14 ms·

Warning: Personal opinion ahead To understand why this keeps happening, you need to understand the product and engineering culture at Google. As a group, Googl

by jfasi 8y ago

Warning: Personal opinion ahead To understand why this keeps happening, you need to understand the product and engineering culture at Google. As a group, Google engineers and PMs are obsessed with promotion. At the heart of every conversation about system design or product proposal lies an unspoken (and sometimes spoken) question: will working on this get me promoted? The criteria for promotion at Google, especially at the higher levels like SWE III -> Senior and especially at Senior -> Staff and above, explicitly talk about impact on the organization and the business. This has consequences for the kind of teams people try to join and kind of work they choose to do. Maintenance engineering is so not-rewarded that it's become an inside joke. Any team that isn't launching products starts bleeding staff, any project that isn't going to make a big splash is going to be neglected, and any design that doesn't "demonstrate technical complexity" will be either rejected or trumped up. This is as important in the product management, people management, and general leadership roles as in engineering. The incentive throughout is to create a product, launch it, apply for promotion, and move on to bigger and better things as soon as possible. In my time at Google I saw organization after organization pay lip service to rewarding maintenance and "preferring landings over launches" and “improving product excellence” but (at least in my experience) nothing stuck. Usually an organization starts with a top-down direction and the rest of the company is compensated for executing it. Not at Google. The "let a thousand flowers bloom" approach that developed from the early days of twenty percent time and total engineering independence has created a disorganized mess of a company. Multiply the individual incentives fifty thousand times and you get a company that throws stuff at the wall to see if it sticks, and if it doesn't kills it immediately. Edit/Addendum: This is also why GMail, YouTube, Search, GCP, Android, and others aren’t going anywhere. They’re making money, they’re core to the business, and there’s plenty of opportunity to work on them and get promoted. They all also share one thing in common: deep down they’re frontends for search or advertising (GCP and Apps are an exception because they make money on their own). Measuring and proving impact on search numbers is a well-known promo narrative at Google, so those products are a safe bet for employees and users. Streaming game services, not so much.

