Without commenting on the direction of the company in recent years and where it is today, I'll be forever grateful to Google for setting the standard for how tech companies need to treat their engineers.
People who have entered the industry in the last decade or two may not realize it, but the sky high salaries, stock grants, benefits, flexible work hours, free gourmet meals, stocked pantries, gym memberships and lots more that you enjoy every day are all thanks to Sergey and Larry in the early 00s wanting to build a Disneyland for nerds. They were so successful at it that the rest of the industry had no option but to emulate them. Being a software engineer was nowhere near as enticing a job pre-Google.
Google does indeed deserve kudos for being an early adopter of great benefits, WLB, cool offices, etc. But I'm not sure how much credit they deserve for "sky high" compensation. I had a classmate with competing offers from both MS and Google back in 2007 - the Google offer was pretty close to MS's offer, both at the time and over the prior decade.
A big reason why silicon valley comp rose so much in the last decade was because major companies were previously colluding to suppress worker wages. And yes, this includes Google. It ended in 2009, and that's when tech compensation really exploded. See link at bottom.
If anything I would give Facebook far more credit. They were willing to pay more than any other major tech company. And more importantly, they also refused to participate in the aforementioned cartel, which resulted in a ton of poaching atcivity which drove up market wages.
it also exploded due to the growling size of such companies and dominance relative to overall US economy, especially post-2008 in which there was a void filled by the collapse of the housing, auto, brick and mortar retail, and financial sectors (except for Walmart, biotech, payment processing, and information tech, everything else weas hurt badly). such companies play a much bigger absolute and relative role, whether it's shopping, cloud and data infrastructure, advertising, AI, etc.. Companies need to pay top dollar for top talent to get products to market before an equally-sized trillion-dollar competitor does. Due to network effects, speed is of upmost importance.
Agreed on many counts but they did collude (with Apple, Intel, Intuit, Pixar, Lucasfilm allegedly due to bullying from Apple) to suppress poaching workers from each other. I'd say Facebook played a huge role in refusing to collude (as can be seen from this exchange between Sheryl Sandberg and Jonathan Rosenberg [1]) and thus driving up worker wages due to competition.
Yes, no. They are also a major proponent of open office space, and (at least the group that I worked for was) very intolerant of remote work pre-pandemic. It was very grating to see all these showy spaces (cafes, gyms, theaters, climbing walls, ball pits etc) that were featured in the REWS (Google Real Estate and Workplace Services) newsletters, with absolutely no focus on the overcrowded open office spaces that engineers were forced to work in.
I left Google for Netflix mostly because of my inability to work remotely for Google, and the group I work for in Netflix being remote friendly.
Netflix is one of the few companies that imo, unequivocally have a better culture than Google (despite working for the latter). However, it's a much smaller engineering org so I don't think it would scale as well.
From what I can tell there isn't any company wide rules. In my building there are directors who sit out in the open, some who have private offices.
Some non manager software engineers share offices with one or two people, and some managers share an office with someone.
Sounds like if an employee really really wants an office, they can be eligible to share one, but it's not the norm. It's actually pretty quiet though in the "cubes" since there are so many meeting rooms to pop into for discussions.
I found this even weirder, frankly. Everyone is sitting out there in the open, but it's impolite to talk, and most people have headphones on. Strange vibe to be in a giant room full of people that is basically silent, IMO.
I don't think I ever saw one. The director of the product I worked on (comprising a few hundred people spread across a few offices and countries) sat a few "pods" over from me for awhile. He did have priority use of a 1:1 meeting room though.
> People who have entered the industry in the last decade or two may not realize it, but the sky high salaries, stock grants, benefits, flexible work hours, free gourmet meals, stocked pantries, gym memberships and lots more that you enjoy every day are all thanks to Sergey and Larry in the early 00s wanting to build a Disneyland for nerds. They were so successful at it that the rest of the industry had no option but to emulate them. Being a software engineer was nowhere near as enticing a job pre-Google.
Were you working in the Valley? Because a decade before Google I co-founded a company and while we didn't have in house chefs (we were a bootstrap) we had the rest. And not because we were special -- the three of us who co-founded the company were used to those things from our own prior employers.
There were certainly some bare-bones companies earlier in my career, but there were also some - Thinking Machines comes readily to mind - that were pretty famous for pampering employees like this. Also, even at the bare-bones companies, people got offices or at least cubes. Of twelve employers in my 30-year career, only two had open offices and less than half of the rest involved sharing with one other. In 2000 I was in the middle of a four-year stint alone in an office with walls, which seems almost unheard-of now. It wasn't until I started WFH full time (2014 I think) that I was able to recapture that experience.
I'll assume you mean offices rather than WFH, since the latter (and its many confounding factors) has been discussed extensively here already. Personally, I was just happier in an office, even if shared with one other. Cubes were second, open plans last, though there were a few times when I appreciated being "in the fray" with everyone else either way. Eliminating the constant sensory assault just makes focus easier, and at the end of the day I could go home with lower cortisol levels. Collaboration was never an issue, as long as there was some space nearby for it and people were willing to use it.
