Hacker News new | past | comments | ask | show | jobs | submit login
Inside Google's Secret Lab (businessweek.com)
98 points by tonez on May 23, 2013 | hide | past | favorite | 36 comments



Perhaps it is just late and I'm cynical but this screamed "hey we're still cool, c'mon see?" There were a series of articles about Microsoft's research projects that had the same vibe. Perhaps it is the BusinessWeek lens. Hard to say.

I cringe though on that title, if its a "secret" lab then you wouldn't know about it, so it isn't secret, so what is it?


It's OK - browsing HN all day can make a lot of people cynical. I'm cynical when it comes to cynicism.

You don't need to analyze tone and phrasing here. Just look at the facts. Google has employed some of the most diverse and qualified people in the world to work in a lab with funding that most researcher-builders could only dream of. That alone should excite the mind to think of the possibilities.

We know the lab continues to develop self-driving cars, and that they've already put a bunch of miles on them. I don't care how cynical you are - that is AMAZING! Self-driving cars will change the world. Even if we can cut the death rate in half, that's over 17,000[1] loved ones who will return home each day rather than dying unexpectedly and without warning. That thought should at least inspire us! This is to say nothing of preventing minor wasteful accidents and injuries so common we measure them in millions.[1]

The lab itself might not be secret, but its projects are. Let's not argue meaningless semantics. Judging by the people who have gone into the lab, and the things that have come out so far, there's some pretty awesome stuff happening inside. I for one am thrilled to see what they come up with.

[1] http://www.census.gov/compendia/statab/2012/tables/12s1103.p...


I certainly agree that its great that Google is willing to invest in this stuff, and that reminds me of Xerox PARC from the 1980's. Thus anyone who is looking to research really "out there" sorts of ideas should seek to join Google. I also think that Google should think about the PR tone (which I know they do on many things) for this sort of article too, caveat you really don't get a huge amount of control on how someone is going to use the information you give them, once they get it.


It's easy to get cynical about business articles and anything with "secret" in the title. I get that way myself. But since the demise of Bell Labs, it feels like there's a big void to be filled. It probably never will be filled in the same way, but I'm happy to see a big company take on big projects. Other than CERN or NASA, I don't hear too much in the news about big budget R&D / science projects.


University teams around the country have partially filled that void.

DARPA is doing some pretty neat things as well.


I believe that IBM still invest significantly in research, including basic research.


The UAV or aerostat balloon broadband relay thing heavily hinted at in this article is one of the more amazing ideas. Totally feasible with current technology and highly beneficial; it seems like an obvious thing to do even now.

I guess I'd care about it more for disaster or conflict zone operations vs. ongoing operations in poor countries (since vastly more money is available in the short term in the first case), but it would probably make sense everywhere. At 50k feet, you can cover a pretty large area with spot beams, and probably do aerostat to aerostat relays. Combined with undersea fiber, you could do a good job of providing high speed communications services to some underserved markets.

I assume you'd use high altitude aerostats to provide low bandwidth coverage to large areas, and then local, lower altitude balloons or uavs, fed by the high altitude stuff, in areas of high user density.



Another use case I see is their potential use at sporting arenas for high profile events, cause you know people get bored/distracted at their football games and congest the network with massive traffic.


Or even more usefully, in random places which have concerts/festivals only infrequently. You'd think a sports stadium could have permanent RF reinforcement gear installed (and just turned on sometimes).


Possibly, but maybe it would be cheaper for them to pay for the service from google or whomever (depending on how frequent their network is overly congested) and not have to worry about having special staff on hand to manage something themselves (maybe even the not having to worry about it is enough to shell out cash in itself).


I am reminded of my former company's CTO who once told me that true and disruptive R&D can only be done by a company with a monopoly power, so they can hide their true margins in blue-sky work. I for one hope they can do this as long as possible, since I love Google and I love Google X and I worked for two great R&D departments that were cut short on all long-term projects by controllers looking to post higher margins in competitive environments.


This sounds like something straight out of Wayne Enterprises. I wish more companies could do things like this and see what can be created and achieved. An area that I could see being able to do something like this without having to face Shareholders, management, etc. that most companies have to do is Universities. University students should be able to take advantage of the resources that universities have and universities should encourage this. They could essentially have a new flock of minds every year to help innovate projects. I believe Georgia Tech started something like this a few years ago.