mrbonner 8y ago
It is general true for any big organization. In my own firm, I could see the rush into machine learning direction as a door way to a promotion. Any engineer or PM who works on a ML related project and is able to ship it would have a higher chance of getting promoted. I could also say the same for the usage of AWS products. In the end, I often see projects resembling the messy mashup of AWS products to form a unnecessary complex ML solution to a mosquito problem. It’s a waste yes, but sure will make it looks challenging and boost people to promotions.
- asark 8y ago
  I increasingly see this as the next level of alienation from work-product. Carpenter makes and sells chairs to people who need chairs, farmer grows and sells apples to people who want apples. Then there's the line worker making just chair legs for people they'll never meet, and the farmer selling their entire apple crop to a big company that'll do who-knows-what with them—sell them, turn them into apple sauce, use them as input to some process that turns out a "natural ingredient" that's some pure, extracted chemical contained in the apples. Then there's the sort of crap we're talking about here. Projects that obviously never should have existed in the first place, and/or that would have been better if done in a simpler, faster, cheaper way, but a ton of make-work was thrown in due to agency problems and perverse incentives. How much of the benefit of computers is being eaten by this kind of thing? The most effective things we do with them may provide 1000x the economic output of that apple farmer, or more, but then we spend a bunch of time doing useless or negative-value stuff, without even getting into a lot of the actually-lucrative work having large negative externalities—spyware-enabled advertising, say, or even just advertising, full stop. What does it do to someone's psyche to spend 90+% of their time doing harmful or useless stuff? Hell some people end up doing nothing of actual value for years on end. What are we doing?
ryandvm 8y ago
I totally agree that this is frustrating for erudite tech users, but I'd say calling Google a "disorganized mess of a company" is kinda stretching things. The last quarterly earnings for GOOG were up 21% year/year. That's pretty insane for a company of this size. The way they operate is annoying when it's one of your favorite products that isn't pulling it's weight, but it's hard to argue that the strategy isn't sound.
- noneeeed 8y ago
  To play devil's advocate, we don't know how sucessful they would be if they followed a more orderly model. Not saying they wouldn't be, but it's entirely possible that they are successfull in spite of, not because of how they operate in much of the business. Given that the vast majority of their revenue comes from just one place, advertising, which seems to be run in a much more cautious manner, there is an argument that this might be the case.
- 1024core 8y ago
  They keep showing more and more (or bigger and bigger) ads. At some point, they'll hit the law of diminishing returns.
- jfasi 8y ago
  I'm disappointed I had to go this far down into the comments to read an actual sensible rebuttal. I stand by my calling it a disorganized mess, but you'll notice I never actually say this strategy isn't paying off for the company...
- TallGuyShort 8y ago
  Agree that "disorganized mess" is a stretch. But pointing to a company's success doesn't make them immune to criticism either. My project often gets compared to a competing offering from Amazon. When discussing this with coworkers, I will point out the drawbacks and trade-offs of their design and why that's a WAY bigger deal for my company's customers than it is for the typical AWS user. "But they made $X last year". Yes. And we're less vulnerable to data loss. Thank you. Next.
throw_away 8y ago
> you get a company that throws stuff at the wall to see if it sticks, and if it doesn't kills it immediately. I would love if they followed this strategy. Instead, you have products like reader and inbox with many passionate users, which demonstrate some wall stickiness, and they're killed rather than monetized. Monetize me. I recognize that the things I like cost money. Figure out how to make money off me! I would have accepted ads in reader or inbox. Hell, I would have paid dollars to make those ads go away even.
- michaelbuckbee 8y ago
  Google reader is the one that kills me. Not only do they not monetize it / promote it, but it's Google subsidized free existence killed the market for alternatives. Then Google+ and their failed social efforts killed Reader and RSS feeds have now disappeared from the landscape leaving us with only proprietary options.
  - jamesgeck0 8y ago
    > it's Google subsidized free existence killed the market for alternatives Did it? IIRC, there were at least six decent alternatives available when Reader closed. More if you count desktop and mobile applications.
    - s0rce 8y ago
      I tested a few that appeared from the ashes. Ended up not liking most of them but eventually found bazqux and pay them money.
      - CobrastanJorji 8y ago
        I switched to Feedly the day they announced the shutdown because it could import from Reader, looked like Reader, and used a lot of the same shortcuts. Is bazqux better?
        8y ago
noneeeed 8y ago
Is this basically why things like Inbox, and the endless attempts at messaging, all come out as seperate products rather than enhancements to existing products, because that way you get to say you launched something new?
- jfasi 8y ago
  To be perfectly honest I don’t know the answer to the chat app case in particular. That’s possible, and if one were to investigate and find each app was made by a team rolling up into a different director then I’d say that’s what happened. However it’s entirely possible that it was caused by an overly complex market segmentation strategy. WhatsApp had the advantage of growing slowly and segmenting their users with time to experiment and tweak things. The Google chat apps came out at once, and it’s possible there were so many because they tried to come up with a working segmentation for all addressable users from day one instead of growing organically and grabbing each segment one at a time.
- nevir 8y ago
  IMO yes: the promotion culture (technical complexity, impact) push people to come up with Big Ideas. The thing with Big Ideas, is that they're often iterations of existing ideas, but just different enough that you can make a case for a fresh start. (And who doesn't want a greenfield project) Plus, the cost of experimentation is far less while you're not supporting a massive userbase.
- tdeck 8y ago
  That's consistent with my own experience at Google. It's not that working on improving a thing isn't rewarded, but it's just much easier to demonstrate impact when you can use the word "launched" in your self-assessment Launching is in itself an accomplishment, while optimizing requires you to also collect data, do analysis on the data, and show a positive impact. "Impact" here is one of the buzzword categories that goes into an assessment of your performance as an employee.
PopePompus 8y ago
It may be that GCP isn't going anywhere, but when our organization was choosing a cloud service, we chose Amazon over Google specifically because of Google's flakiness. Google could drop cloud services tomorrow, and it would not be out of character.
- linsomniac 8y ago
  We are in the same boat. My boss came to me last week saying he had lunch with a friend that just moved their services all to GCP and got big savings. But we just recently were impacted by the sudden, unexpected Maps 14x price increase.
  - saurik 8y ago
    Similar drastic changes happened to AppEngine, where Google let it rot for something like a year with increasing, extremely bad, latencies and then a sudden very large pricing model shift that made it much more expensive. I was an early adopter of AppEngine, and just can't imagine ever screwing around with Google cloud technology ever again.
  - Moru 8y ago
    We got stuck there too, moving away everything to alternatives. Just analytics left to remove now.
- jfasi 8y ago
  I can’t share with you how much GCP is making in revenue, but I can assure you it’s enough that it won’t get dropped, ever. I’ll put my money where my mouth is: I’m now preparing to expand my own team’s operations into GCP and away from AWS because of how good their ML offerings are.
  - jdgoesmarching 8y ago
    Maybe the core product won't go away, but you could still see the damage from this approach in features or services you depend on within GCP. Not to mention Google's love for mixing around interfaces every five seconds (with poor documentation of course). Consistency and reliability matter more than flashy new tech. I wish the culture at Google would understand that.
TheTruth1234 8y ago
Thanks, but to you and any other GOOGLE EMPLOYEES Are you not internally aware of this? Do you just carry on regardless? What happens when you voice this issue internally? As an ex-Google fan boy, the title of this article has been true for years and it's what poisoned me (and no doubt many others) against you. So you must be aware of it. What's your take? I'm super curious as to why you gave up tech (brand) leadership so seemingly nonchalantly.
- geofft 8y ago
  Google employees in an individual-contributor role have limited power to change this if they want to remain Google employees. They're subject to performance reviews on this system, and the expected value of any individual advocating for a change is so tiny, compared to the expected value of playing along with the system and getting their salary and bonus check. If you want this culture to change, you need the people signing the checks to feel compelled to change it. The practical effect is likely to be people leaving Google because they can get rewarded for maintaining existing products elsewhere.
- steven2012 8y ago
  Google as a company is worth almost $1T. It's hard to argue that's the wrong approach when they are so ridiculously successful.
  - falsedan 8y ago
    Amazon and Apple have similar market caps and yet don't have this product discombobulation
  - scarface74 8y ago
    How well is GCP doing compared to AWS and Azure? How well is GSuite doing compared to MS Office? How well is Android doing as far as revenue and profit compared to iOS? According to information that came out during the Oracle trial, Android has made Google less over 10 years than iOS makes Apple during a down quarter. How well is G+ doing against FB? Google hasn’t had but one successful revenue generating product - ads.
s3r3nity 8y ago
I totally understand this - but it doesn't mean I like it. I could get through having two podcast apps on my Android (Google Podcasts, Google Play Music) that give me notifications for my podcasts, with no plans on phasing out one or the other. I could get over having two different music apps - Google Play Music and YouTube Music - because it seemed like there might actually be a plan here - albeit we haven't heard anything in ages. I could get over having two email apps - Inbox and Gmail - because I knew Inbox was just an experiment that was going to be a tease of cool features that may / may not be ported over...therefore I tried not to take the bait. I could get over having a bunch of different messaging apps - I didn't like it, but 95% of my messages are SMS texts anyway, so it didn't impact me. The FINAL straw was falling in love with a PixelBook that I bought during the Black Friday sales, as I moved into a role in an org that was 1) deeply invested in G Suite and 2) allowed me to do all my work from the browser - making the Chrome OS concept finally fit for me. THEN Google closes that part of their hardware division. Now I just have angst over _any_ Google product / service that isn't the core that I use and "know" knock on wood won't go away (Search / Gmail / Chrome). Android wear? Google Home? Hell, I was seriously finally moving to a Pixel 3, as the tight integration with some Chrome OS features on my PixelBook would've been ideal - but I can't even trust that the Pixel phone line will last that much longer.
- Wowfunhappy 8y ago
  > The FINAL straw was falling in love with a PixelBook that I bought during the Black Friday sales, as I moved into a role in an org that was 1) deeply invested in G Suite and 2) allowed me to do all my work from the browser - making the Chrome OS concept finally fit for me. THEN Google closes that part of their hardware division. The PixelBook isn't going to stop getting security updates any time soon, so if it was a good machine before the division was closed, why is it a less good machine now?