Obviously it can be easier to slack off in an office, or there could be conflict with one's office-mate, and so on, but overall I think offices were worth it. I've also been told by multiple CFOs that the reason for cubes and then for open plans has mostly to do with cost and how various kinds of space are taxed. Every one admitted that the productivity/collaboration claims were just excuses. Which brings us back to WFH vs. RTO, I guess. ;)
What was the actual reasoning? Like, Zuck is a survivor and genius I respect (he's not perfect tho obviously)...but why does he do open plan at FB? Why did any of the tech companies do open plan? Are they thinking about it? Do they measure it? What's the rationale?
Another side of “for nerds” is that early Google managed to attract employees from academia. I remember the buzz in the academic world when a professor (Matt Welsh) gave up tenure at Harvard to join Google: some others followed. Even more, it attracted top students who otherwise had their heart set on academia: within the (delusional) world of students who consider being a professor as the highest ideal, I remember hearing the opinion that if you can't get a tenure-track job, Google is a pretty good second choice because it's “just like a university”: you had (so the story went) freedom to do whatever you wanted, smart peers, “talks at Google” from the best speakers, … and there were perks (free food, laundry, etc) that appealed to students.
I myself ended up joining Google when my academic path was interrupted, despite thinking I'd never do that. Google employees were already complaining in the early 2010s about the changing culture, but there was still enough of it left, and by all accounts, early Google was a magical place.
I joined google after self-interrupting my academic path. Later I realized that Matt Welsh was the guy who wrote the first Linux manual I ever had (https://www.mdw.la/papers/linux-getting-started.pdf) which was pretty much the gateway for me learning what I needed to to work at Google.
My goal in joining google was to treat it as academia and use my position there to return to a stronger position in academia. I didn't manage to do this, but I was a more productive researcher as a SWE at Google than I ever was working in academia.
For a while Google wasn't just disneyland for nerds- it was nerdtopia. It was the first place I worked that had more people who were more nerdy than me. At first that was great but eventually it got tiresome.
R&D in pharma/biotech can be like this. But there are fewer and fewer free-wheeling R&D groups in industry these days, probably because people saw Xerox PARC and thought it was a waste of money (a single invention at PARC paid for all of PARC).
I'm not sure how much of it is Google and how much is just a broader collection of SV companies, but engineering salaries, including software, in general were much more conventional professional salaries until at least the dot-com era. I started with a computer company for about $40K in the mid-80s (with a grad degree) so a bit over $100K today with nothing special with respect to other benefits.
One thing that has definitely happened (at least where I am) is that other professional/office salaries seem to have stagnated somewhat and not kept pace with inflation, while tech salaries continued to climb upwards. Until recently, anyways.
It's clearly more complicated than this, as people are pointing out, but I think your point is directionally true, and worth making. Maybe somebody else would have raised the bar, maybe the bar wasn't raised as high as was claimed, maybe they even raised the wrong bar. But, Google did pressure many other employers to change in a generally positive direction, so I'd give them that.
Analogously: I'm not a fan of Starbucks coffee, but I recognize that (where I live) specialty coffee places mostly did not even exist before SBUX popularized the idea that coffee didn't have to come out of a #10 can. You can raise the level of competition while not being the best yourself.
Hmm, I still fail to realize that, as except flexible work hours, none of this is matching my experience as a software engineer. I live in Europe though.
It's not that I complain of my situation: I'm healthy, have a loving family, and work with a great team which is full of kindness and which is part of an international group, something that let me exchange with people from all around the world as part of my job in a full remote position.
So this is just to say that not all SE live in this ultra-privileged bubble. :)
You really need to read some mid-90's books like Microserfs or some of the early Mac histories for some historical context. None of this was new or particularly unusual in the valley at the time.
One of the thigns that attracted me to Silicon Valley was the book Hackers, which has this classic line:
"During this period when everybody at Apple was celebrating the increasing
revenues piles of money that would make many of them so rich that they would be
beyond millionaires, in the ozone of Croesus Mode, where wealth is counted in
units of tens of millions John Draper was at home, playing with his Apple"
"wealth is counted in units of tens of millions" how cute!
Most of what Google did was copying other companies, like Genentech (I've worked at both, twice). It's almost as if Google got a list from Genentech and then just added 25% to everything. There were a few benefits that I haven't seen anywhere else, such as the death benefit, which led me to joke that I was worth more dead than alive.
You say it like paying 25% more than the prevailing wage is trivial for any company to do. Yes all they did was spend more money on talent than everyone else, and that's exactly what boosted the market.
We are talking about the bay area. No one – IBM, Cisco, Apple, Oracle, HP, Novell, Intel, Motorola, 3Com, Yahoo or any of the other big industry players that employed software engineers – paid anywhere near the salary and benefits of Google.
When I chose to return to Genentech I had an offer from Apple and it certainly met or exceeeded everything in my previous Google offers. This is consistent with my read on other employers like Facebook and Microsoft.
> sky high salaries, stock grants, benefits, flexible work hours, free gourmet meals, stocked pantries, gym memberships and lots more that you enjoy every day
Most of us never have and never will. This hasn't become the standard just a perk for a small subsection of mostly American FANNG employees.
As a young teenager learning to code in the mid/late 90s I distinctly remember two influential movies: Office Space and Hackers. It wasn’t until the dot com boom that I realized you could have a lot of fun and stay out of prison.
haha I have thought about this. I saw office Space right before I got my first coding job. So I expected the lifestyle of the typical engineer to be in line with the 'drive an old car, live in a small apartment' representation in the movie.
