Live data from Hacker News

Things I've learned at Google so far

bentilly.blogspot.com

101–110 of 197 posts

Re: Things I've learned at Google so far

#101

Here's a question, and I don't mean this in a snarky way at all. It's a genuine question. With all these smart people, why has the experience of google search not improved substantially over the last 10 years? Sure, there's more little geegaws in search now, but the experience of finding what I want hasn't gotten any easier.

I would assume that for most people, the experience of search is easy and is working well. Otherwise I don't think it would continue to be #1. It sounds like you might be outside the norm.

The fact that they're #1 just means they're the best of the available options. It doesn't mean the experience is great, just better than what else is out there.

Re: Things I've learned at Google so far

#102

Earlier quoted context omitted.

Alright, I'll bite. We've all worked with really bright individuals before. What magically transformative thing happens when they become the average that would be so incredible and unbelievable to non-googlers? I would describe CS grad school as having "really bright people" as the norm, but while pretty cool, I would never deign to think it as incomprehensible to outsiders or that "nobody would believe [me] anyway".…

In my experience, Google employees individually are not much smarter or more productive than what you'd find in a top-tier computer science department or successful startup. Google's engineering reputation comes from having 20,000 of them, all with access to the same code & information, similar cultural values, and mostly aligned goals. It's a critical mass problem. One smart person is a useful open-source library. T…

Interesting, thanks!

I'd still like to someday understand what makes Google qualitatively different from other companies. Because I don't think 20k smart engineers necessarily yields a functional company nor is immediate access to the top levels always a good thing (for obvious reasons).

Re: Things I've learned at Google so far

#103
post #67

Earlier quoted context omitted.

I don't see the site I used to estimate my IQ before. But http://www.davidpbrown.co.uk/psychology/iq-conversion.html presents a calculator. And if you google around you can find other conversion charts that vary slightly from each other.

2340 on the calculator puts you at the 100th percentile, and at about 2100 on the pre-95 SAT (which only went up to 1600). You sure this has the right version of the score?

The GRE has 3 parts. Verbal, analytical and quantitative. For that you're supposed to put in the sum of the first 2 of them, not the sum of all three.

I got 800 on the quantitative part, so put in 1540.

Re: Things I've learned at Google so far

#105

Earlier quoted context omitted.

Alright, I'll bite. We've all worked with really bright individuals before. What magically transformative thing happens when they become the average that would be so incredible and unbelievable to non-googlers? I would describe CS grad school as having "really bright people" as the norm, but while pretty cool, I would never deign to think it as incomprehensible to outsiders or that "nobody would believe [me] anyway".…

In my experience, Google employees individually are not much smarter or more productive than what you'd find in a top-tier computer science department or successful startup. Google's engineering reputation comes from having 20,000 of them, all with access to the same code & information, similar cultural values, and mostly aligned goals. It's a critical mass problem. One smart person is a useful open-source library. T…

Something else I read someplace (so it must be true!) is that Google has maniacally strict rules about what programming languages are allowed and that everything must go into company-wide version control. If that's true, and if they also have good code-conventions, I can see how that would have a magnifying effect on productivity and technology improvement.

Re: Things I've learned at Google so far

#106
post #72

Earlier quoted context omitted.

Why should that be off-putting? Standardized tests are pretty much the only way to objectively measure intelligence. Writing ability doesn't capture stuff like mathematical skill. Yes, yes, tests are imperfect. Everyone understands this. But they're the best measurement we have -- short of getting to know someone over the course of a few months or years. Also, tests, because of the way they establish some people as "…

I dunno about others but when I read it sounded a little arrogant. For example, I have a Mensa gold certificate saying that I scored in the top 0.5% percentile on a test (actual IQ scores vary based on the type of test but mine was 143.5). Sounds pretty arrogant, eh? Generally any statement that equates to "I am intelligent" gets taken this way.

Any sentence containing the word "Mensa" is arrogant.

Re: Things I've learned at Google so far

#107
post #12
post #10

As my boss likes to say, they have the free meals because they expect you to be there for breakfast, lunch, and dinner.

Ha! I've visited Mountain View enough to know that it's a ghost town when breakfast is actually being served.

This is true. I made it to breakfast once in an entire year.

Re: Things I've learned at Google so far

#108

Earlier quoted context omitted.

In my experience, Google employees individually are not much smarter or more productive than what you'd find in a top-tier computer science department or successful startup. Google's engineering reputation comes from having 20,000 of them, all with access to the same code & information, similar cultural values, and mostly aligned goals. It's a critical mass problem. One smart person is a useful open-source library. T…

Interesting, thanks! I'd still like to someday understand what makes Google qualitatively different from other companies. Because I don't think 20k smart engineers necessarily yields a functional company nor is immediate access to the top levels always a good thing (for obvious reasons).

I was at Google in 2007, here's my take on it.