  - sooheon 8y ago
    What do you think the half life of most electronics is? Why do you think so many people get anxious about Apple "abandoning" the Mac? A trickle of security updates does not make a well supported and thriving product line.
geofft 8y ago
This went into my decision not to accept a Google offer. AMP is a product that should not exist in the first place. Everything AMP does could be done in other ways—having Google Search measure load time, improving load times, working on webpackages as step 1 instead of an afterthought, etc. But because AMP is a coherent, well-defined product, and Google rewards launching it over accomplishing the same goals in less-obvious ways, the person behind AMP was incentivized to go that route for the sake of his own career at Google, and ruined the internet for the rest of us. I didn't want to be subject to those engineering standards.
- jfasi 8y ago
  Ehh, I disagree with your assessment, and to illustrate why I think you need to focus on asblockers. AMP serves a critical strategic purpose: it severely disincentivizes crappy, JS-heavy websites on search results and as a result makes people less likely to install adblockers. Google is terrified of adblockers going mainstream, to the point where they’re willing to risk anticompetitive suits by creating their own. AMP is a part of the effort to prevent that from happening.
  - geofft 8y ago
    That's an interesting viewpoint and makes a lot of sense, but still - why couldn't Google have accomplished this goal through web search policies? Have the crawler say that if it sees JavaScript with certain behavior it will be de-prioritized compared to a similar website with less annoying JavaScript. Which, to some extent, you get just by focusing on load time: if the crawler emulates a normal browser and measures the time until everything has rendered, it'd be able to penalize annoying JS ads that block loading. Or if it emulates a normal browser and indexes text rendered on the screen, it'd be able to penalize sites that display a pop-over ad that you have to interact with. (Would that have been more prone to an antitrust lawsuit than AMP?)
    - jfasi 8y ago
      I don’t want to speculate too much on the legals of the situation, so I won’t talk about anticompetitive matters. As for why not do this in indexing, of course they’re doing that. They’ve been doing that for a very long time, but that approach can only go so far. There’s a constant cat and mouse game between websites and search providers, and AMP is a way to end it once and for all. It’s important to stress that the motivation here is largely user-focused. People tend to ascribe to Google these shady Oracle-like motivations around screwing users and developers, but the truth is that moves like this are largely motivated by a desire to improve the experience. Not out of any sort of altruism, mind you, but because what’s good for the user is generally good for Google.
dna_polymerase 8y ago
I think Google doesn't care for most services as they do not depend on any of those. Sure Android, GMail, and YouTube will stay, because they serve their core business: ads. Everything else could be over at any time. This is supported by their recent push for China. Suddenly employee voices aren't worth anything anymore (remember the drone story, that was over after employees pushed back). Maybe they have realised by now that diversification isn't that bad after all and will continue to push GCP but I wouldn't want to bet on GCP's life-span.
- RestlessMind 8y ago
  I am willing to bet on GCP's life-span - it will be around for 10 more years at least. What would you like to bet for?
  - iainmerrick 8y ago
    GCP in its current form, though? In 10 years, will you have to choose between the deprecated version and the version that doesn't work yet?
snarf21 8y ago
Well said, you always get what you optimize for.
chibg10 8y ago
In my experience at a different FANG, the same culture exists here. If I had to guess, I would say it probably applies to the others as well.
- jonathankoren 8y ago
  Oh just go ahead an name Facebook. The irony, is that this pattern at FB and and Big G isn’t indicative of a good engineering culture. It’s pathological engineering.
- jedberg 8y ago
  It definitely doesn't apply to Netflix. You can get good raises and promotions by optimizing an existing process. If you can show improvement in internal metrics, such as improving latency or lowering cost, that's a huge win. Even if you increase costs you can get a raise and promotion if you show that it increased code quality, such as creating the Chaos Monkey/Gorilla/Kong.
  - anoncake 8y ago
    I guess it helps that Netflix essentially has one product.
    - jedberg 8y ago
      Yes that probably helps a lot. Also they don’t hire fresh college grads, so you generally have less of the “wow this was clearly written by a junior engineer” type of maintenance. The low hanging fruit is mostly picked when it comes to architecture and optimization.
- phjordon 8y ago
  I work at Amazon and I don't believe that applies here. There certainly is an expectation that you can coordinate increasingly broad efforts as you rise through the ranks, but the focus isn't on launching new products (in my experience). Impact can be anything that is positive and measurable. At the end of the day if you make the company $1million then it doesn't really matter where you do it.
iagovar 8y ago
Cool for them, but that hurts my wallet. I never felt safe using google products, and the only one I paid for (adwords) decreased in quality, so...
amyjess 8y ago
> Usually an organization starts with a top-down direction and the rest of the company is compensated for executing it. Not at Google. The "let a thousand flowers bloom" approach that developed from the early days of twenty percent time and total engineering independence has created a disorganized mess of a company. Multiply the individual incentives fifty thousand times and you get a company that throws stuff at the wall to see if it sticks, and if it doesn't kills it immediately. This is also why we had a botched rollout of Material Design across Google's apps following the release of Lollipop. Apps rolled out Material redesigns whenever, often months and sometimes even years after Lollipop's launch. Apps used various different versions of the Material support library: the Drive Apps in particular had text and widget alignment issues that made them look very unlike the other Material apps. And for a while, I was tracking which apps exhibited correct behavior of the status bar and the hamburger menu, because there were multiple ways apps could implement each one, but only one correct way (ironically enough, I was doing so in a Google Doc) [0]. To me, this is an awful way to roll out your new company-wide branding initiative. You want a new brand to be consistent across all your products and rolled out simultaneously. In retrospect, it was the moment when I began to lose faith in Google as a company. Some time after that, somebody explained to me why, and it was exactly as you said: engineering groups operate independently of each other with no top-down chain of command. There's nobody telling them "you have to roll out the new branding by this date", "you have to make sure you're following these company-wide standards", etc. Basically, I was told that at Google, "company-wide" isn't a thing that exists. And that's awful. And honestly, it makes me never want to work for Google. That kind of chaos is exactly the opposite of the kind of working environment I could thrive in. [0] Correct behavior: status bar is a darker shade of the toolbar's color, the hamburger menu covers the screen from top to bottom with the status bar as a translucent overlay on top of the menu bar, and the hamburger icon doesn't animate at all while you drag the menu out. Following Lollipop, there were Material apps that broke every one of those rules, but in completely different combinations from app to app.
dalbasal 8y ago
Corporate political economy (ecology? anthropology?) is always pathological like this, one way or another. Seems inevitable, like factionalism and partisanship taking over parliamentary politics in every example we have. That aside, it kind of makes sense that Google places this much emphasis on the revenue producing parts of the business. On "what about new products and innovation" see the first point. While every corporate pathology is different (Google's flavour seems unique) a near-constant is doublespeak. There might be (or might have been) a logic behind it, it just gets communicated in corporate-speak because plain language is too harsh. Ultimately, at Google's scale (like a VC) a meaningful innovation is >$1bn annual revenue or the equivalent of that. Not many projects have that kind of potential. Letting a thousand flowers bloom means letting 998.3 flowers die.
olalonde 8y ago
> Multiply the individual incentives fifty thousand times and you get a company that throws stuff at the wall to see if it sticks, and if it doesn't kills it immediately. It sucks for some users but isn't it good for Google as a business? The way you describe it, it sounds like Google is internally approximating a free market to allocate resources. I understand the risk of damaging the brand but for a lot of new Google products, the only competition comes from startups which equally are a risky bet.
gotocake 8y ago
I agree with most of what you’ve said, but I’m curious about whether or not YouTube actually makes money. I was under the impression that on a good day it just about breaks even, but has that changed?
jarjoura 8y ago
Facebook operates similarly. They created a culture of working on things that interest you as long as it can show meaningful impact. If you can show maintenance of something impactful, then you can work on it. It's really no different in the startup world. If your startup can't show growth in users or profits, then you shutdown.
bradstewart 8y ago
Any thoughts on whether Google Photos will stick around? I find it's far better than any alternatives I've come across, and the recent Inbox shutdown made me realize how "dependent" I am on it.
- rock_hard 8y ago
  That’s gonna be an interesting one to watch. Hosting the worlds photos costs many billions of dollars each year! That said it’s a key driver to sell android phones (“infinite storage”)...so for the time being you will be good.
  - scarface74 8y ago
    But then again according to information that came out during the Oracle lawsuit, they only made $36Billion from Android during its entire existence.
  - dillondoyle 8y ago
    Just googled around and I'm guessing the terms allow google to use the photos and associated metadata which sounds to me like it would be hugely valuable. For ML research, search, advertising, etc. If Google knows where my photos were taken, what products are in the photo, what car, etc.
    - rock_hard 8y ago
      Yeah, people keep saying that...IMO its huge and risky bet to hope for some future way to leverage the meta data for ML
president 8y ago
This mindset is commonplace in most of the jobs I've held in the last 10 years. Personal gain trumps morals, company values, and even company performance. It's what happens when you put a bunch of high-performers in a box and let them play the corporate ladder game.
- jimmy1 8y ago
  It's better than the situation at most companies I have been at: It's basically the personal gain trumps all, but without the actual results. You just have a bunch of populists that talk up a good game, but when it comes time to execute or actually launch something, they are the last person you really want at the helm, but because they have built up a reputation with the higher ups, they are impossible to displace. From my perspective, this appears to be the situation of most "Health IT" corps and VP of Engineering and above. I've since "sold out" and moved to a digital marketing firm, but in many ways the work here is much more honest than what happened at places I worked at in the name of "health"
- tdeck 8y ago
  I think that's a bit of an unfair way to characterize what's going on here. In fact, it's Google's values (as expressed through internal incentives) that govern this process. If the company wanted people to, say, make X better rather than killing it and launching a completely new X, all it would have to do is reward that rather than setting up a system that punishes it through reduced career growth.
mshockwave 8y ago
> Maintenance engineering is so not-rewarded that it's become an inside joke highlighted
Animats 8y ago