The awesome things you can do when your key product masquerades as a mint.


Here's an honest question: how do the unluckier 90 percent, who face calibration scores and Perf and closed allocation, not find themselves in revolt when they learn that others work in a blue-sky environment on challenging problems that will build their careers?

How is the fact that most people at Google won't get to work on stuff like this not a gigantic HR issue?


Most of us in the non-X parts of Google are actually pretty content. I for one am happy for the X team, not jealous. Though I do admire them.

You might as well be asking how NASA plans to avoid a revolt of the rank-and-file technicians because only a small number of people get to be astronauts.


There are different kinds of people though (uh oh, one of those analogies).

There are people who are content to work on unambitious stuff. People who, say, are happy to do python stuff at reddit working on translation and obsessing over starcraft, all while transitioning their personal appearance towards more of an unkempt, yet smiling, middle-eastern shop owner vibe.

Then, there are people who have a larger, possibly delusional, view of themselves. They don't see the point in non-self-directed work. These people tend to see the world as "I could be making something to help millions of people — why am I sitting here fighting mysql crapping itself the same way for the 40th time this year? Is the computer working for me or am I working for it?"

Why does the second group of people put up with not being in the prestigious X teams? That's between them and their lack of backbone to assert themselves (or lack of social connections to become part of the 'in crowd'), but I'd wager there are more silent resentful types than happy lower-rank people.


And then there's the 99% between those two extremes, who have ambitious long-term goals but feel the best way to reach them is through patient perseverance, as opposed to getting frustrated right away and quitting in disgust at the first sign of a setback.

> Why does the second group of people put up with not being in the prestigious X teams?

In my case, it's because there are plenty of impactful, prestigious, and fulfilling positions elsewhere in the company, and many of them are a better match for my skill set.

BTW, I decided to pick someone at random who had helped millions of people, and see what their career was like. The first name to come to mind was Jonas Salk. Here's an excerpt from his Wikipedia page:

By 1947, Salk decided to find an institution where he could direct his own laboratory. After three institutions turned him down, he received from William McEllroy, the dean of the University of Pittsburgh School of Medicine, an offer which included a promise that he would run his own lab. He accepted, and in the fall of that year, left Michigan and relocated to Pennsylvania. But the promise was not quite what he expected. After Salk arrived at Pittsburgh, "he discovered that he had been relegated to cramped, unequipped quarters in the basement of the old Municipal Hospital," writes Bookchin. As time went on, however, Salk began securing grants from the Mellon family and was able to build a working virology laboratory, where he continued his research on flu vaccines.

And it was a result of that long period of hard work that he got the invitation to work on a polio vaccine. It's a good thing he didn't quit in a huff when the going got tough.


The second group of people, once they try to actually build something to help millions of people on their own without anybody to blame, often find that fighting MySQL crapping itself is an integral part of building anything that helps millions of people.


Postgres.


....is handy, I used it myself for my startup and liked it a lot better than MySQL. Regardless, you're going to face annoying schleps whatever you work on.


That is true. Even the most exciting work has mundane and difficult elements. There's no avoiding that.

What's infuriating is when it's all the mundane stuff, because management took the exciting parts and made them political favors.


> What's infuriating is when it's all the mundane stuff, because management took the exciting parts and made them political favors.

Yeah, that would be awful. Thankfully Google isn't that sort of place.


Most people learn pretty quickly that comparing yourself to other people is a pretty quick way to make yourself unhappy. The unluckier 90 percent (actually, it's more like 99%) realize that they are phenomenally lucky to be in the 0.1% of people who are hired by Google in the first place.

(More generally - most people simply outgrow thinking like this around college or their first job. I have a bunch of really smart friends. Many of them are like, "I thought I was hot shit in high school, but then I got to college and realized I'm pretty ordinary" or "I thought I was hot shit in college, but then I got to the working world and realized I'm pretty ordinary" or "I thought I was hot shit in my last employer, but then I got to Google and realized I was pretty ordinary" or "I thought I was hot shit at Google, but then I got to Search/Research/X and realized I was pretty ordinary." And then they're forced to develop a self-image that doesn't involve being "better" than other people, because you realize that no matter how high you go there will be someone higher and it'll just cause you pain and disappointment for your thought process to fixate on that. I know a lot of people that enjoy challenging themselves, but they find the challenge in playing the hand that they're dealt, not in whining that they weren't dealt a better hand.)