Twenty something years later never in my wildest dreams did I imagine I'd have insane compensation and perks and yet an even shittier desk situation than depicted in that movie lol.
> People who have entered the industry in the last decade or two may not realize it, but the sky high salaries, stock grants, benefits, flexible work hours, free gourmet meals, stocked pantries, gym memberships and lots more that you enjoy every day are all thanks to Sergey and Larry in the early 00s wanting to build a Disneyland for nerds.
Two comments:
1. They wanted to encourage people to work later
2. Where the f%%% was all that stuff when I was at AWS /s
I did work remotely the entire time I was there (my department didn’t have an RTO mandate). But I did travel to corporate offices in DC, Nashville, Dallas, Seattle and San Francisco after Covid lifted and I was…disappointed.
By 1998 many of the things you mentioned where common in Silicon Valley. So, Google was hardly the first company to pay very well or have great perks.
Historically a huge range of perks have been tossed around including some that have gone away like company cars due to tax code changes that eliminated the benefits. There’s long been ebb and flow around what perks are common, and which ones stick around as the company grows.
One that I wish caught on was Microsoft being rather famous for offering engineers an actual office with doors that closed.
I'm curious about this. Was it true that engineers in the Valley made good money before Google and the like?
I joined the industry after the dotcom boom, and I got the impression that older companies such as IBM were very very mediocre in pay, nothing like Google.
Was that not always true? You see that many people who joined after the IPO were able to retire after 15 years at Google (or Amazon), did that really happen at places like IBM for the average senior engineer?
You'd here about engineers who sometimes became multi-millionaires from an IPO that went well, but now people can become multi-millionaires from W-2 RSUs and a little bit of stock growth.
IBM wasn't Silicon Valley. IBM was IBM, and then there was Silicon Valley.
IBM is cut from a very different cloth than Silicon Valley companies that were actually founded and ran in SV in the 80s / 90s. I wouldn't use them as the measuring stick.
Microsoft was pretty well known to pay really well too (granted out of Seattle / Redmond, but they grafted the SV culture really well). They also had perks and bonuses and big stock grants. Netscape, Yahoo! etc. They all had some variation of this going on.
It’s hard to do direct comparisons when most of this isn’t public information, the job requirements are so different, and inflation plays such a big role in these comparisons. A dollar is nominally worth 1/2 what it was in 1995, but if you’re looking at housing etc the difference is much larger.
My understanding is adjusted for inflation the minimally skilled developer was actually better off on the mid to late 90’s. The median developer was in a roughly similar situation. The elite’s pay was comparable, but today we have a larger percentage of people in that elite category.
PS: IBM isn’t Silicon Valley. Many developers today are still making sub 120k, but that’s really not what we’re talking about.
Netscape and SGI are two good examples that were making people into millionaires in the 1990's. Not to mention all the "Microsoft millionaires" (stock options, baby)
Engineers in Silicon Valley did well, but the multiple over other work wasn't as high. I think that's still true for many employers in Silicon Valley- they pay better than average, but when you factor in all the details (long commute, high prices for everything, decaying infra) it's not really a premium.
I seem to recall that at one point Microsoft guaranteed to match Google's offers, because Google had clearly figured out how to hire the best engineers.
It certainly doesn't seem that way now, unless there is a massive skew in titles (maybe an L4 at Google is given Senior Engineer 2 at MSFT or something?)
> the sky high salaries, stock grants, benefits, flexible work hours, free gourmet meals, stocked pantries, gym memberships and lots more that you enjoy every day
Just a reminder that outside of SV/FAANG (so, 99% of devs?) all of that is not even close to be a reality.
I've never lived in SV, but many many companies at least attempt to offer some of the perks big tech companies have come to be known for. It suddenly wasn't enough just to have free snacks and coke, companies started offering more free meals (if not every day), larger bonuses, better workstations or laptops, etc. It really did resonate out to the rest of the industry.
And how were the conditions 25 years ago for these same developers? Just because everyone doesn't get paid the same doesn't mean the overall compensation in the profession hasn't gone up.
I'm not sure how much it has relative to other engineering and arguably comparable professions more broadly. Maybe somewhat? But $100K+ as an entry-level in enterprise IT is probably pretty solid.
There are dramatically more software developers today in the US than 25 years ago. 1.6 million with a median salary (before benefits) of $109,000. So the field got a lot larger in terms of labor (2-3 fold depending on how you want to count it), while simultaneously paying so well - a particularly remarkable thing. Those jobs tend to come with good benefits as well, including nice health insurance (which is obviously extremely expensive in the US).
Granted, this is peak software developer and we're just starting on a downslope for the profession (both in the US and globally).
I partially agree; they did make a big splash with all of those benefits, but also Google excludes 54% of their workforce (TVCs) from these benefits. Its a bit of a stretch to say tech workers owe Google for these benefits at other companies. To me, they're a shrewd US corporation (not that there is anything particularly wrong with that) where I don't see a lot of "goodness".
The spoiling was already quite obscene during the Dotcom Bubble 1996-00. IPO valuations were directly related to the number of programmers in a company, so they did everything to hire and retain staff.
Google is a great example of "it's difficult to get someone to understand something if their paycheck depends on not understanding it".
Even before google was founded and the founders discussed how they could monetize their search, they understood that paid ads create a conflict of interest: the point of search which is to to return the most relevant results, but the incentive in paid ads is to return the best paying ads. One of their papers contains this point _in writing_.
However, every single time I pointed this out to a person working at google they immediately jump into Olympic level acrobatics to explain why the two things are not in conflict. I especially enjoy hearing the argument that in a free market the most relevant advertiser also happens be the one who can afford to pay the most. I've never worked at Google, but I would bet that part of their induction is to train their employees how to combat these ideas.
One of the most shocking thing I experienced working at Google compared to my previous job (e-commerce with very thin operating margins) is that hardly anyone cared about potential revenue vs cost tradeoffs or monetization more broadly.
Vast majority of people are working on building some feature functionality or enhancing an underlying system infra for more scale or performance or such non-functional capability which they genuinely think will improve user experience and hence make users prefer to use their product more than anything else.
People didn't focus on business metrics (most people don't know what they are) – this was most disorienting for me. It took me sometime to understand the luxury provided by strong margin business that is google ads which allows you gloss over short-term inefficiencies while exploring new product spaces.
It is assumed monetization in the form of ads or subscription can be done if there is strong user engagement for the product and that revenue vs cost equation will work out. (If not, they kill it, which they do badly). Another culture at Google is people are openly critical of their product in very large internal discussion forums.
I skimmed both articles and they seem to be about anti-trust, which is orthogonal to the relevance-revenue conflict you described. Have I missed anything, or did you think a random unrelated fact proves your point somehow?
> Even before google was founded and the founders discussed how they could monetize their search, they understood that paid ads create a conflict of interest: the point of search which is to to return the most relevant results, but the incentive in paid ads is to return the best paying ads. One of their papers contains this point _in writing_.
It's not just their papers. Their philosophy - which was a big part of everyone moving from other search providers in the late 90s - was never having ads in the search results.
I had forgotten about that, but now that you remind me I used to make that point when I recommended Google to friends. Now I feel like a sucker all over again.
> part of their induction is to train their employees how to combat these ideas
You dont need to indoctrinate people; the selection process takes care of most of the heavy lifting and the rest is taken care of by the fat paycheque and stock options.
imo there is absolutely nothing wrong with the conflict of interest, pretending it doesnt exist however suggests either dishonesty or a special kind of stupidity.
> imo there is absolutely nothing wrong with the conflict of interest, pretending it doesnt exist however suggests either dishonesty or a special kind of stupidity
Right, in principle I don't strongly disagree with you. Certainly in the short term the intelectual dishonesty bothers me more than making money by bumping up results.
But whether that is a real issue or not, depends on how relevant Google is in our society. If there's alternatives, people can try them. If there's no alternatives, having the power to control the results to benefit yourself is maybe too much power concentrated in too few hands. It's unclear to me where we stand right now. I haven't used Google search in many years (DDG), but I'm aware they're hugely influential to basically everyone else.
Now think of ChatGPT with this framework. Because it is paid to do, it will slowly try to convince you to buy something or even worse think of some politician or party differently. Scary stuff!
The nice thing with ChatGPT is they've established a viable funding model through paid subscriptions from day 1.
On the other hand I predict that we will see the next ten years as a 'golden age' of steadily improving performance from ChatGPT, which will then plateau and 'enshittify'.
I don't think we know yet if it's viable or not. All they've said is that Plus is break even on "compute" (i.e. probably not training, definitely not staff/data costs).
> I would bet that part of their induction is to train their employees how to combat these ideas.
Exactly. This has been reported [1] and expanded on [2]. In a nutshell, they have internal learning materials designed to train employees to not recognize anti-competitive behavior.
Some of the arguments aren't actually Olympian, although if you have a pre-existing bias against commerce or capitalism it may feel that way.
Most obvious example query of where PageRank totally fails: [flowers]. The correct top result here for 99.9% of searchers is a local flower shop or possibly a big flower delivery chain, but it's not the Wikipedia page for flowers. Likewise for [hotel near me] which an algorithm like PageRank will yield garbage for, [oppenheimer showtimes] and many other very popular query types.
Link based ranking just doesn't work for this sort of result. You need data that isn't on the web which means you then need a way to rank it. And how exactly do you plan to rank flower shops or movie theatres? You've got no data on which to do that. Even if they have a website at all (big if) people don't organically link to them. Attempts to gain ranking data via things like user reviews will fail, almost nobody real reviews local stores or cinemas so if you rely on that then either you'll be flooded with fake reviews, or companies will aggressively push people to write them, or both, and users hate both outcomes.
Ad auctions work much better. Users don't have to do anything directly, they just buy things from the best businesses as they see it, revenue flows to those businesses enabling them to bid higher for ads, so users see the most successful businesses as determined by the market as a whole, which is much harder to game than user reviews or PageRank.
When the original Google was built, it was built by students. Their first ever test query was the name of the guy who ran the university. Their edge was ranking publicly available web pages using link graphs, so they just weren't thinking about commercial queries at all and if that's the scope of your ambition then yeah it will look like ads are always a conflict of interest because the only model you can imagine is re-ranking organic informational results surreptitiously. Once they turned on their first search backend deal and got experience running a search engine with serious non-nerd user traffic for the first time, they realized there were tons of popular queries that their whole approach just sucked at and would always suck at. That's when they realized that in some cases "ads are information", as they put it.
That wasn't some sort of paycheque-induced intellectual blindness, if it was then Google would have gone bankrupt a long time ago. It was an actual fundamental realization about the nature of what people use search engines for.
At it's heart advertising is about informing potential customers about products or services [1].
One aspect you've missed though is the free-versus-paid dimension - here the competition is between people creating links, versus paid for promotion.
You could argue that reflects the value perceived by the paid or free 'customers'. However I think in the area of politics and ideas there is a dangerous imbalance - because of the power of professional promotion.
In essence common sense is being drowned out by paid agendas - because of the difference in impact a paid professional can make.
[1] Let's ignore all the dark arts aspects for now.
That's one interpretation. An alternative interpretation is they realized that making non-paid results worse increases the clickthru rate for the paid results.
Yes, there is a conflict, but it's not an unsolvable one.
Google implemented a compromise - paid ads get displayed at the top, but they are marked as ads, so if you want the most relevant results, you scroll below the paid ads.
Indeed. Google is far from being a saint. The idea I'm disputing is that the conflict between organic results and ads is impossible to solve. Google solved it many years ago, but is now regressing from it.
If you have to scroll, they are not displayed. You seem to have a working definition of "display" which does not include making something visible, but that conflicts with every actual definition of the term.
> if any of you google employees reading this thread, if you love open and free web, get out of google. you're part of the problem.
Shouldn’t we give the opposite advice? To stay at Google in order to gain influence as to create initiatives to guide the company into a better direction?
The same way libraries are free: they're actually paid for by taxpayers.
If the internet was invented in America's earlier days, a lot of these information tools would be a public service, similar to libraries, and we'd all be better for it.
You're under the mistaken impression that Americans like their taxes to go towards public services. Many public libraries are underfunded, and what funding they get is sometimes wrapped up in censorship laws serving an increasingly hostile and paranoid constitutency. "Public service" is synonymous with "government controlled service."
Google would be no different. Forget any "innovation" any of Google's satellite projects may have had, they simply wouldn't exist, too expensive. The service would be stripped down to a bare bones minimum of functionality and farmed out to lowest-bid contractors to save money. The CIA, NSA and law enforcement would have backdoors everywhere (so no difference there), "controversial" sites would be delisted or fined (because broadcast decency laws would probably apply, and the taxpayers would insist any indecency be removed) and the whole thing would wind up privatized by the Republicans anyway, so it would still run ads at the end.
They bought doubleclick, which was already doing that. I don't think Google invented this, and if they hadn't existed, wouldn't it have continued to happen anyway?
While I understand what you mean, you should state at least one reason of why you think they are deeply evil people.
To me, it's the story of how they systematically convince small firms that they want to do business, only to screw them over. Like Google Earth and Terra Vision, for example.
Also, from the beginning, the proximity to the U.S. government was very noticeable, such as Keyhole or leasing military land.
Edit: To make it very clear, I think Google is evil and was evil all the time. It is at least one step further to a brave new world dystopia.
One can make a strong argument that actively seeking collaboration with an organization that runs a global network of extralegal and extrajudicial torture centers qualifies as evil.
AWS, for example, operates a special airgapped "region" at Langley for the CIA to store and process their data, which presumably includes operational information related to the global network of assassination and torture personnel they develop, own, and operate.
Many people give the military and military-adjacent a moral pass on things like mass murder (or mass murder adjacent things like extrajudicial assassination, kidnapping, or torture), but not all of us do. You may or may not personally disagree, but it is still a reasonable and cogent argument that seeking active collaboration with DoD qualifies for the title of "evil" on the grounds that it will more effectively enable mass murder, assassination, torture, and other deprivations of human rights.
Google also rebooted their censored search project, known as Dragonfly, to re-enter
the Chinese market, which included the specific feature that would display government-provided AQI indicators/values instead of the real ones. Eventually word got out and they stopped again.
This shows that they are willing and able to collaborate with any government, no matter how repugnant their stance on human rights, if it will result in increased revenue or market access for Google. That's evil, in my book.
> This shows that they are willing and able to collaborate with any government, no matter how repugnant their stance on human rights, if it will result in increased revenue or market access for Google. That's evil, in my book.
Google actually left China and didn't go back. That's rather different to other large tech companies. Consider Apple, which not only never left China but become totally dependent on it for manufacturing the phones they sell elsewhere, and direct revenue from iPhones and the App Store. For example Apple is regularly giving user data to the Chinese authorities -- they get thousands of requests per year and handing over the data in the vast majority of cases [0]. Google has accepted one request in a decade [1].
So it's pretty odd to use China as an example of why they're evil. How can their leaving and aborting a later attempt to re-enter be worse than Tim Cook being held over the barrel by the CCP and doing little to get out?
At present, China is the only place on all of Earth that can manufacture 50,000 iPhones per hour, which is on the close order of how many they sell. There's literally no other way for Apple to come up with the 250 million units a year that they sell.
In a certain sense, the iPhone is a wholly Chinese product. Without China, there is no iPhone, and there is no Apple-as-we-know-it-in-2023.
I am of the belief that Apple is doing everything in their power to change that, as even a cold war with China would put a rapid end to that.
Apple is investing huge amounts of money in India to create a parallel supply chain for iPhones. I imagine that people in the USG are acutely aware that the CCP has, at present, ultimate control over the revenue stream of one of their largest and most profitable companies, and have coordinated a response plan to address this (ultimately untenable) state of affairs.
Yes, I agree. That makes Apple completely beholden to the CCP. If they anger the CCP, their whole business is gone.
But I fail to see how that makes any of this more acceptable. They had 15 years to diversify. They didn't, but chose to instead spent that time just becoming more involved with a brutal autocracy, gambling their whole company on it.
It was stupid (one of the rare strategic missteps by Apple), it was amoral, and somehow still this is seen as being totally fine while Google toying around with launching a single product in China but not doing it was quoted as evidence of them being evil in this very thread. And I kind of feel like your reply reinforces that point.
For me it's the extreme surveillance they have introduced. Every website activity you do hits google. Smartphones log everything to google. Look on google maps, you can see exactly where you were every day in the last 10 years. This whole infrastructure should just not exist.
Yes I know you can opt out of some of it but they make it as hard as they can and many things like google analytics just can't be opted out of.
But I edited the deeply evil part as I also thought it was too harsh :) It was only there for a minute so I'm surprised you saw it. I do really believe they made the world a much worse place though. And yes if they hadn't someone else would have (like Zuckerberg). But that doesn't forgive it.
Just judge them by their own supposed goals, organising information. They fail. And they fail not for inability, but for greed. You can quibble about what types of non-good behaviour constitutes evil, but first accept that "evil" was cute language.
They are a distorting middleman and their scale means the cultural damage is incalculable. They create a near monopoly search that offers no user control over it's behaviour and it functions to vanish small independent creators under commercial interests and adverts. Just look at one property - youtube - the inability to find title matched videos by search, the shovelling of shorts, the pruning of valuable user features etc. None of these have been about increasing user experience, user power or the access of information, it's about increasing Google's Smaug like horde of gold.
I see it as a guilty conscience in anticipation of the inevitable.
Even the naive do not take long to realise that capitalism will fill the immoral corners of an opportunity like water. If you wanted a real commitment, you don't talk to yourself about sobriety as you walk into a bar, you have to tie yourself to the mast Odysseus-like. It needed a different form of governance, a non-profit that could have created sustainable non-evil. An IPO sells your soul not just your company.
What's also incredible about early Google is how FAST the company created unique technology that led to decades-long profits.
The company IPO'd in August 2004, less than 6 years after it was founded. I think the market cap was around $20 B then, and now it's in the trillions.
I don't think any other company since has done it that fast -- Stripe, Uber, AirBNB, Dropbox, etc.
Compare Facebook, a company that also grew quickly. It was founded in 2004 and IPO'd in 2012, so more like 8 years.
And not just IPO'd, but created BOTH interesting tech and product. Back in those days I loved using Google -- it's a shame that most people don't remember that, and don't have the chance to experience it.
People would say the same things about Google that they say today about generative AI -- I remember an acquaintance wrote a blog post "Is Google God?"
Also, if you compare OpenAI, they're about ~8 years in now, and obviously have done more than 99.9% of organizations in 8 years. But they're also not fantastically profitable, and there's a big "research preview" disclaimer below their main product.
...
I joined Google less than 7 years in, right after the IPO, and it was already a smoothly running machine. It already had 3,000 employees and a bunch of offices, and good management.
I do think the IPO changed the company permanently. There was a bunch of turnover around that time. I remember Wayne Rosing, the VP of engineering, left shortly after I joined. I maintained/rewrote tools written by early employees, and I'm sure they were all (rightfully) taking a victory lap, at least in their minds ...
Definitely sad to see the hollowing out of all that talent in the last ~10 years. I think the founders just weren't prepared for how big it would become, and the company got beyond them.
Adobe had sky high growth back in the 80s. It was the first company out of SV that achieved profitability in less than a year (thanks to a sweetheart deal with Steve Jobs / Apple), however it did really well through the decade. Inflation adjusted, I think it'd hit similar numbers, though I think Google is still bigger by some margin.
Adobe today maybe not so much, but at one point they were a powerhouse
Oh yeah for sure, I can only imagine Adobe in its early days was just as exceptional as Microsoft or Apple back then.
I would have thought another Google would come along in the last 25 years, but weirdly/arguably it was Apple, Amazon, and Microsoft that out-competed them! (in mobile, e-commerce, and cloud)
Those companies are all older than Google!
And there's definitely a timing issue play. The market in the early 80's was "ripe" for the PC boom and Apple, and the market was ripe for desktop publishing and Adobe.
The Internet was likewise "ripe" in the early 2000's.
It feels like the last 10 years have been full of grasping at straws for the next big thing:
- google glass was definitely a "jump the shark" moment
- VR / metaverse seems to be "forced"
- there was a bunch of hype around 3D printing for consumers, rather than for industrial use
- In 2023, self-driving cars have achieved 1% or less of what people thought they would in 2013 or 2016
- crypto and NFTs also feel "forced". Bitcoin is 15 years old now. Whatever you think of it, it's not like the PC era or the Internet era. 15 years in, those were way more impactful.
- I mentioned generative AI above -- it's interesting, but it also has big question marks in terms of BOTH the product and business. Google had neither of those. It's also 11 years in to deep learning.
Of course, this could be just an old guy talking ... and probably missing a lot of what's useful and new. But it does seem like Google was singular -- both the right thing, and the right time. I'm sad to see how bad the products have gotten and how disappointed people are in the brand
Even a relatively large 3,000 person company is much different from a much larger company and especially if you've dome well financially it's pretty natural to either leave or just coast for a bit. And depending upon where you are you maybe don't want to take a new job but just take your pay for a while.
But I'm sure they love being billionaires now. At some point most people will sell out. You see it all the time with companies, mainly consumer product companies, that sell to private equity who then create objectively worse products than before.
That might be true, but I don't believe people sell out for the money. There is no concrete difference in lifestyle between having a hundred millions and a billion. And no matter how luxuriously you live, there will be an enormous amount of wealth left to your heirs.
When you have that kind of money, changing the world and the future becomes more interesting than getting as rich as possible. After all our time on earth is finite. And as a tech giant you have a good possibility of doing just that.
So when extremely rich people sell out, I think the reason is rather some other kind of pressure and influence, that they might not be able to talk openly about.
If a dog steals a sausage, it's because it was hungry. But if a dog that has just eaten a hundred sausages and has his dog house full does it - then it's not hunger.
If this was about legacy, not money, do you think they would let enshittification proceed at the pace it has? No, it's about money. Always has been. Hence the quote.
You understood the opposite of what I wrote. What I'm saying is that people who are ultra-rich are not selling out because they want more money, but rather because of threats, extortion and such shadowy pressures. Combined with money. The carrot and the stick.
The user demographics changed a lot. When google started, most people didn't even use the internet. Today, most people go online to shop, watch videos, or socialize. Today's Google serves that target pretty well, at the expense of the relatively small number of people who are searching for technical information about programming languages or frameworks.
Not coincidentally, a lot of ads target that demograpic as well. So naturally google's top results are also ad-laden.
Google still serves me well when I look for technical information, but do you know a search engine that's even better for that purpose? (I also use ChatGPT, and that's sometimes better, sometimes worse, and always slower.)
There's no perfect business model. The alternative they had in mind was running it out of academia, with the growth of the web that was not financially viable.
As another comment in this thread points out, when google started it was only available from Standford and Berkely. It could have been another JSTOR. I'm glad it's available to everyone, and that's thanks to ads.
And here's 20 Billion dollars that disagrees with you. If switching was just a "click away" why would Google pay Apple $20B per year to be the default search for iOS etc.
Competition being a click away is what keeps quality between Google and Bing comparable, and what keeps Google always trying to invest to stay a step ahead.
But once quality is close like that, defaults matter, which is why Google pays Apple.
If Bing wasn't there (or barely worked) and there was no competition to Google, then Google wouldn't need to pay Apple, because Apple would never consider Bing as a default in the first place.
The payment is a sign of how strong competition is, not a sign of a lack of it.
In this case we have two ad supported search engines that deliver a very similar experience. As you note the differentiation is not great and thus, defaults matter. So in this case they are no longer competing on product quality, but on distribution. A high barrier to entry ($20B is high) but at <10% of Google's revenue it's affordable to them but basically no one else.
There is also brand competition and Google has done a fantastic job building very strong brand preference. Honestly no one has even tried to build an alternative search brand for over a decade in any meaningful way (DDG and Bing combined is the closest we could probably come and that's like Fresca competing with Coke).
Google Chrome has ~63% market share and that's not up for auction. They pay for ~23% or so at the $20B discussed thus no one is going to be able to out-distribute Google any time soon.
Furthermore no one, including Microsoft, is going to risk $20B on overcoming Google's brand preference. It would take years and probably $100B or more to play this game for an uncertain outcome. Better to invest in OpenAI than take that bet.
Credit to Google; they have built a very strong business with a strong moat. But no, real competition is not a "click away."
> So in this case they are no longer competing on product quality, but on distribution.
Incorrect. Quality is a moving target as search gets better. Competition is in quality and distribution. If Google stops improving and lets Bing move ahead on quality, it'll be dead. Witness the recent worries about incorporating LLM's into search.
> Furthermore no one, including Microsoft, is going to risk $20B on overcoming Google's brand preference.
Earlier this year there were reports that Samsung was considering switching to Bing, and it was suggested there was a similar $3B contract involved. So of course Microsoft is in the same game, and it's quite obvious that Google wouldn't be paying these sums at all if Microsoft weren't willing to offer something as well (but keeps getting outbid).
E.g. if Google weren't paying Apple $20B, it's quite likely that Microsoft would be paying Apple $15B instead. One of them would be.
Okay, they complete on quality but they are "close" to use your term.
My point was that paying to be the default is not enough. A challenger needs to also overcome the brand preference has built. A one-time investment of $20B (or even $3B) would be wasted. A true competitor would have to be in this for the long haul and willing to outbid Google for many years. Additionally they'd need to invest in building their brand image against Google which would be expensive and risky.