Streaming game services, not so much. Yes. The game industry is very reluctant to use Improbable's Spatial OS, because, through a deal with Google, they force you to host your game, expensively, on Google's servers.[1] Nobody wants to spend $100 million developing an AAA title using that system, and then suddenly get an email that Google is discontinuing the product. (Spatial OS is technically interesting, and, hopefully, someone else will do something like it, with reasonable licensing terms.) [1] https://massivelyop.com/2018/01/12/chronicles-of-elyria-is-dropping-virtual-world-platform-spatialos/ https://massivelyop.com/2018/01/12/chronicles-of-elyria-is-d...
- ernopp 8y ago
  > then suddenly get an email that Google is discontinuing the product. Google is unlikely to discontinue GCP
  - sgift 8y ago
    The list of discontinued things that people said that about is getting longer by the day. That's kind of the point here.
- user5994461 8y ago
  Game servers typically don't last more than a few years, even for an AAA game. They pretty much never last a whole decade outside of major MMORPG. It's a bit surprising that the game industry would be scared of GCP going bust, given the short release cycle of the industry.
- ocdtrekkie 8y ago
  Didn't know SpatialOS was attached specifically to Google Cloud. This seems like a risky business move, people should've learned from the fine example of companies built exclusively on the Facebook platform by now. https://www.hadean.com/ https://www.hadean.com/ is presumably a direct competitor to SpatialOS then, I take it? And it's tied to Azure, I believe.
  - Animats 8y ago
stcredzero 8y ago
GMail, YouTube, Search, GCP, Android, and others...deep down they’re frontends for search or advertising... Streaming game services, not so much. Streaming games has tremendous potential synergy with advertising. Games are yet another form of media, after all. There is obvious tremendous potential synergy with YouTube. Also, if one can stream games, then why not stream other kinds of applications? There's tremendous opportunities for synergies there as well.
- dageshi 8y ago
  Basically identical to youtube. Forget putting AAA games on it, imagine a massively improved version of the flash games that were popular a decade ago.
  - stcredzero 8y ago
    Forget putting AAA games on it, imagine a massively improved version of the flash games that were popular a decade ago. I wouldn't be so confident about AAA games being a barrier. I guess AAA games emphasizing "realism" might always have a high-end PC niche, since the genre could always ramp up graphics to the point of un-streamability. I could also totally imagine, let's say, Nintendo, going whole-hog into streaming games with cartoony graphics, emphasizing multiplayer with friends. I don't see any reason why the industry couldn't target and achieve something like 30 ms round trip latency. So long as the gameplay is good, the potential audience for 1080P gaming with even last generation level graphics is huge. To think they would be limited to the level of flash games is hugely naive.
    - dageshi 8y ago
      I guess I was thinking more in terms of what's going to happen when you can literally play a game with the click of a button in a browser like you could in flash. My thought is there will be massive amounts of games flooding the market in much the same ways there are on mobile devices and the price will trend towards zero with microtransactions in the same fashion.
josteink 8y ago
> why this keeps happening, you need to understand the product and engineering culture at Google. As a group, Google engineers and PMs are obsessed with promotion. For people outside the San Francisco and Silicon Valley bubble which are confused by this, it may be worth noting that promotion here does not need mean getting a new job. Everyone are not constantly getting new jobs. “Promotion” here means internal, magic Google-points and seem to work much like Chinese social credit. Source: answer by Googler here on HN when I was confused about this very same thing. https://news.ycombinator.com/item?id=18575875 https://news.ycombinator.com/item?id=18575875
- QuercusMax 8y ago
  Promotion means you get promoted to a higher level within Google[1]. This typically means you get more money[2], better opportunities, and more visibility. People often do change jobs within Google/Alphabet after getting promoted. It's a signal that you're doing a good job, as verified by a promotion committee, and hiring managers will often look at a recent promotion as a signal that someone is a high-performer. But there are lots of folks who don't change jobs after a promotion (especially if you're happy with what you're doing) - it's an opportunity to keep doing what you're doing, just more of the same, with official recognition that you've been doing a good job. 1. Level 3 = SWE I - New Grad (Bachelors/Masters). Level 4 = SWE II (Newly minted PhDs start here). Level 5 = Senior SWE. Level 6 = Staff SWE. Level 7 = Senior Staff SWE. Level 8 = Principal SWE / Director. 2. This is not strictly true; a high-performer at Level N may make just as much as a low- or average-performer at Level N+1. But in practice you start to hit salary caps, and sustaining very high performance at Level N may be more difficult than sustaining average performance at Level N+1.
rifung 8y ago
Indeed that is also what I thought when reading the very relevant article from yesterday: https://news.ycombinator.com/item?id=19541475 https://news.ycombinator.com/item?id=19541475
neilv 8y ago
How common are career ambition incentives like this, among engineers (not middle management) right now? It sounds like, in Google's case, it's very intentional. Other than the job-hopping that I started hearing about when dotcom IPOs started, engineers, from what I (perhaps naively) understood of my anecdotal experience, seemed to mostly just like to do engineering, and/or also had a sense of obligation to their project/team/duty. About the worst I recall hearing about would be an engineer picking a language/tool because they wanted to learn it or add it to their resume, rather than because it made the most sense for corporate goals for the project.
- iainmerrick 8y ago
  Not necessarily common, but the ambitious engineers are the ones who get promoted and they're the ones who get to set the technical direction. So the incentives on offer to (over-)ambitious people are very important.
korethr 8y ago
>This is also why GMail, YouTube, Search, GCP, Android, and others aren’t going anywhere. They’re making money, they’re core to the business, and there’s plenty of opportunity to work on them and get promoted. They all also share one thing in common: deep down they’re frontends for search or advertising (GCP and Apps are an exception because they make money on their own). Measuring and proving impact on search numbers is a well-known promo narrative at Google, so those products are a safe bet for employees and users. Streaming game services, not so much. This, right here is the key, and what I think most people miss, including the original article. Hell, I missed it myself for a long time: Google is not a platform company as the original article asserts. They are an advertising company. That is Google's bread and butter, and has been since day one. The products and platforms that stick around do so because they are or become successful funnels or front ends for advertising. As much as the tech culture of the Internet loves to bitch about advertising (myself included), unto itself, it is not a solely an evil thing. Oh, it absolutely can be turned to evil ends; see any of a myriad complaints about Facebook. However, I think the pairing of ads with search has been damn useful in a way that balances with the potential for evil. As an example, when I'm searching for specific lubricating grease for my bicycle's bearings, I very much appreciate seeing an ad for a vendor who's willing to sell me a single tube thereof, instead of having only the web search listings of vendors who require a minimum order size of 1 shipping pallet. Google did right paring advertising with search. To use a 'corporate bullshit' word, Search and ads synergize well. The froth of Google's other products and platforms makes more sense when looked at in that light. A company that sits still will eventually get passed by someone else who correctly sensed changes in the market and business environment, and thus eventually die. So of course Google is going to be trying to come up with all manner of things that leverage the Internet and their massive infrastructure; one of those may well be the next big thing that funnels more money in to advertising. No small part of this is that Google employs a great many nerds like you and I, people who have no trouble inventing cool new things you can do on the Internet. But sadly, a great many of those cool things don't fit well with Google's advertising-based business model, as evidenced by the constant product froth.
- scarface74 8y ago
  Advertising is not evil, but I prefer business models that involve me giving companies my money and them giving me stuff.
eeeeeeeeeeeee 8y ago
Probably also explains why their messaging strategy is completely incomprehensible too. I still have no idea what apps do what, which ones provide video or audio or text or both. I have a hangouts and voice app on my phone and they appear to have duplicate functionality. I’m not joking — I receive the same notifications on both.
- munk-a 8y ago
  I was super pleased the other day when I needed to fall back on a personal device to join a meeting via GSuite and realized that because it was a meetup my hangouts app didn't work for it. I also enjoy that calendar is also a separate and distinct thing so it's possible for me to get three notifications for a single meeting. Now that's what I call usability and a solid product! If only they could improve this experience by raising the price of their business suite by 20% then I'd be really tickled!
- chewz 8y ago
  I have Google Assistant app but to change settings I must use Google app but some settings like volume, room etc. could be controlled via Google Home app.
avip 8y ago
If so, why is YT (sampled randomly from your list) buggy as hell? Is it possible no-one at G noticed that (sampling randomly again) for example playlist shuffle does not work? How come this is not fixed?
- shawabawa3 8y ago
  Following OPs argument (I have no idea either way) Someone got promoted for introducing playlist shuffle Nobody gets promoted for fixing it
unlimit 8y ago
> As a group, Google engineers and PMs are obsessed with promotion. At the heart of every conversation about system design or product proposal lies an unspoken (and sometimes spoken) question: will working on this get me promoted? > The criteria for promotion at Google, especially at the higher levels like SWE III -> Senior and especially at Senior -> Staff and above, explicitly talk about impact on the organization and the business. Wow! Its exactly like my job at a my IT off-sourcing job. I am very surprised.
layoutIfNeeded 8y ago
Hmm, I’m thinking whether this is an example of Goodhart’s law (when a measure becomes a target it ceases to be a good measure - the measure being the number of product launches) or “tragedy of the commons” (i.e. some sucker will do the maintenance, I’m going for the promotion!). Either way, it’s frustrating as a user. I will never forgive them killing Google Reader.
chippy 8y ago
Note that whilst GMail, YouTube and Search might not be going anywhere they also are not being improved or developed upon. Just being kept running which is the same symptom
ken 8y ago
Stupid question: why are Google engineers so obsessed with promotion? I'm sure any entry-level engineer there makes more than I ever have in my life, and anyone who hasn't cashed in their billion-dollar startup lottery ticket by now is far too late to the party to ever do so. What exactly is the benefit of promotion? Is a 6-figure salary and all the perqs in the world not enough? At all the places I've worked, from tiny startups to big stable companies, I haven't seen this phenomenon. Does Google's rapid rise simply attract the sorts of people who feel they deserve it?
- Aeolun 8y ago
  I’m not sure I understand this perspective. If I’m making 100k, and I could be making 150k by being promoted, while working on interesting new things at the same time (regardless of success), then that is what I would be doing. Little point in sticking at 100k my whole career for some misplaced sense of honor.
- bradlys 8y ago
  You could go from making $200k/yr to $400k/yr with one to two promotions. It's a big jump and for people who live in an area where homes are regularly $2-4m (they're not even nice homes), you have to make big money in order to afford one.

