #26 on the rep list and I am employed (which means there is a workplace I have to go to every day and be there during the business hours).

I cannot tell much about my current job's details (I'm bound by an NDA), but mostly it involves not coding as such but rather answering the other people's programming-related questions, in other words exactly what I'm doing on Stack Overflow.

I also have a blog where people ask me the questions and I answer them.

SO and the blog just help me to be in form for the job: almost any problem the people contact me with at work I have been asked about (and answered) a dozen times on SO and the blog.

It would be interesting doing some sort of data gathering on that question. I'm sure it's the jibe Stackoverflow has gotten used to. Who wants an employee that wastes all his time asking & answering questions?

Might be worth proving it wrong.

So one guy == typical?

Look at http://stackoverflow.com/users, those are all famous Netizens, you should be able to figure out if they're high achievers or just basement StackOverflow nerds

I feel like I'm looking at a list of favorite celebrities in a country I've never heard of. I have no idea who any of these people are or what they might be famous for.

I recognize Jon Skeet, Quassnoi, Alex Martelli and cletus... but I know them all from Stack Overflow. I wouldn't say all of those people are famous anywhere in particular other than there?

Alex Martelli I recognise; he's a python guru (behind the Python Cookbook, among other things) now working at google. There's a couple of excellent tech-talks on (mostly) advanced python by him that are well worth watching if you use python at all.

I think you're making it more complicated than it needs to be.

Invested heavily (time) in SO -> more likely to continue to invest in new SO features

I'm actually impressed that they get away with running a site with so much traffic on such minimal hardware. Almost an argument against the cloud/NoSQL/scale-out trends :)

Everybody wants to believe their site is going to be The Next Twitter, and as good engineers, they believe that they should build it super-duper-scalable from the beginning so they don't have to re-architect it if it really does take off.

Pragmatically, 99.99% of startups don't become Twitter and will manage fine with one DB server and one web server, especially if you use a compiled language and a commercial-grade database and nice hardware. And even if you become Twitter, well, they had terrible scaling problems, and it doesn't seem to have hurt them one bit.

Are you guys still in BizSpark, btw?

I heard in a Podcast that they are; and it is valid for a duration of 3 years.

So, they still would be.

That is a good point that is often overlooked. If you're willing to maintain your own servers, you can get _much_ better bang for the buck than leasing them from Amazon (and way better IO performance which matters a lot for your DB).

I sometimes feel like there's two types of people interested in the NoSQL stuff:

1. People building internet scale sites (Facebook, Twitter, Google, etc)

2. People who don't know how to optimize their database.

But maybe I'm just cynical. :)

3. People who don't like sql and don't like row oriented storage.

People who don't know/understand SQL and fail to grasp row-oriented storage.

Are there any examples of domain specific jobs sites that have been very successful?

Craiglist's old "Erotic Services" page :)

You have to pay $30 to upload your CV?! Why would anyone ever pay to get a job?

You're not paying to get a job, you're paying to have your CV seen by employers, which is a heck of a lot less :-)

Having a small barrier to listing a CV ensures that most CVs in the system represent people who are genuinely interested in finding a job and actually have the confidence that they are moderately qualified.

If you've ever been on the employer side, and looked at the resumes you get when you put an open job listing on a high traffic site like monster.com, you get enormous piles of junk, people that are completely unqualified. Our theory is that having 5 qualified candidates is better than having, say, 10 qualified candidates in a pile of 300, both for the employer AND for the employee who would rather be in the smaller pile.

In any case, $29 is a tiny amount. You'll spend a lot more than that on the haircut you get the day before your interview.

I spent $14 for a haircut a few days before my last interview a few weeks ago. It didn't help. I should have spent that money playing craps ;)

The standard is definitely to have the employer pay, but technically they'll just have less to pay you with after paying the finder fees. So, in a way, you're still paying with a head hunter.

I once went to an interview, made an excellent impression with the interviewer, who totally wanted to hire me; but his boss wouldn't let him because the head hunter fee was too expensive.

I don't plan on ever submitting my resume to stack overflow careers. I'm pretty confident in the content, but i feel my lack of participating on the site (no reputation) will work against me, thus making the site useless.

of course I haven't looked for a new job since the economy tanked, for all I know dice may not work any more :)

It depends on where you are and who you're competing against. In smaller markets, or if you have unique skills, employers searching will be finding a small number of employees, and won't care about Stack Overflow participation. So for example if you know lisp and want a job in Austin, you're competing against exactly one person, who only has 424 reputation, so you're pretty likely to get a call anyway even without StackOverflow reputation.

If you are competing against more people, say, if you're looking for a Java job in the Bay Area, it may work against you, but I doubt if it will make the site useless. In my own recruiting using Stack Overflow careers I found plenty of good people who didn't even use Stack Overflow and I still contacted them because their CVs were great.

PS when I search StackOverflow I look for lisp, OCaml, ML, Smalltalk, etc., not because we need them, but because we like to hire people who know them.

Joel, on a slight tangent...we've been seeing all sorts of Jeff and Joel bashing posts lately, which personally I don't get why people get so worked up.

If I had to take a guess though, it would be that as you guys get more and more successful there's a fear that you will succeed in creating the number one method in the world for getting a programming job and it will end up discriminating against people who don't answer questions on StackOverflow or fit your model of a programmer.

That somehow you guys end up being the gatekeepers for defining what a good programmer is and how they get hired.

I was wondering what your thoughts are on this?

That's crazy fantasy extrapolation, almost too absurd to address, but I will anyway since you asked :)

There are 1492 CVs on StackOverflow right now, and maybe 9,000,000 programmers in the world. Even if we're monumentally successful we're still going to be only a tiny fraction of the hiring of programmers that goes on in the world... a TINY fraction.

I should also add that in the bizarre scenario that StackOverflow actually became a common way of getting programming jobs, it would be a metric ton better than the current system of Monster and Dice and emailing around Word resumes which are scanned in by stupid software that looks at keywords.

I think this post is on the money. You have to bear in mind that right now the "gatekeepers" you're referring to are mostly completely non-technical HR people who might as well be using an Ouija board for all their proclivity finding good programmers. One can argue the merits and demerits of judging someone based on their Stack Overflow participation, but it can't be worse than the current reality.

Thanks. I agree, although perhaps you underestimate your ability to influence some of the top recruiters in the world?

I'm just trying to understand where the criticism is coming from because I don't get it.

> That's crazy fantasy extrapolation, almost too absurd to address

There was an uptick in my respect level for your operation when I read that.