Another ivory tower joins those built by Spolsky and Yegge. A nice theory on hiring that is completely impractical, bordering on the surreal.
1) You wouldn't recognize a Dennis-Ritchie- or Claude-Shannon-to-be if they would spit you in the face, because at the point in their lives where they would consider to work for you (if there ever was such a point), their achievements would not stand out nearly as much. Robin Milner died saturday and I just learned he didn't even have a Ph.D. Pretending you'd be even remotely capable of recognizing
2) What gives you the idea that people with such potential would apply at RethinkDB? You are describing a group of maybe a thousand people worldwide. Many have academic ambitions; others are interested in completely different subjects. There are more than a thousand companies looking for these people. If you'd manage to hire one, you'd be lucky.
3) Even if you were able to recognize the ability for such achievements and get them to apply, it is doubtful they would able to achieve those peaks in your corporate environment. Claude Shannon didn't have to satisfy any customers. He could spend years developing a theory, whose practical value was only clear after it was sufficiently developed.
4) A corollary of 2): the people you've hired so far are not the kind of people you describe.
It would be nice if, just for once, someone would try to describe the people they actually hire, instead of pretending they only hire potential Field medalists, Nobel laureates and other people of such fame that all of their actual developers know them.
[..] I am often unable to solve complex problems during
the interview because of the nervous pressure. But I don’t
want to hire people as good as me, I want to hire people
much better than me.
5) People that botch their interview because they are nervous can't be better than you? Perhaps nervousness/self-doubt go hand-in-hand with the ability to be better than you and you're throwing out the babies with the bathwater. Especially considering that someone passionate about working with you would be prone to nervousness.
6) And finally, about that Spolsky quote: I'm convinced he put it there to make sure no one else would hire the ones he would like to hire. If you don't have any doubts about a candidate, then you haven't pried enough to expose their weaknesses. That, or they are too glib. You should have doubts about every single one of your interviewees.
PS: this article, as well as the ones by Spolsky and Yegge referenced in the article, contain sound advice. However, all three of them go overboard at some point, which inspired the above. This comment might be considered as tearing apart a specific point, while ignoring the larger whole. This is not my intention: the article is worth reading.
Great post. Approached from an angle not often taken here.
Oddly, you said one thing that just didn't sit well with me. I re-read it several times trying to figure out why I was so uncomfortable with it:
Claude Shannon didn't have to satisfy any customers. He could spend years developing a theory, whose practical value was only clear after it was sufficiently developed.
This is exactly the opposite of my thinking. I've been in academia and in the trenches. I prefer the trenches.
Where I come from you don't have anything until you're able to satisfy customers. Building theories, tools, and new approaches is great, so great, in fact, that some of the best breakthroughs I've ever seen were because a customer demanded something people like us never imagined or thought possible.
I love satisfying customers with whatever I have in my back pocket. And lots of that stuff was developed by the giants whose shoulders I stand on. I always imagined that they were satisfying their own customers, but I guess quite a few of them were just doing "basic research".
You reminded us that there needs to be room for both the "trench warrior" and the "dreamer". When you're building a team, you need to decide who would be better for the role at hand and approach the recruiting accordingly.
Not relevant to the "hiring process" discussion, but I think the idea that:
Claude Shannon didn't have to satisfy any customers. He could spend years developing a theory, whose practical value was only clear after it was sufficiently developed
is historically wrong; certainly reading the wikipedia page on Shannon makes it sound like he developed his ideas while doing quite practical work during WW II.
Someone please correct me if my impression is incorrect.
With respect, isn't it normal to start with something less than would satisfy customers?
I think the difference you're describing is in direction, not position. On the side you prefer, you figure out what someone wants, and heads toward that. On the other, you head towards something you want or feel like doing. I assume you've found the right project for you when the sides are the same.
I think you make interesting points, but I respectfully disagree with your conclusions.
You wouldn't recognize a Dennis-Ritchie- or Claude-Shannon-to-be if they would spit you in the face
Nine out of ten times, sure, nobody would recognize Dennis Ritchie in the rough. But if you're hiring the first five employees, you don't care about nine out of ten - you care about that one time you do recognize a promising candidate. Every selective process has a false negative rate, there is no way around that. Amazing people get turned away by all kinds of smart people in all kinds of places all the time. If you have ideas on how to lower the false negative rate without increasing the false positive rate, I'd love to hear how you'd do it.