by jfasi 8y ago

mrbonner 8y ago

It is general true for any big organization. In my own firm, I could see the rush into machine learning direction as a door way to a promotion. Any engineer or PM who works on a ML related project and is able to ship it would have a higher chance of getting promoted. I could also say the same for the usage of AWS products. In the end, I often see projects resembling the messy mashup of AWS products to form a unnecessary complex ML solution to a mosquito problem. It’s a waste yes, but sure will make it looks challenging and boost people to promotions.

asark 8y ago
I increasingly see this as the next level of alienation from work-product. Carpenter makes and sells chairs to people who need chairs, farmer grows and sells apples to people who want apples. Then there's the line worker making just chair legs for people they'll never meet, and the farmer selling their entire apple crop to a big company that'll do who-knows-what with them—sell them, turn them into apple sauce, use them as input to some process that turns out a "natural ingredient" that's some pure, extracted chemical contained in the apples. Then there's the sort of crap we're talking about here. Projects that obviously never should have existed in the first place, and/or that would have been better if done in a simpler, faster, cheaper way, but a ton of make-work was thrown in due to agency problems and perverse incentives. How much of the benefit of computers is being eaten by this kind of thing? The most effective things we do with them may provide 1000x the economic output of that apple farmer, or more, but then we spend a bunch of time doing useless or negative-value stuff, without even getting into a lot of the actually-lucrative work having large negative externalities—spyware-enabled advertising, say, or even just advertising, full stop. What does it do to someone's psyche to spend 90+% of their time doing harmful or useless stuff? Hell some people end up doing nothing of actual value for years on end. What are we doing?

ryandvm 8y ago

I totally agree that this is frustrating for erudite tech users, but I'd say calling Google a "disorganized mess of a company" is kinda stretching things. The last quarterly earnings for GOOG were up 21% year/year. That's pretty insane for a company of this size. The way they operate is annoying when it's one of your favorite products that isn't pulling it's weight, but it's hard to argue that the strategy isn't sound.

noneeeed 8y ago
To play devil's advocate, we don't know how sucessful they would be if they followed a more orderly model. Not saying they wouldn't be, but it's entirely possible that they are successfull in spite of, not because of how they operate in much of the business. Given that the vast majority of their revenue comes from just one place, advertising, which seems to be run in a much more cautious manner, there is an argument that this might be the case.
1024core 8y ago
They keep showing more and more (or bigger and bigger) ads. At some point, they'll hit the law of diminishing returns.
jfasi 8y ago
I'm disappointed I had to go this far down into the comments to read an actual sensible rebuttal. I stand by my calling it a disorganized mess, but you'll notice I never actually say this strategy isn't paying off for the company...
TallGuyShort 8y ago
Agree that "disorganized mess" is a stretch. But pointing to a company's success doesn't make them immune to criticism either. My project often gets compared to a competing offering from Amazon. When discussing this with coworkers, I will point out the drawbacks and trade-offs of their design and why that's a WAY bigger deal for my company's customers than it is for the typical AWS user. "But they made $X last year". Yes. And we're less vulnerable to data loss. Thank you. Next.

throw_away 8y ago

> you get a company that throws stuff at the wall to see if it sticks, and if it doesn't kills it immediately. I would love if they followed this strategy. Instead, you have products like reader and inbox with many passionate users, which demonstrate some wall stickiness, and they're killed rather than monetized. Monetize me. I recognize that the things I like cost money. Figure out how to make money off me! I would have accepted ads in reader or inbox. Hell, I would have paid dollars to make those ads go away even.

michaelbuckbee 8y ago
Google reader is the one that kills me. Not only do they not monetize it / promote it, but it's Google subsidized free existence killed the market for alternatives. Then Google+ and their failed social efforts killed Reader and RSS feeds have now disappeared from the landscape leaving us with only proprietary options.
- jamesgeck0 8y ago
  > it's Google subsidized free existence killed the market for alternatives Did it? IIRC, there were at least six decent alternatives available when Reader closed. More if you count desktop and mobile applications.
  - s0rce 8y ago
    I tested a few that appeared from the ashes. Ended up not liking most of them but eventually found bazqux and pay them money.
    - CobrastanJorji 8y ago
      I switched to Feedly the day they announced the shutdown because it could import from Reader, looked like Reader, and used a lot of the same shortcuts. Is bazqux better?
      - 8y ago

noneeeed 8y ago

Is this basically why things like Inbox, and the endless attempts at messaging, all come out as seperate products rather than enhancements to existing products, because that way you get to say you launched something new?

jfasi 8y ago
To be perfectly honest I don’t know the answer to the chat app case in particular. That’s possible, and if one were to investigate and find each app was made by a team rolling up into a different director then I’d say that’s what happened. However it’s entirely possible that it was caused by an overly complex market segmentation strategy. WhatsApp had the advantage of growing slowly and segmenting their users with time to experiment and tweak things. The Google chat apps came out at once, and it’s possible there were so many because they tried to come up with a working segmentation for all addressable users from day one instead of growing organically and grabbing each segment one at a time.
nevir 8y ago
IMO yes: the promotion culture (technical complexity, impact) push people to come up with Big Ideas. The thing with Big Ideas, is that they're often iterations of existing ideas, but just different enough that you can make a case for a fresh start. (And who doesn't want a greenfield project) Plus, the cost of experimentation is far less while you're not supporting a massive userbase.
tdeck 8y ago
That's consistent with my own experience at Google. It's not that working on improving a thing isn't rewarded, but it's just much easier to demonstrate impact when you can use the word "launched" in your self-assessment Launching is in itself an accomplishment, while optimizing requires you to also collect data, do analysis on the data, and show a positive impact. "Impact" here is one of the buzzword categories that goes into an assessment of your performance as an employee.

PopePompus 8y ago

It may be that GCP isn't going anywhere, but when our organization was choosing a cloud service, we chose Amazon over Google specifically because of Google's flakiness. Google could drop cloud services tomorrow, and it would not be out of character.

linsomniac 8y ago
We are in the same boat. My boss came to me last week saying he had lunch with a friend that just moved their services all to GCP and got big savings. But we just recently were impacted by the sudden, unexpected Maps 14x price increase.
- saurik 8y ago
  Similar drastic changes happened to AppEngine, where Google let it rot for something like a year with increasing, extremely bad, latencies and then a sudden very large pricing model shift that made it much more expensive. I was an early adopter of AppEngine, and just can't imagine ever screwing around with Google cloud technology ever again.
- Moru 8y ago
  We got stuck there too, moving away everything to alternatives. Just analytics left to remove now.
jfasi 8y ago
I can’t share with you how much GCP is making in revenue, but I can assure you it’s enough that it won’t get dropped, ever. I’ll put my money where my mouth is: I’m now preparing to expand my own team’s operations into GCP and away from AWS because of how good their ML offerings are.
- jdgoesmarching 8y ago
  Maybe the core product won't go away, but you could still see the damage from this approach in features or services you depend on within GCP. Not to mention Google's love for mixing around interfaces every five seconds (with poor documentation of course). Consistency and reliability matter more than flashy new tech. I wish the culture at Google would understand that.

