Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Things I've learned at Google so far

bentilly.blogspot.com

Post reply on HN
Print view View original

181–190 of 197 posts

Re: Things I've learned at Google so far

#181

Post by btilly »

elblanco wrote: post #169
btilly wrote: post #123

Earlier quoted context omitted.

Try to build something that deals with the amount of data that Google has, as quickly as Google handles it, with everything distributed, replicated, load balanced, and with failover to keep the site up when bad things happen anywhere in the system. (Which happens more often than we would like.) Before you're done you'll have built a lot of pieces of software that do different things that have complex little dependenc…

Quick, slightly off-topic question. There's a belief in the industry (I think wrongly) that pretty much the only way to build big complex systems, you need a central "Enterprise Architect" to model all the business processes and use those to drive down to the technology. As somebody external to Google, it seems to me that this approach largely does not exist at Google (or if it does, it exists only for particular iso…

Excellent question.

I don't think I'm supposed to answer it.

However Google's answer would be part of the, "nobody would believe me" piece of neilk's top-rated comment.

Re: Things I've learned at Google so far

#182

Post by btilly »

neilk wrote: post #156
btilly wrote: post #63

Earlier quoted context omitted.

#1 sounds very familiar to me. :-) I hope I never reach #3. That sounds like I'd become a kind of person I don't like being.

Really? For me it took about a week. It's just a matter of psychology. You still have friends outside the company... but you can't talk about what you're doing. You can't even talk about the perks without becoming deeply annoying. You can't talk about the scale without giving away competitive information, or sounding like you're bragging. So at some level, when it comes to your work life, you start putting up a mask…

I can understand that.

However the part that I'm objecting to is By the time you've reached (3) you've come to believe that Google is years ahead of anyone else; you've transferred all loyalties to the company; you accept the idea that you're in an elite of humanity. I have a lot of loyalties outside of Google, and recognize a lot of talent that isn't there. (Some of whom I'd like to get there, but that is another story.) There are also a lot of kinds of valuable talent which I don't think are even particularly well represented at Google.

However I'm really enjoying it.

Re: Things I've learned at Google so far

#183

Post by _delirium »

nostrademons wrote: post #122
waterlesscloud wrote: post #97

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.

What are some of the tasks you've had a bad experience with? I work in Google Search, and we actively try to identify areas that people are having difficulty with and then come up with ways to make their search experience better. Unfortunately, most of the low-hanging fruit has already been picked. If we knew precisely where the pain points were, we'd be much better able to solve them.

One big annoyance: 10 years ago, when I searched for a snippet of song lyrics, I got a fan site for the band that accurately transcribed the lyrics. Now I get nothing but pages and pages of shady sites selling me ringtones, with inaccurate lyrics, even when I know there are fansites for the band with better lyrics.

A guess is that this is the result of the reliance on domain names for ranking. The high-quality lyrics are distributed across millions of separate domain names, since fansites typically are dedicated to only one band. The low-quality, ring-tones-heavy lyrics sites, though, are centralized, with every band's lyrics on the same domain.

Re: Things I've learned at Google so far

#184

Post by _delirium »

dschobel wrote: post #47
hello_moto wrote: post #40

Earlier quoted context omitted.

Oh... I was referring to the fact that the author just completed his first month at Google... It's hard to judge a company just in one month especially when some people might spend their first week (or two) doing company orientation.

Agreed. I don't doubt that it's an amazing place to work, but every job has its honeymoon phase and the author is definitely in that. I'd be eager to hear an honest assessment from him after a year though. There has to be a reason why brilliant and motivated people leave Google, after all.

The people I know who've left have left mainly due to the secrecy this article also talks about: they developed cool algorithms they wanted to publish research papers about, and weren't allowed to.

Re: Things I've learned at Google so far

#185

Post by kaj_sotala »

sgk284 wrote: post #108
dschobel wrote: post #102

Earlier quoted context omitted.

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 Li…

Do policy issues such as "how to deal with China" get discussed in the TGIFs or e-mail lists? I would imagine that such discussions might become very long-winded and very prone to flame wars very quickly.

Re: Things I've learned at Google so far

#186

Post by waterlesscloud »

_delirium wrote: post #183
nostrademons wrote: post #122

Earlier quoted context omitted.

What are some of the tasks you've had a bad experience with? I work in Google Search, and we actively try to identify areas that people are having difficulty with and then come up with ways to make their search experience better. Unfortunately, most of the low-hanging fruit has already been picked. If we knew precisely where the pain points were, we'd be much better able to solve them.

One big annoyance: 10 years ago, when I searched for a snippet of song lyrics, I got a fan site for the band that accurately transcribed the lyrics. Now I get nothing but pages and pages of shady sites selling me ringtones, with inaccurate lyrics, even when I know there are fansites for the band with better lyrics. A guess is that this is the result of the reliance on domain names for ranking. The high-quality lyrics…

Good point, and probably symptomatic of a wider problem. Sites with a ton of "good enough" information will outrank sites with a very specific set of "excellent" information. The high ranking of wikipedia and the various answers sites is a similar effect.

Re: Things I've learned at Google so far

#187

Post by EricBurnett »

kaj_sotala wrote: post #185
sgk284 wrote: post #108

Earlier quoted context omitted.

