4 ms·

I don't think there is much intrisincally different between Facebook and Google. That is, when Google had only 2.5K employees it was attracting people like Lars

I don't think there is much intrisincally different between Facebook and Google. That is, when Google had only 2.5K employees it was attracting people like Lars from other companies like MS and Yahoo, and for the same reasons. The problem is scale. It's much easier to find 2.5K smart people than it is to find 25K smart people, and so when you're at 25K the average intelligence of your population is much lower. I don't think it's so much the extra management than the reduced signal-to-noise ratio that Lars is referring to. If you've worked at companies of different sizes (I have) the difference is tangible.

idoh 16y ago
In your opinion, is there anything a company can do to keep or improve the average level of employees as the company grows, or is it just inevitable?
- staktrace 16y ago
  Interesting question. I think it's possible to maintain and even improve the average level of intelligence of employees, but past a certain size it's so ridiculously difficult as to be infeasible, practically speaking. Also, I'm not convinced that it's in the company's best interests, but I guess that depends on what the goal of the company is. If, like most companies, it is to make money, then having too many smart people could be a liability; after a certain point you want to hire people for reliability and teamwork rather than individual intelligence. Even if the company has some other goal, I would say it's probably easier to try to keep the company small and automate as much work as possible than it is to keep hiring really smart people.
- numbchuckskills 16y ago
  Google employs the Lake Wobegon strategy: http://googleresearch.blogspot.com/2006/03/hiring-lake-wobegon-strategy.html http://googleresearch.blogspot.com/2006/03/hiring-lake-wobeg... "only hire candidates who are above the mean of your current employees"
nostrademons 16y ago
I don't think the average intelligence at Google has declined very much. The people I work with are still all very, very smart - Lars said much the same thing in his article. There've been quite a few very intelligent latecomers recently - Alex Russell and Joshua Schacter both joined in 2009, even if Schacter did leave eventually. And I've heard a bunch of old-timer friends - people hired in 2005, or even in 2002 - say that there are recent, fairly inexperienced hires doing amazing things. I think the problem is that the intelligence of a group is defined by the minimal intelligence of its members, not by the average. If you have even one member who just doesn't get it, then everyone else needs to dumb down their ambitions so that he doesn't get left behind. That's why the "no false positives" rule is so important in startup hiring, and why the public school system sucks so much. I think it's a fairly good bet that with 25k employees, the dumbest person at Google is dumber than the dumbest person in FaceBook. And with Google's shared infrastructure, it's fairly likely that you'll run across some design decision made by him.
- staktrace 16y ago
  I agree that the minimal intelligence is going to be lower, but I also don't think it matters as much as you think. Consider the extreme: say Google has 24,999 really smart people and one dumb person, do you really think that one person is going to have that much of an impact? there are recent, fairly inexperienced hires doing amazing things Maybe, but is that true of the average new hire? I'm sure Google is still hiring a fair number of smart people who will do amazing things; my argument is that the average has dropped. You can cite individual examples of smart people but those aren't a counter-argument unless you have a statistically significant number of them. Remember, if you're going to use free food and candy stations as hiring tools, don't be surprised when the type of people you hire are more interested in eating free food and candy than building something awesome.
dasil003 16y ago
Scale may have something to do with it, but I don't think it's intelligence. People at Google are not sitting around complaining about the idiots they have to work with. That happens almost anywhere except Google, but Google has had pick of the litter for years now selecting heavily for intelligence. As an early-stage startup guy and hired gun, my outside perspective is of course biased towards the smaller company, so take this with a grain of salt: I think the main thing is that Facebook has a unified product and vision, with really strong UI, and a lot of hardcore infrastructure all centered around a global experience. Meanwhile Google has one strong business, and is repeatedly throwing spaghetti at the wall to find new areas to expand to. Sure there's a lot of great products there, but nothing with the potential Facebook has.
- Mpdreamz 16y ago
  I might eat this words but nothing facebook offers to the masses has had the same impact google search and maps has. Google might have a terrible track record with sideproject but its core ventures still have immense potential.
  - dasil003 16y ago
    Well I'm actually took the other side of this argument in a comment 4 hours ago, in what I thought was a pretty solid argument, however I'm getting reamed in voting, so maybe I'm wrong... On the other hand, hackers (myself included) value raw utility highly and maybe on average we just prefer google search and google maps as tools due to the vapidity of Facebook. I certainly feel that way. However, comparing the impact on the larger world, I really really think Facebook has made way more impact than we care to admit. They got the whole world on a single social network. They created the newsfeed, some thing which is ubiquitous across all sites now. The Facebook platform, though buggy and heinous was completely revolutionary when they launched it. They created a single sign on that is used by a staggering number of people across the web. Personally, Facebook's innovation over MySpace is every bit as impressive as Google over Altavista or MapQuest. We may not value the ends, but 500,000,000 users with orders of magnitude more engagement than Google is undeniable impact.