TheTruth1234 8y ago

Thanks, but to you and any other GOOGLE EMPLOYEES Are you not internally aware of this? Do you just carry on regardless? What happens when you voice this issue internally? As an ex-Google fan boy, the title of this article has been true for years and it's what poisoned me (and no doubt many others) against you. So you must be aware of it. What's your take? I'm super curious as to why you gave up tech (brand) leadership so seemingly nonchalantly.

geofft 8y ago
Google employees in an individual-contributor role have limited power to change this if they want to remain Google employees. They're subject to performance reviews on this system, and the expected value of any individual advocating for a change is so tiny, compared to the expected value of playing along with the system and getting their salary and bonus check. If you want this culture to change, you need the people signing the checks to feel compelled to change it. The practical effect is likely to be people leaving Google because they can get rewarded for maintaining existing products elsewhere.
steven2012 8y ago
Google as a company is worth almost $1T. It's hard to argue that's the wrong approach when they are so ridiculously successful.
- falsedan 8y ago
  Amazon and Apple have similar market caps and yet don't have this product discombobulation
- scarface74 8y ago
  How well is GCP doing compared to AWS and Azure? How well is GSuite doing compared to MS Office? How well is Android doing as far as revenue and profit compared to iOS? According to information that came out during the Oracle trial, Android has made Google less over 10 years than iOS makes Apple during a down quarter. How well is G+ doing against FB? Google hasn’t had but one successful revenue generating product - ads.

s3r3nity 8y ago

I totally understand this - but it doesn't mean I like it. I could get through having two podcast apps on my Android (Google Podcasts, Google Play Music) that give me notifications for my podcasts, with no plans on phasing out one or the other. I could get over having two different music apps - Google Play Music and YouTube Music - because it seemed like there might actually be a plan here - albeit we haven't heard anything in ages. I could get over having two email apps - Inbox and Gmail - because I knew Inbox was just an experiment that was going to be a tease of cool features that may / may not be ported over...therefore I tried not to take the bait. I could get over having a bunch of different messaging apps - I didn't like it, but 95% of my messages are SMS texts anyway, so it didn't impact me. The FINAL straw was falling in love with a PixelBook that I bought during the Black Friday sales, as I moved into a role in an org that was 1) deeply invested in G Suite and 2) allowed me to do all my work from the browser - making the Chrome OS concept finally fit for me. THEN Google closes that part of their hardware division. Now I just have angst over _any_ Google product / service that isn't the core that I use and "know" knock on wood won't go away (Search / Gmail / Chrome). Android wear? Google Home? Hell, I was seriously finally moving to a Pixel 3, as the tight integration with some Chrome OS features on my PixelBook would've been ideal - but I can't even trust that the Pixel phone line will last that much longer.

Wowfunhappy 8y ago
> The FINAL straw was falling in love with a PixelBook that I bought during the Black Friday sales, as I moved into a role in an org that was 1) deeply invested in G Suite and 2) allowed me to do all my work from the browser - making the Chrome OS concept finally fit for me. THEN Google closes that part of their hardware division. The PixelBook isn't going to stop getting security updates any time soon, so if it was a good machine before the division was closed, why is it a less good machine now?
- sooheon 8y ago
  What do you think the half life of most electronics is? Why do you think so many people get anxious about Apple "abandoning" the Mac? A trickle of security updates does not make a well supported and thriving product line.

geofft 8y ago

This went into my decision not to accept a Google offer. AMP is a product that should not exist in the first place. Everything AMP does could be done in other ways—having Google Search measure load time, improving load times, working on webpackages as step 1 instead of an afterthought, etc. But because AMP is a coherent, well-defined product, and Google rewards launching it over accomplishing the same goals in less-obvious ways, the person behind AMP was incentivized to go that route for the sake of his own career at Google, and ruined the internet for the rest of us. I didn't want to be subject to those engineering standards.

jfasi 8y ago
Ehh, I disagree with your assessment, and to illustrate why I think you need to focus on asblockers. AMP serves a critical strategic purpose: it severely disincentivizes crappy, JS-heavy websites on search results and as a result makes people less likely to install adblockers. Google is terrified of adblockers going mainstream, to the point where they’re willing to risk anticompetitive suits by creating their own. AMP is a part of the effort to prevent that from happening.
- geofft 8y ago
  That's an interesting viewpoint and makes a lot of sense, but still - why couldn't Google have accomplished this goal through web search policies? Have the crawler say that if it sees JavaScript with certain behavior it will be de-prioritized compared to a similar website with less annoying JavaScript. Which, to some extent, you get just by focusing on load time: if the crawler emulates a normal browser and measures the time until everything has rendered, it'd be able to penalize annoying JS ads that block loading. Or if it emulates a normal browser and indexes text rendered on the screen, it'd be able to penalize sites that display a pop-over ad that you have to interact with. (Would that have been more prone to an antitrust lawsuit than AMP?)
  - jfasi 8y ago
    I don’t want to speculate too much on the legals of the situation, so I won’t talk about anticompetitive matters. As for why not do this in indexing, of course they’re doing that. They’ve been doing that for a very long time, but that approach can only go so far. There’s a constant cat and mouse game between websites and search providers, and AMP is a way to end it once and for all. It’s important to stress that the motivation here is largely user-focused. People tend to ascribe to Google these shady Oracle-like motivations around screwing users and developers, but the truth is that moves like this are largely motivated by a desire to improve the experience. Not out of any sort of altruism, mind you, but because what’s good for the user is generally good for Google.

dna_polymerase 8y ago

I think Google doesn't care for most services as they do not depend on any of those. Sure Android, GMail, and YouTube will stay, because they serve their core business: ads. Everything else could be over at any time. This is supported by their recent push for China. Suddenly employee voices aren't worth anything anymore (remember the drone story, that was over after employees pushed back). Maybe they have realised by now that diversification isn't that bad after all and will continue to push GCP but I wouldn't want to bet on GCP's life-span.

RestlessMind 8y ago
I am willing to bet on GCP's life-span - it will be around for 10 more years at least. What would you like to bet for?
- iainmerrick 8y ago
  GCP in its current form, though? In 10 years, will you have to choose between the deprecated version and the version that doesn't work yet?

snarf21 8y ago

Well said, you always get what you optimize for.

chibg10 8y ago

In my experience at a different FANG, the same culture exists here. If I had to guess, I would say it probably applies to the others as well.

jonathankoren 8y ago
Oh just go ahead an name Facebook. The irony, is that this pattern at FB and and Big G isn’t indicative of a good engineering culture. It’s pathological engineering.
jedberg 8y ago
It definitely doesn't apply to Netflix. You can get good raises and promotions by optimizing an existing process. If you can show improvement in internal metrics, such as improving latency or lowering cost, that's a huge win. Even if you increase costs you can get a raise and promotion if you show that it increased code quality, such as creating the Chaos Monkey/Gorilla/Kong.
- anoncake 8y ago
  I guess it helps that Netflix essentially has one product.
  - jedberg 8y ago
    Yes that probably helps a lot. Also they don’t hire fresh college grads, so you generally have less of the “wow this was clearly written by a junior engineer” type of maintenance. The low hanging fruit is mostly picked when it comes to architecture and optimization.
phjordon 8y ago
I work at Amazon and I don't believe that applies here. There certainly is an expectation that you can coordinate increasingly broad efforts as you rise through the ranks, but the focus isn't on launching new products (in my experience). Impact can be anything that is positive and measurable. At the end of the day if you make the company $1million then it doesn't really matter where you do it.

iagovar 8y ago

Cool for them, but that hurts my wallet. I never felt safe using google products, and the only one I paid for (adwords) decreased in quality, so...