When Google later went public in 2004, Larry, Sergey, and Eric made a pact to work together on Google for the following 20 years. That's coming due soon in 2024.
which is the worst outcome possible- sergey is very disruptive. He will come into a project that is dealing with some challenging and real problem and say "well, why don't you just..." and while his idea sounds plausible, after you halt your current approach, context switch to chase the billionare, you learn that the idea wasn't a very good one. I know a few managers at google who had to work to keep sergey away from his employees because sergey would waste everybody's time.
(sergey is a genius and should have run his own R&D group and not bother other groups)
Wikipedia is probably the biggest contender. I'd say Wikipedia's growth is even more impressive because it hasn't been fueled by a bunch of acquisitions.
To name a few (US centric ones) from guesses.... Ford, the British East India Company, Bethlehem Steel, Standard Oil, Bell, Apple, Boeing, and probably Amazon.
I have the same first reaction. Google's progress and impact on the society are very impressive, independently of how we feel about it.
But come to think of it, advancements in technology catch up quickly. Looking back into history: Ford Motor Company, Bell Telephone Company: they created a profound impact on the society in less than a couple of decades. I am sure there are other examples. I wonder: at what point in time did it accelerate? With the start of the Industrial Revolution?
Isn't it a bit too early to reason about it? We are not sure yet of the extent of its impact, there is a lot of hype around it recently. I'm not diminishing it, just saying we will be able to tell better once things settle down.
About a year before, in mid-1997, word got around that there was this cool new search engine project that was giving great results. You could get to it only from the Berkeley and Stanford campuses at http://google.stanford.edu
Berkeley and Stanford, despite their sports rivalry, actually work together a lot and at the time had direct connect networks between the two schools (which also made file sharing across dorms at both campuses a lot easier!), which was the only reason we could access it.
We all quickly switched to using it because it gave far better results than the best alternative at the time, AltaVista.
don't forget akebono.stanford.edu... the early version of yahoo! It's strange to think that I saw a company go from a college website to a multibillion dollar media conglomerate only to run itself into the ground.
> Bechtolsheim and Cheriton were two of the first investors in Google, investing US$100,000 each in September 1998. When he gave the check to Larry Page and Sergey Brin, Google's founders, the company had not yet been legally incorporated.
"Masterpeace" lol :) But I understand what you mean.
I didn't know McDonnell Douglas was such a bad company but I heard the same sentiment from Boeing people in that documentary about the 737 MAX, that after the merger they were being forced to cut corners by the new management.
Anyone else remember how much better Google was at the beginning than anything else?
I had always tried the latest search engines like Ask Jeeves, Altavista, Hotbot, Webcrawler, etc. Once Google came around they all looked like total crap, it was simply a revolution. The clean interface, the performance, the much better results, it was beautiful.
I do remember. Everything else looked like a directory rather than a search engine. And it wasn't just better, it was way faster too. It felt like magic.
This is basically happen with almost every SaaS. They start with clean interface, performance while being used-friendly running by burning VC money to capture market. Then end up being user hostile products that locked in users to milk them before alternative come out.
TY! That was interesting and I learned that their first servers can be viewed in various museums. While web searching that, I found this interesting tangent, a Lego cased storage array from 1996:
I don't know...don't you always purchase domains for your personal projects as soon as you get excited about them and devise a name--and then sit (or build) on them for months before actually having something to launch?
Chromium/ChromiumOS and AOSP are some of the most valuable open source projects on earth. Without android, smartphones would be a luxury, thanks to android even poor people have one, in many cases their only general purpose computing device.
And 25 years in they more or less control the whole internet. With WEI they will be able to decide who is allowed to use the web, and with what software. It was good while it lasted.
If you think the world deserves some more innovation in the search space, email me at josh@sirch.org - we have a team (mostly ex-Meta/Twitter/Insta), an MVP, and early revenue.
It's certainly been interesting watching my opinion of Google change over the years from great to awful. Or maybe I'm just old and grumpy, being over twice as old as Google. :)
Lot of messages in that that picture. Elements of simplicity, couple ordinary people rolling up their sleeves and saying why not them, classic American dream. Attire is also interesting, I think more for them to feel like they are at work and not hanging around a friends garage.
If you've ever had to wake up in the morning and set your own direction, you'll appreciate the picture.
Google was cool but isn't now and haven't been any kind of cool for well over a decade. I recall dumping Altavista for Google and now I note I have recently dumped Chrome for Firefox and G for DDG.
It doesn't really matter: Google and co now has a hold over us proles that Big Brother could only dream of. In 1984, you had a surveillance screen installed. Nowadays, you buy the bloody things yourself (Smart TVs, home IoT cams etc). Also you carry them on your person - smart phones.
People who have entered the industry in the last decade or two may not realize it, but the sky high salaries, stock grants, benefits, flexible work hours, free gourmet meals, stocked pantries, gym memberships and lots more that you enjoy every day are all thanks to Sergey and Larry in the early 00s wanting to build a Disneyland for nerds. They were so successful at it that the rest of the industry had no option but to emulate them. Being a software engineer was nowhere near as enticing a job pre-Google.