The people that are working there aren't just the most brilliant in the world, they literally invented a good portion of the internet, computers, computer science, and programming as we know it. They have "the father of the internet" (Vint Cerf), the creator of Vim (Moolenaar), the creator of Python (Guido), the lead developer of Firefox (Goodger), several of the lead Linux kernel developers, the creator of memcahced (Fitzpatrick), Ken Thompson and Rob Pike (both notable for several things), and hundreds of other developers that have already proven their worth (Peter Norvig and Jeff Dean come to mind).

These aren't just smart people, they're smart people who have changed the world.

To put it into context, one day I'm sitting in a meeting and someone says "Mike is working on an algorithm for that." Another person asks, "Burrows?", to which the first person says "Yea." They were talking about Mike Burrows, as in the guy who created the Burrows-Wheeler Transform (among other things). Names of people that have had major impacts like this are routinely brought in casual conversation. It's like working in a programming mecca. The crazy thing is, even the names you wouldn't recognize have most likely had substantial impacts as well.

As for critical mass, Google has it. I was standing in a hallway one day reading stats about search traffic on one of the walls (the wall slightly beyond Larry and Sergey's office, for those who have been there), and sure enough Larry, Steve Jobs, and some other guys walk on right by me. When you're big enough to have people like Steve Jobs walking around your halls... some doors open up that simply aren't open to most start ups and companies.

The thing is though that the company is really laid back. I was eating lunch one day with my team and there was an open seat. Sergey Brin came over and asked if he could eat with us. Another time it happened during TGIF, and we just sat around and drank beers. At a different TGIF I drank beers with Alex Martelli and we talked about some of the Python books he's written and how he got into hacking on Python. TGIF, in case you didn't know, is a weekly meeting that the entire company has. Every Friday the company gets together, Larry and Sergey get up and talk about what Google has done during the past week and then the floor opens up for Googlers to criticize the company, recommend things they can do better, or show appreciation for things well done. Read that again... Every week the company gets together as a whole and does a self-analysis. Any issues that are brought up will be resolved by the next Friday. When Google does something stupid, it's addressed and fixed in under a week. Most companies have a meeting once a year, and there is no self-analysis going on. After the self-analysis is done, Google has beer, food, and often live music. The company just kicks back, relaxes, and bullshits over beer. They do this every week.

As for people that haven't yet made a name for themselves... Google will give you the opportunity to. I was there for two weeks before I was running distributed programs across thousands of machines (prior to being there I had never even done serious multi-threaded programming, yet alone distributed computing). I had a copy of the Internet on one of our teams shares that I would run tests and experiments against. Yea, you read that right, a copy of the Internet (text-only). You are effectively given unlimited resources to do whatever you want. Money is never an issue, only brain power is ever in short supply. You'll never be given hard deadlines, you'll never even really be told what to work on... everything is more of a suggestion. You can find a bug, fix it, and push it live in under 4 hours (I've seen this done) and no one will stop you. If you ever get bored and want to join a different team... your desk and items will be moved to that team within 24 hours (assuming that team can take you on). Engineers are viewed as stem cells... if everyone is brilliant than anyone can learn to fill any role, and they do.

The teams bond really well. No one eats lunch alone, you always eat with your team. All the food is free, so you don't have people packing lunches, and all the food is delicious, so you don't have people leaving campus for lunch. Whenever there was a big blockbuster movie(like Transformers when I was there), Google would buy out the local theaters on opening day and take the company to see them. They made sure that you were having fun, and they made sure that you had time to interact with your team in non-professional settings.

Also, everyone in the company is at your disposal to help. The organization is nearly flat, but it doesn't matter... you'll never see someone say "I'm the Director of Blah", everyone always just uses the title "Software Engineer". It's quite a sensation. I haven't even begun to scratch the surface of what it's like working there. It is truly quite difficult to explain to someone who's never been there.

All that said, it's not perfect and there are lots of sore spots (especially as they keep growing). I won't get into them though simply because this is already long winded enough.

Re: Things I've learned at Google so far

#109

Earlier quoted context omitted.

I've never seen the problem with being at work for all three meals, given that breakfast and lunch closely bracket work hours. If you are young and don't have a family, having your employer cook you healthy meals is probably a net time-saver, even if you do work 9am - 7pm.

If you are young and don't have a family, having your employer cook you healthy meals is probably a net time-saver, even if you do work 9am - 7pm. Unfortunately, maintaining that work schedule means I'd have to continue to not have a family.

My father's pushing 60 and still working those hours, in IT. If you're going to be working those hours, you might as well spend it with smart people, get paid well, and have free food while you're at it.

Re: Things I've learned at Google so far

#110

Earlier quoted context omitted.

In my experience, Google employees individually are not much smarter or more productive than what you'd find in a top-tier computer science department or successful startup. Google's engineering reputation comes from having 20,000 of them, all with access to the same code & information, similar cultural values, and mostly aligned goals. It's a critical mass problem. One smart person is a useful open-source library. T…

Something else I read someplace (so it must be true!) is that Google has maniacally strict rules about what programming languages are allowed and that everything must go into company-wide version control. If that's true, and if they also have good code-conventions, I can see how that would have a magnifying effect on productivity and technology improvement.

It's generally true for production code. For prototyping we can use whatever we want. Typically you don't want to build off prototype code, though.
Post reply on HN