idoh 16y ago

In your opinion, is there anything a company can do to keep or improve the average level of employees as the company grows, or is it just inevitable?

staktrace 16y ago
Interesting question. I think it's possible to maintain and even improve the average level of intelligence of employees, but past a certain size it's so ridiculously difficult as to be infeasible, practically speaking. Also, I'm not convinced that it's in the company's best interests, but I guess that depends on what the goal of the company is. If, like most companies, it is to make money, then having too many smart people could be a liability; after a certain point you want to hire people for reliability and teamwork rather than individual intelligence. Even if the company has some other goal, I would say it's probably easier to try to keep the company small and automate as much work as possible than it is to keep hiring really smart people.
numbchuckskills 16y ago
Google employs the Lake Wobegon strategy: http://googleresearch.blogspot.com/2006/03/hiring-lake-wobegon-strategy.html http://googleresearch.blogspot.com/2006/03/hiring-lake-wobeg... "only hire candidates who are above the mean of your current employees"

nostrademons 16y ago

I don't think the average intelligence at Google has declined very much. The people I work with are still all very, very smart - Lars said much the same thing in his article. There've been quite a few very intelligent latecomers recently - Alex Russell and Joshua Schacter both joined in 2009, even if Schacter did leave eventually. And I've heard a bunch of old-timer friends - people hired in 2005, or even in 2002 - say that there are recent, fairly inexperienced hires doing amazing things. I think the problem is that the intelligence of a group is defined by the minimal intelligence of its members, not by the average. If you have even one member who just doesn't get it, then everyone else needs to dumb down their ambitions so that he doesn't get left behind. That's why the "no false positives" rule is so important in startup hiring, and why the public school system sucks so much. I think it's a fairly good bet that with 25k employees, the dumbest person at Google is dumber than the dumbest person in FaceBook. And with Google's shared infrastructure, it's fairly likely that you'll run across some design decision made by him.

staktrace 16y ago
I agree that the minimal intelligence is going to be lower, but I also don't think it matters as much as you think. Consider the extreme: say Google has 24,999 really smart people and one dumb person, do you really think that one person is going to have that much of an impact? there are recent, fairly inexperienced hires doing amazing things Maybe, but is that true of the average new hire? I'm sure Google is still hiring a fair number of smart people who will do amazing things; my argument is that the average has dropped. You can cite individual examples of smart people but those aren't a counter-argument unless you have a statistically significant number of them. Remember, if you're going to use free food and candy stations as hiring tools, don't be surprised when the type of people you hire are more interested in eating free food and candy than building something awesome.

dasil003 16y ago

Scale may have something to do with it, but I don't think it's intelligence. People at Google are not sitting around complaining about the idiots they have to work with. That happens almost anywhere except Google, but Google has had pick of the litter for years now selecting heavily for intelligence. As an early-stage startup guy and hired gun, my outside perspective is of course biased towards the smaller company, so take this with a grain of salt: I think the main thing is that Facebook has a unified product and vision, with really strong UI, and a lot of hardcore infrastructure all centered around a global experience. Meanwhile Google has one strong business, and is repeatedly throwing spaghetti at the wall to find new areas to expand to. Sure there's a lot of great products there, but nothing with the potential Facebook has.

Mpdreamz 16y ago
I might eat this words but nothing facebook offers to the masses has had the same impact google search and maps has. Google might have a terrible track record with sideproject but its core ventures still have immense potential.
- dasil003 16y ago
  Well I'm actually took the other side of this argument in a comment 4 hours ago, in what I thought was a pretty solid argument, however I'm getting reamed in voting, so maybe I'm wrong... On the other hand, hackers (myself included) value raw utility highly and maybe on average we just prefer google search and google maps as tools due to the vapidity of Facebook. I certainly feel that way. However, comparing the impact on the larger world, I really really think Facebook has made way more impact than we care to admit. They got the whole world on a single social network. They created the newsfeed, some thing which is ubiquitous across all sites now. The Facebook platform, though buggy and heinous was completely revolutionary when they launched it. They created a single sign on that is used by a staggering number of people across the web. Personally, Facebook's innovation over MySpace is every bit as impressive as Google over Altavista or MapQuest. We may not value the ends, but 500,000,000 users with orders of magnitude more engagement than Google is undeniable impact.