What gives you the idea that people with such potential would apply at RethinkDB?
Because they already have. Smart and driven people apply for the same reasons smart and driven people do anything - they're interested in the problem, and have an innate desire to solve it. In any case, if you sit around and wait for great people to apply, the sourcing part of your recruiting process isn't done right. If you want great people, you have to go out, find them, contact them directly, and then do everything in your power to close them.
People that botch their interview because they are nervous can't be better than you?
Of course they can, but again, it's about increasing the false negative rate to lower the rate of false positives. I can afford to turn some great people away, but I can't afford to hire someone who isn't great. Any software startup that works on hard technology has the same constraints - if you hire people that are anything less than stellar, you significantly lower your chance of success. In a startup you (mostly) don't have geographical limits for your customers, but the corollary is that you also don't have geographical limits for your competition. When you're targeting every customer in the world, you're also competing with every company in the world. Going into that with anything but the top-notch team is suicide.
What gives you the idea that people with such potential would apply at RethinkDB?
Because they already have. Smart and driven people apply for the same reasons smart and driven people do anything - they're interested in the problem, and have an innate desire to solve it. In any case, if you sit around and wait for great people to apply, the sourcing part of your recruiting process isn't done right. If you want great people, you have to go out, find them, contact them directly, and then do everything in your power to close them.
I'm going to hazard a pretty safe guess that one of the reasons you can blithely assert this is that you've never actually personally known someone with the abilities of a Dennis Ritchie or Claude Shannon. It is easy to overestimate the abilities of the best you've known if you've never known the truly best.
This is not a slight on the abilities of some of the developers you are lucky to have. It is a well deserved compliment to some of the giants of computing.
>> What gives you the idea that people with such potential would apply at RethinkDB?
> Because they already have.
So how much of the company do they own? I ask because I have noticed that people with the combination of talent, experience and work ethic you are looking for are rarely content to be just someone else's employee. They'll either be founders of their own company, or hold a low-employee-number position in someone else's business with plenty of stock options, or work as independent consultants, or otherwise do something where what's in it for them is on a par with what they bring to the table.
The only guys I know like this are holed up in some pure research job at universities, working on problems that interest them so much they barely leave their offices to shower. I'm talking top 50 living programmer material though.
Comments
Another ivory tower joins those built by Spolsky and Yegge. A nice theory on hiring that is completely impractical, bordering on the surreal.
1) You wouldn't recognize a Dennis-Ritchie- or Claude-Shannon-to-be if they would spit you in the face, because at the point in their lives where they would consider to work for you (if there ever was such a point), their achievements would not stand out nearly as much. Robin Milner died saturday and I just learned he didn't even have a Ph.D. Pretending you'd be even remotely capable of recognizing
2) What gives you the idea that people with such potential would apply at RethinkDB? You are describing a group of maybe a thousand people worldwide. Many have academic ambitions; others are interested in completely different subjects. There are more than a thousand companies looking for these people. If you'd manage to hire one, you'd be lucky.
3) Even if you were able to recognize the ability for such achievements and get them to apply, it is doubtful they would able to achieve those peaks in your corporate environment. Claude Shannon didn't have to satisfy any customers. He could spend years developing a theory, whose practical value was only clear after it was sufficiently developed.
4) A corollary of 2): the people you've hired so far are not the kind of people you describe.
It would be nice if, just for once, someone would try to describe the people they actually hire, instead of pretending they only hire potential Field medalists, Nobel laureates and other people of such fame that all of their actual developers know them.
5) People that botch their interview because they are nervous can't be better than you? Perhaps nervousness/self-doubt go hand-in-hand with the ability to be better than you and you're throwing out the babies with the bathwater. Especially considering that someone passionate about working with you would be prone to nervousness.6) And finally, about that Spolsky quote: I'm convinced he put it there to make sure no one else would hire the ones he would like to hire. If you don't have any doubts about a candidate, then you haven't pried enough to expose their weaknesses. That, or they are too glib. You should have doubts about every single one of your interviewees.
PS: this article, as well as the ones by Spolsky and Yegge referenced in the article, contain sound advice. However, all three of them go overboard at some point, which inspired the above. This comment might be considered as tearing apart a specific point, while ignoring the larger whole. This is not my intention: the article is worth reading.
Great post. Approached from an angle not often taken here.
Oddly, you said one thing that just didn't sit well with me. I re-read it several times trying to figure out why I was so uncomfortable with it:
Claude Shannon didn't have to satisfy any customers. He could spend years developing a theory, whose practical value was only clear after it was sufficiently developed.
This is exactly the opposite of my thinking. I've been in academia and in the trenches. I prefer the trenches.
Where I come from you don't have anything until you're able to satisfy customers. Building theories, tools, and new approaches is great, so great, in fact, that some of the best breakthroughs I've ever seen were because a customer demanded something people like us never imagined or thought possible.
I love satisfying customers with whatever I have in my back pocket. And lots of that stuff was developed by the giants whose shoulders I stand on. I always imagined that they were satisfying their own customers, but I guess quite a few of them were just doing "basic research".
You reminded us that there needs to be room for both the "trench warrior" and the "dreamer". When you're building a team, you need to decide who would be better for the role at hand and approach the recruiting accordingly.
Not relevant to the "hiring process" discussion, but I think the idea that:
Claude Shannon didn't have to satisfy any customers. He could spend years developing a theory, whose practical value was only clear after it was sufficiently developed
is historically wrong; certainly reading the wikipedia page on Shannon makes it sound like he developed his ideas while doing quite practical work during WW II.
Someone please correct me if my impression is incorrect.
With respect, isn't it normal to start with something less than would satisfy customers?
I think the difference you're describing is in direction, not position. On the side you prefer, you figure out what someone wants, and heads toward that. On the other, you head towards something you want or feel like doing. I assume you've found the right project for you when the sides are the same.
I think you make interesting points, but I respectfully disagree with your conclusions.
You wouldn't recognize a Dennis-Ritchie- or Claude-Shannon-to-be if they would spit you in the face
Nine out of ten times, sure, nobody would recognize Dennis Ritchie in the rough. But if you're hiring the first five employees, you don't care about nine out of ten - you care about that one time you do recognize a promising candidate. Every selective process has a false negative rate, there is no way around that. Amazing people get turned away by all kinds of smart people in all kinds of places all the time. If you have ideas on how to lower the false negative rate without increasing the false positive rate, I'd love to hear how you'd do it.
What gives you the idea that people with such potential would apply at RethinkDB?
Because they already have. Smart and driven people apply for the same reasons smart and driven people do anything - they're interested in the problem, and have an innate desire to solve it. In any case, if you sit around and wait for great people to apply, the sourcing part of your recruiting process isn't done right. If you want great people, you have to go out, find them, contact them directly, and then do everything in your power to close them.
People that botch their interview because they are nervous can't be better than you?
Of course they can, but again, it's about increasing the false negative rate to lower the rate of false positives. I can afford to turn some great people away, but I can't afford to hire someone who isn't great. Any software startup that works on hard technology has the same constraints - if you hire people that are anything less than stellar, you significantly lower your chance of success. In a startup you (mostly) don't have geographical limits for your customers, but the corollary is that you also don't have geographical limits for your competition. When you're targeting every customer in the world, you're also competing with every company in the world. Going into that with anything but the top-notch team is suicide.
What gives you the idea that people with such potential would apply at RethinkDB?
Because they already have. Smart and driven people apply for the same reasons smart and driven people do anything - they're interested in the problem, and have an innate desire to solve it. In any case, if you sit around and wait for great people to apply, the sourcing part of your recruiting process isn't done right. If you want great people, you have to go out, find them, contact them directly, and then do everything in your power to close them.
I'm going to hazard a pretty safe guess that one of the reasons you can blithely assert this is that you've never actually personally known someone with the abilities of a Dennis Ritchie or Claude Shannon. It is easy to overestimate the abilities of the best you've known if you've never known the truly best.
This is not a slight on the abilities of some of the developers you are lucky to have. It is a well deserved compliment to some of the giants of computing.
>> What gives you the idea that people with such potential would apply at RethinkDB?
> Because they already have.
So how much of the company do they own? I ask because I have noticed that people with the combination of talent, experience and work ethic you are looking for are rarely content to be just someone else's employee. They'll either be founders of their own company, or hold a low-employee-number position in someone else's business with plenty of stock options, or work as independent consultants, or otherwise do something where what's in it for them is on a par with what they bring to the table.
The only guys I know like this are holed up in some pure research job at universities, working on problems that interest them so much they barely leave their offices to shower. I'm talking top 50 living programmer material though.
Nobody should genuinely expect a "Dennis-Ritchie- or Claude-Shannon-to-be". That's just hyperbolic way of saying "good employee".