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 Li…

Do policy issues such as "how to deal with China" get discussed in the TGIFs or e-mail lists? I would imagine that such discussions might become very long-winded and very prone to flame wars very quickly.

Yes, they certainly do. In the TGIF presentation itself there isn't too much back-and-forth (although the panel members may have different opinions), but people will discuss it themselves after. For something as big as the China issue, it was probably discussed in some other channel(s) as well.

As far as email lists go, the discussions can get quite lively, although I wouldn't generally call them flame wars - civility still holds in general. But they can definitely get long; the term centithread gets thrown around a lot. When I was there one of the longest I saw was about people putting open chocolate milk back in the fridge, and whether or not that was a terrible thing to do :p..

Re: Things I've learned at Google so far

#188

Post by elblanco »

btilly wrote: post #180
elblanco wrote: post #174

Earlier quoted context omitted.

I think one thing that people definitely forget is that many of those were startups at one time. Lots of the peripheral technology offerings Google has were not developed originally in-house -- they were acquired with ad-revenue from their flagship. And as I posted in another comment ( http://news.ycombinator.com/item?id=1093455 ), outside of Apps for business, really don't make any money for anybody. A quick analysi…

You're missing something obvious. When Google buys something they tend to put ads on it and let people use it for free. That means that any revenue that web property brings in is counted as ad revenue, and is not counted as coming from that web property. Therefore looking at the SEC statement gives you no way to figure out how much revenue Google is or is not getting from a lot of their acquisitions.

That's true I suppose. But it still means that any products and services Google actually sells still accounts for about $200mil/yr and about $100mil/yr of that is purely Google Apps for businesses.

That means,

Google Earth Enterprise Geospatial Servers + Google Earth Pro + Google Products/Checkout (minus advertising revenue from that) + Google Appliances + Google Sketchup Pro + More Picassa Space + Google Voice long distance service fees + whatever else all account for $100mil/yr in revenue. Those aren't ad supported services and nearly all of those were acquisitions.

Google Appliances should be a multi-billion dollar/yr business, supplanting everything from document and content management systems, to Sharepoint, to many types of traditional Oracle installations, yet we know that it's entire business sits at under $100mil/yr. Let's be generous, and in a flight of fancy say it sits at $90mil/yr, that means that all the rest of those products would have to do less than $10mil/yr.

Too much you say! Google voice Long Distance has to do something in the neighborhood of $5mil/yr! And I know personally of at least a dozen or so Google Earth Enterprise Server Installs, at ~$20k a pop (or whatever the pricing is), that's not a whole lot of money. ESRI has a similar business at just under $.75 billion/yr. Assuming there are quite a few more installs than I personally know about, let's say all added up, $2mil/yr in geospatial products (that's about 10x the value of what I've personally seen so I feel I'm being pretty generous here). So the Appliance business couldn't possibly be that "big".

In other words, no matter how you mix the numbers, the performance of these offerings is not very good. I'm currently working in a startup, and we're doing numbers like this and we're considered very under-performing by our investors right now. It would rankle me in all kinds of ways were I in google management that these products are not each at least $20-30mil/yr interests and one or two of them should be $billion+/yr interests.

Again, don't get me wrong, I love Google and their products and services. I probably use most of what I just listed here at least 2-3x a week. But Google is an advertising agency that uses a novel distribution channel for it's advertising clients. It's not really much of a products company in the same way Microsoft of Apple is. Google should be analyzed more in terms of comparison to other ad channels and agencies not in terms of direct competition with technology vendors. This is the flaw analysts, users, employees and competitors make.

Re: Things I've learned at Google so far

#189

Post by jlees »

gcheong wrote: post #134
jlees wrote: post #113

Earlier quoted context omitted.

Aw man, I wish I'd joined as an engineer now. The lens I'm seeing the company through (product management) isn't half so cool. I mean, a lot of the above is still true, but we definitely have deadlines, tasks, responsibilities, and the team doesn't seem to eat together all that much. Obviously teams vary though, and this one has been thrown together fairly recently from a very disparate set of people, so that cohesiv…

Can't you switch roles?

Not that easily actually. You have to re-interview - and I don't actually want to be in a coding role, so I doubt engineering would suit me ;)

Re: Things I've learned at Google so far

#190

Post by EricBurnett »

elblanco wrote: post #57
johnyzee wrote: post #8

I am blown away by some of the stuff Google puts out, I can only imagine what their internal technologies and tools must look like. To me, that more than anything validates their approach to hiring, development and business strategy, of which I would otherwise be very sceptical. I'd never get hired at Google, and rightly so; I look at some of their stuff and know that I could never do that. I used to think (not witho…

I think that you can. To date, Google really only has one successful product (success meaning "makes money for Google") and that's Ad-sense. Sure it's wildly successful. But if that stopped making money for Google tomorrow (for example, I automatically mentally filter out all advertisements in my search results, how long before everybody does that), what would their revenue and profitability picture look like? In alm…

A lot of that is because Google doesn't define success that way. Their view is that anything that stimulates the internet will indirectly benefit them, so that makes products like gmail or chrome clear winners. Gmail brought us threaded emailing, ajax (and good javascript-supported interfaces in general), etc. Chrome brought us tear off tabs, fast javascript execution, etc. I think if the focus was stand alone profit generators the products they put out would be very different - what would Android look like?

*brought us = made popular, not necessarily invented.

Post reply on HN