Your "honest question" contains several false assumptions, as you well know, Michael. Who do you think you're fooling?

I'm not sure who the "unluckier 90 percent" are, because a) X is much, much smaller than 1% of the company, and b) you're not laying out any other 90%-sized subset of the company with your other comments.

People inside X also have calibration and perf, of course.

Your ongoing ranting about "closed allocation" is more complicated; I think you make (elsewhere) an in-theory-mildly-interesting point, but you hugely overstate the significance. It's true that new googlers get relatively little control over allocation (though often not as little as you claim), but it's well-understood, even by you, that 12 to 24 months of good work at your starting post gets you a lot of flexibility. (Most knowledge workers, at least those past the first blush of youth, accept that a year or two of merely-awesome work are a fine stepping stone; we already know that you disagree and claim to want to change the world immediately, starting today.... how's that approach working for you?) If people don't want to work on something, they don't work on it.

Nearly everyone at Google works on challenging problems (well... I have no idea if the cafeteria staff think their work is challenging, so let's restrict this to engineering, sales, legal, and finance). Everyone at Google is building their careers. Can anyone doubt that?

Some of us are jealous of the people who work in X (I am!), most of us are not (I don't know why), but in what universe does it make sense for the existence of X to make me less happy than the non-existence of it?

----

More generally, most people do not believe that they are entitled to dictate their salary AND what they work on. And that goes double for people who don't yet have a proven track record of success.

I can pick what I want to work on, if I'm willing to accept whatever salary comes, including $0. (I've tried that, heh. What a surprise, no one wants to pay me for my opinions about board game design, nor my amateurish programming-language-design skills. I know, I know, I can't believe it either.) Alternately, I can make a great wage, working on something that's pretty awesome, but not exactly what I want to work on. Why would I be angry about that choice? If I want the best of both, obviously I'll have to work on making myself valuable enough that it makes sense for X to hire me.


"I have no idea if the cafeteria staff think their work is challenging"

The cafeteria staff works amazingly hard at their jobs and there're a bunch of challenges involved in feeding 30,000 employees well on a weekly budget. Go to Chef's Table at Jia (if you're in Mountain View) and ask about the job, or even just find the cafe's head chef (they're usually standing around the food looking proud of it) and strike up a conversation. They usually love to talk about their work.


There's your one. Now stay off until tomorrow.


In the heyday of Bell Labs and PARC, AT&T still had plenty of linemen and operators, and Xerox had toner salespeople and service technicians.


This analogy doesn't work because Google is extremely selective and hires for "cream of the crop" CS graduates who can pass an algorithms/data structures whiteboard coding gauntlet, as well as highly accomplished/experienced engineers who have made already made huge contributions at other companies or in open source.

When you hire people like that, you have to give them hard, interesting problems to work on, or they're going to be bored.


it's usually a mix of things.

first, while you may be drawn to that type of work, not everyone is. some people really enjoy the work they do in other parts of the company.

second, if you're drawn to it, you need to identify the path to it and the needed qualifications. if you don't make the cut, then you know why.

that keeps it from being an exclusive "clique" with no obvious merit-based way in, which helps keep the resentment others may feel (that you bring up) down.


second, if you're drawn to it, you need to identify the path to it and the needed qualifications. if you don't make the cut, then you know why.

That's a very Big Corp way of thinking. We don't believe in that. Nobody here starts off in the mailroom and works their way up to CEO in 30 years. You start off as a CEO at 21, say to hell with "experience" and "qualifications" — I'll just learn by doing and figure it out as we go along.

Not enough companies have a "learn by doing" on-boarding mentality. Nobody knows what they're doing without the chance to actually do it. (the worst programmers tend to be fresh CS graduates with no real world experience)


>You start off as a CEO at 21, say to hell with "experience" and "qualifications" — I'll just learn by doing and figure it out as we go along.

This reminds me that IMO hiring CEOs based on previous experience at the same job is fundamentally flawed because it creates a paradox.


Here's hoping the lab is NOT ONLINE. Because maybe it seems that China might possibly want to kind of hack it.


come on google glass in china ! so that the cizitens will photograph and video all the abuse that they are subjected to with the world in real time . i do not think so


they already have cellphones, you know :)


It's all going to be made in China eventually.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: