9 ms·

The problem is that everyone is trying to hire the best, partly because they know how good that looks when fundraising, partly because that's how you win in win

by hiou 12y ago

The problem is that everyone is trying to hire the best, partly because they know how good that looks when fundraising, partly because that's how you win in winner take all tournament type economies. This ends up similar to NFL Quarterbacks. Even the worst quarterback in the NFL is exceptional, but everyone wants the best one because it's hard to win a Super Bowl with the worst quarterback in the NFL. So it's not that talent is hard to find. There are 1,000s of of people that are very good quarterbacks. It's that the best talent is hard to find. And this is because there can only be a couple of bests at anything.

derefr 12y ago
People also try to hire the best because the value of a team in an acquihire scenario is basically the sum of their "ranking scores"—if you're the #1 most-wanted developer in the whole talent pool, the acquirer will pay exponentially more than they will for the #2 most-wanted, and so forth. If you want to succeed as a startup, recruit by setting a bar and hiring anyone who passes it, not by comparing people to other people. "Good in an absolute sense" is much more useful knowledge than "better than X."
- wuliwong 12y ago
  Where are these rankings kept? I honestly never had a clue that there were rankings for developers. I've never once seen that or heard it talked about.
  - derefr 12y ago
    I didn't mean that there are literal rankings; to be the "#1 most-wanted" developer is just to be more well-known and sought-after (and thus to have higher social status and clout in the industry) than the "#2 most-wanted." Google collects highly-ranked developers (e.g. Rob Pike)—for no other reason than their high rank—and calls them Fellows. Heroku hired Ruby's creator Matz, and let him continue working on Ruby itself in an open-source fashion, just to say they had him. It's like collecting baseball cards, basically.
bootload 12y ago
"... This ends up similar to NFL Quarterbacks. Even the worst quarterback in the NFL is exceptional, but everyone wants the best one because it's hard to win a Super Bowl with the worst quarterback in the NFL. ..." This reminds me of Moneyball and Billy Beane. Then I found Kirk Goldsberry doing the same for basketball. [1] It's going to happen to programming. [1] Mark McClusky, 'This Guy’s Quest to Track Every Shot in the NBA Changed Basketball Forever' ~ http://www.wired.com/2014/10/faster-higher-stronger/ http://www.wired.com/2014/10/faster-higher-stronger/
dasil003 12y ago
Where the analogy breaks down is that Quarterback is a very specific job that millions of people train for from a young age. Despite the uncertainty over how top prospects will pan out in the NFL, and the influence of the supporting cast, etc, there still is a huge amount of objective data to look at about performance over time. To make a startup succeed you need some very special people. Many brilliant programmers, designers, sales people, executives, etc do not have what it takes to build a successful company. Heck, many people have what it takes to build one successful company, but not another one, or they have what it takes if they are with certain complementing individuals but not others. This is because unlike sports teams, every company is different, and every successful one succeeds in a unique way. The number one criteria for the initial hires should be people who, in addition to whatever concrete skill you need right now, also have a strong ability to feel out this invisible path and separate the 99% chaff of potentially promising tasks from the 1% that actually get you closer to a viable business. Now how do you filter for that? Hell if I know.
fleitz 12y ago
Sabre metrics.

by hiou 12y ago

derefr 12y ago

People also try to hire the best because the value of a team in an acquihire scenario is basically the sum of their "ranking scores"—if you're the #1 most-wanted developer in the whole talent pool, the acquirer will pay exponentially more than they will for the #2 most-wanted, and so forth. If you want to succeed as a startup, recruit by setting a bar and hiring anyone who passes it, not by comparing people to other people. "Good in an absolute sense" is much more useful knowledge than "better than X."

wuliwong 12y ago
Where are these rankings kept? I honestly never had a clue that there were rankings for developers. I've never once seen that or heard it talked about.
- derefr 12y ago
  I didn't mean that there are literal rankings; to be the "#1 most-wanted" developer is just to be more well-known and sought-after (and thus to have higher social status and clout in the industry) than the "#2 most-wanted." Google collects highly-ranked developers (e.g. Rob Pike)—for no other reason than their high rank—and calls them Fellows. Heroku hired Ruby's creator Matz, and let him continue working on Ruby itself in an open-source fashion, just to say they had him. It's like collecting baseball cards, basically.

bootload 12y ago

"... This ends up similar to NFL Quarterbacks. Even the worst quarterback in the NFL is exceptional, but everyone wants the best one because it's hard to win a Super Bowl with the worst quarterback in the NFL. ..." This reminds me of Moneyball and Billy Beane. Then I found Kirk Goldsberry doing the same for basketball. [1] It's going to happen to programming. [1] Mark McClusky, 'This Guy’s Quest to Track Every Shot in the NBA Changed Basketball Forever' ~ http://www.wired.com/2014/10/faster-higher-stronger/ http://www.wired.com/2014/10/faster-higher-stronger/

dasil003 12y ago

Where the analogy breaks down is that Quarterback is a very specific job that millions of people train for from a young age. Despite the uncertainty over how top prospects will pan out in the NFL, and the influence of the supporting cast, etc, there still is a huge amount of objective data to look at about performance over time. To make a startup succeed you need some very special people. Many brilliant programmers, designers, sales people, executives, etc do not have what it takes to build a successful company. Heck, many people have what it takes to build one successful company, but not another one, or they have what it takes if they are with certain complementing individuals but not others. This is because unlike sports teams, every company is different, and every successful one succeeds in a unique way. The number one criteria for the initial hires should be people who, in addition to whatever concrete skill you need right now, also have a strong ability to feel out this invisible path and separate the 99% chaff of potentially promising tasks from the 1% that actually get you closer to a viable business. Now how do you filter for that? Hell if I know.

fleitz 12y ago

Sabre metrics.