amyjess 8y ago

> Usually an organization starts with a top-down direction and the rest of the company is compensated for executing it. Not at Google. The "let a thousand flowers bloom" approach that developed from the early days of twenty percent time and total engineering independence has created a disorganized mess of a company. Multiply the individual incentives fifty thousand times and you get a company that throws stuff at the wall to see if it sticks, and if it doesn't kills it immediately. This is also why we had a botched rollout of Material Design across Google's apps following the release of Lollipop. Apps rolled out Material redesigns whenever, often months and sometimes even years after Lollipop's launch. Apps used various different versions of the Material support library: the Drive Apps in particular had text and widget alignment issues that made them look very unlike the other Material apps. And for a while, I was tracking which apps exhibited correct behavior of the status bar and the hamburger menu, because there were multiple ways apps could implement each one, but only one correct way (ironically enough, I was doing so in a Google Doc) [0]. To me, this is an awful way to roll out your new company-wide branding initiative. You want a new brand to be consistent across all your products and rolled out simultaneously. In retrospect, it was the moment when I began to lose faith in Google as a company. Some time after that, somebody explained to me why, and it was exactly as you said: engineering groups operate independently of each other with no top-down chain of command. There's nobody telling them "you have to roll out the new branding by this date", "you have to make sure you're following these company-wide standards", etc. Basically, I was told that at Google, "company-wide" isn't a thing that exists. And that's awful. And honestly, it makes me never want to work for Google. That kind of chaos is exactly the opposite of the kind of working environment I could thrive in. [0] Correct behavior: status bar is a darker shade of the toolbar's color, the hamburger menu covers the screen from top to bottom with the status bar as a translucent overlay on top of the menu bar, and the hamburger icon doesn't animate at all while you drag the menu out. Following Lollipop, there were Material apps that broke every one of those rules, but in completely different combinations from app to app.

dalbasal 8y ago

Corporate political economy (ecology? anthropology?) is always pathological like this, one way or another. Seems inevitable, like factionalism and partisanship taking over parliamentary politics in every example we have. That aside, it kind of makes sense that Google places this much emphasis on the revenue producing parts of the business. On "what about new products and innovation" see the first point. While every corporate pathology is different (Google's flavour seems unique) a near-constant is doublespeak. There might be (or might have been) a logic behind it, it just gets communicated in corporate-speak because plain language is too harsh. Ultimately, at Google's scale (like a VC) a meaningful innovation is >$1bn annual revenue or the equivalent of that. Not many projects have that kind of potential. Letting a thousand flowers bloom means letting 998.3 flowers die.

olalonde 8y ago

> Multiply the individual incentives fifty thousand times and you get a company that throws stuff at the wall to see if it sticks, and if it doesn't kills it immediately. It sucks for some users but isn't it good for Google as a business? The way you describe it, it sounds like Google is internally approximating a free market to allocate resources. I understand the risk of damaging the brand but for a lot of new Google products, the only competition comes from startups which equally are a risky bet.

gotocake 8y ago

I agree with most of what you’ve said, but I’m curious about whether or not YouTube actually makes money. I was under the impression that on a good day it just about breaks even, but has that changed?

jarjoura 8y ago

Facebook operates similarly. They created a culture of working on things that interest you as long as it can show meaningful impact. If you can show maintenance of something impactful, then you can work on it. It's really no different in the startup world. If your startup can't show growth in users or profits, then you shutdown.

bradstewart 8y ago

Any thoughts on whether Google Photos will stick around? I find it's far better than any alternatives I've come across, and the recent Inbox shutdown made me realize how "dependent" I am on it.

rock_hard 8y ago
That’s gonna be an interesting one to watch. Hosting the worlds photos costs many billions of dollars each year! That said it’s a key driver to sell android phones (“infinite storage”)...so for the time being you will be good.
- scarface74 8y ago
  But then again according to information that came out during the Oracle lawsuit, they only made $36Billion from Android during its entire existence.
- dillondoyle 8y ago
  Just googled around and I'm guessing the terms allow google to use the photos and associated metadata which sounds to me like it would be hugely valuable. For ML research, search, advertising, etc. If Google knows where my photos were taken, what products are in the photo, what car, etc.
  - rock_hard 8y ago
    Yeah, people keep saying that...IMO its huge and risky bet to hope for some future way to leverage the meta data for ML

president 8y ago

This mindset is commonplace in most of the jobs I've held in the last 10 years. Personal gain trumps morals, company values, and even company performance. It's what happens when you put a bunch of high-performers in a box and let them play the corporate ladder game.

jimmy1 8y ago
It's better than the situation at most companies I have been at: It's basically the personal gain trumps all, but without the actual results. You just have a bunch of populists that talk up a good game, but when it comes time to execute or actually launch something, they are the last person you really want at the helm, but because they have built up a reputation with the higher ups, they are impossible to displace. From my perspective, this appears to be the situation of most "Health IT" corps and VP of Engineering and above. I've since "sold out" and moved to a digital marketing firm, but in many ways the work here is much more honest than what happened at places I worked at in the name of "health"
tdeck 8y ago
I think that's a bit of an unfair way to characterize what's going on here. In fact, it's Google's values (as expressed through internal incentives) that govern this process. If the company wanted people to, say, make X better rather than killing it and launching a completely new X, all it would have to do is reward that rather than setting up a system that punishes it through reduced career growth.

mshockwave 8y ago

> Maintenance engineering is so not-rewarded that it's become an inside joke highlighted

Animats 8y ago

Streaming game services, not so much. Yes. The game industry is very reluctant to use Improbable's Spatial OS, because, through a deal with Google, they force you to host your game, expensively, on Google's servers.[1] Nobody wants to spend $100 million developing an AAA title using that system, and then suddenly get an email that Google is discontinuing the product. (Spatial OS is technically interesting, and, hopefully, someone else will do something like it, with reasonable licensing terms.) [1] https://massivelyop.com/2018/01/12/chronicles-of-elyria-is-dropping-virtual-world-platform-spatialos/ https://massivelyop.com/2018/01/12/chronicles-of-elyria-is-d...

ernopp 8y ago
> then suddenly get an email that Google is discontinuing the product. Google is unlikely to discontinue GCP
- sgift 8y ago
  The list of discontinued things that people said that about is getting longer by the day. That's kind of the point here.
user5994461 8y ago
Game servers typically don't last more than a few years, even for an AAA game. They pretty much never last a whole decade outside of major MMORPG. It's a bit surprising that the game industry would be scared of GCP going bust, given the short release cycle of the industry.
ocdtrekkie 8y ago
Didn't know SpatialOS was attached specifically to Google Cloud. This seems like a risky business move, people should've learned from the fine example of companies built exclusively on the Facebook platform by now. https://www.hadean.com/ https://www.hadean.com/ is presumably a direct competitor to SpatialOS then, I take it? And it's tied to Azure, I believe.
- Animats 8y ago

stcredzero 8y ago

GMail, YouTube, Search, GCP, Android, and others...deep down they’re frontends for search or advertising... Streaming game services, not so much. Streaming games has tremendous potential synergy with advertising. Games are yet another form of media, after all. There is obvious tremendous potential synergy with YouTube. Also, if one can stream games, then why not stream other kinds of applications? There's tremendous opportunities for synergies there as well.

dageshi 8y ago
Basically identical to youtube. Forget putting AAA games on it, imagine a massively improved version of the flash games that were popular a decade ago.
- stcredzero 8y ago
  Forget putting AAA games on it, imagine a massively improved version of the flash games that were popular a decade ago. I wouldn't be so confident about AAA games being a barrier. I guess AAA games emphasizing "realism" might always have a high-end PC niche, since the genre could always ramp up graphics to the point of un-streamability. I could also totally imagine, let's say, Nintendo, going whole-hog into streaming games with cartoony graphics, emphasizing multiplayer with friends. I don't see any reason why the industry couldn't target and achieve something like 30 ms round trip latency. So long as the gameplay is good, the potential audience for 1080P gaming with even last generation level graphics is huge. To think they would be limited to the level of flash games is hugely naive.
  - dageshi 8y ago
    I guess I was thinking more in terms of what's going to happen when you can literally play a game with the click of a button in a browser like you could in flash. My thought is there will be massive amounts of games flooding the market in much the same ways there are on mobile devices and the price will trend towards zero with microtransactions in the same fashion.

josteink 8y ago

> why this keeps happening, you need to understand the product and engineering culture at Google. As a group, Google engineers and PMs are obsessed with promotion. For people outside the San Francisco and Silicon Valley bubble which are confused by this, it may be worth noting that promotion here does not need mean getting a new job. Everyone are not constantly getting new jobs. “Promotion” here means internal, magic Google-points and seem to work much like Chinese social credit. Source: answer by Googler here on HN when I was confused about this very same thing. https://news.ycombinator.com/item?id=18575875 https://news.ycombinator.com/item?id=18575875

QuercusMax 8y ago
Promotion means you get promoted to a higher level within Google[1]. This typically means you get more money[2], better opportunities, and more visibility. People often do change jobs within Google/Alphabet after getting promoted. It's a signal that you're doing a good job, as verified by a promotion committee, and hiring managers will often look at a recent promotion as a signal that someone is a high-performer. But there are lots of folks who don't change jobs after a promotion (especially if you're happy with what you're doing) - it's an opportunity to keep doing what you're doing, just more of the same, with official recognition that you've been doing a good job. 1. Level 3 = SWE I - New Grad (Bachelors/Masters). Level 4 = SWE II (Newly minted PhDs start here). Level 5 = Senior SWE. Level 6 = Staff SWE. Level 7 = Senior Staff SWE. Level 8 = Principal SWE / Director. 2. This is not strictly true; a high-performer at Level N may make just as much as a low- or average-performer at Level N+1. But in practice you start to hit salary caps, and sustaining very high performance at Level N may be more difficult than sustaining average performance at Level N+1.

rifung 8y ago

Indeed that is also what I thought when reading the very relevant article from yesterday: https://news.ycombinator.com/item?id=19541475 https://news.ycombinator.com/item?id=19541475

neilv 8y ago

How common are career ambition incentives like this, among engineers (not middle management) right now? It sounds like, in Google's case, it's very intentional. Other than the job-hopping that I started hearing about when dotcom IPOs started, engineers, from what I (perhaps naively) understood of my anecdotal experience, seemed to mostly just like to do engineering, and/or also had a sense of obligation to their project/team/duty. About the worst I recall hearing about would be an engineer picking a language/tool because they wanted to learn it or add it to their resume, rather than because it made the most sense for corporate goals for the project.

iainmerrick 8y ago
Not necessarily common, but the ambitious engineers are the ones who get promoted and they're the ones who get to set the technical direction. So the incentives on offer to (over-)ambitious people are very important.

korethr 8y ago

>This is also why GMail, YouTube, Search, GCP, Android, and others aren’t going anywhere. They’re making money, they’re core to the business, and there’s plenty of opportunity to work on them and get promoted. They all also share one thing in common: deep down they’re frontends for search or advertising (GCP and Apps are an exception because they make money on their own). Measuring and proving impact on search numbers is a well-known promo narrative at Google, so those products are a safe bet for employees and users. Streaming game services, not so much. This, right here is the key, and what I think most people miss, including the original article. Hell, I missed it myself for a long time: Google is not a platform company as the original article asserts. They are an advertising company. That is Google's bread and butter, and has been since day one. The products and platforms that stick around do so because they are or become successful funnels or front ends for advertising. As much as the tech culture of the Internet loves to bitch about advertising (myself included), unto itself, it is not a solely an evil thing. Oh, it absolutely can be turned to evil ends; see any of a myriad complaints about Facebook. However, I think the pairing of ads with search has been damn useful in a way that balances with the potential for evil. As an example, when I'm searching for specific lubricating grease for my bicycle's bearings, I very much appreciate seeing an ad for a vendor who's willing to sell me a single tube thereof, instead of having only the web search listings of vendors who require a minimum order size of 1 shipping pallet. Google did right paring advertising with search. To use a 'corporate bullshit' word, Search and ads synergize well. The froth of Google's other products and platforms makes more sense when looked at in that light. A company that sits still will eventually get passed by someone else who correctly sensed changes in the market and business environment, and thus eventually die. So of course Google is going to be trying to come up with all manner of things that leverage the Internet and their massive infrastructure; one of those may well be the next big thing that funnels more money in to advertising. No small part of this is that Google employs a great many nerds like you and I, people who have no trouble inventing cool new things you can do on the Internet. But sadly, a great many of those cool things don't fit well with Google's advertising-based business model, as evidenced by the constant product froth.

scarface74 8y ago
Advertising is not evil, but I prefer business models that involve me giving companies my money and them giving me stuff.

eeeeeeeeeeeee 8y ago

Probably also explains why their messaging strategy is completely incomprehensible too. I still have no idea what apps do what, which ones provide video or audio or text or both. I have a hangouts and voice app on my phone and they appear to have duplicate functionality. I’m not joking — I receive the same notifications on both.

munk-a 8y ago
I was super pleased the other day when I needed to fall back on a personal device to join a meeting via GSuite and realized that because it was a meetup my hangouts app didn't work for it. I also enjoy that calendar is also a separate and distinct thing so it's possible for me to get three notifications for a single meeting. Now that's what I call usability and a solid product! If only they could improve this experience by raising the price of their business suite by 20% then I'd be really tickled!
chewz 8y ago
I have Google Assistant app but to change settings I must use Google app but some settings like volume, room etc. could be controlled via Google Home app.

avip 8y ago

If so, why is YT (sampled randomly from your list) buggy as hell? Is it possible no-one at G noticed that (sampling randomly again) for example playlist shuffle does not work? How come this is not fixed?

shawabawa3 8y ago
Following OPs argument (I have no idea either way) Someone got promoted for introducing playlist shuffle Nobody gets promoted for fixing it

unlimit 8y ago

> As a group, Google engineers and PMs are obsessed with promotion. At the heart of every conversation about system design or product proposal lies an unspoken (and sometimes spoken) question: will working on this get me promoted? > The criteria for promotion at Google, especially at the higher levels like SWE III -> Senior and especially at Senior -> Staff and above, explicitly talk about impact on the organization and the business. Wow! Its exactly like my job at a my IT off-sourcing job. I am very surprised.

layoutIfNeeded 8y ago

Hmm, I’m thinking whether this is an example of Goodhart’s law (when a measure becomes a target it ceases to be a good measure - the measure being the number of product launches) or “tragedy of the commons” (i.e. some sucker will do the maintenance, I’m going for the promotion!). Either way, it’s frustrating as a user. I will never forgive them killing Google Reader.

chippy 8y ago

Note that whilst GMail, YouTube and Search might not be going anywhere they also are not being improved or developed upon. Just being kept running which is the same symptom

ken 8y ago

Stupid question: why are Google engineers so obsessed with promotion? I'm sure any entry-level engineer there makes more than I ever have in my life, and anyone who hasn't cashed in their billion-dollar startup lottery ticket by now is far too late to the party to ever do so. What exactly is the benefit of promotion? Is a 6-figure salary and all the perqs in the world not enough? At all the places I've worked, from tiny startups to big stable companies, I haven't seen this phenomenon. Does Google's rapid rise simply attract the sorts of people who feel they deserve it?

Aeolun 8y ago
I’m not sure I understand this perspective. If I’m making 100k, and I could be making 150k by being promoted, while working on interesting new things at the same time (regardless of success), then that is what I would be doing. Little point in sticking at 100k my whole career for some misplaced sense of honor.
bradlys 8y ago
You could go from making $200k/yr to $400k/yr with one to two promotions. It's a big jump and for people who live in an area where homes are regularly $2-4m (they're not even nice homes), you have to make big money in order to afford one.