Google is the new IBM::Years of being one-upped on AI and cracking down on innovation turned the poster child for Silicon Valley cool into a dinosaur.
If you’ve ever interviewed at Google, you know why this is happening. They hire people who are as much like the people they already employ as possible, to the point that employees don’t know who they’re interviewing or even for what. The person getting hired is pre-screened for all sorts of “desirable traits” before being matched with a team. The people who succeeded there all think the same, and so they all end up having the same ideas, and the number of novel ideas nose dives.
IBM has an internal motto that they really push when you get hired there: “Treasure wild ducks”. Beyond the regular buzz word bingo of 'think different ’ and ‘move fast break things’ it means “when someone else has a crazy idea that might just work, fucking listen to them”, and it’s what’s kept them in business for literally a century. I don’t think Google has that fundamental non-self-centered DNA. Every product they’ve ever put out was a result of their intellectual monoculture and the hyper competitive mire of sameness it breeds.
I’m an ex-IBMer (For 3 years, recent years) and never heard of this motto. Not saying it is a lie, just that it might not be as widespread.
But I agree with the commenter. My opinion is that IBM went from being run by engineers to be run by “used car salesmen” that care little for the tech and much more for the end of year bonus. I’ve seen leadership pushing fron clear multi million undeliverable projects just to get their signing bonus and bail.
Google seems to be in the same direction and this is a consequence of its own size.
I think there is a research opportunity on those big tech companies to generate another excellent theory like “Innovator’s Dilemma “
Never heard that phrase used there either. But uh… can confirm the rest. It’s a bummer to say the least.
This is a great overview of the problem, I never even really thought about. It kind of really explains why they’ve binned and remade almost the identical apps multiple times. Maybe if they’d made more QOL and supported their apps with new novel ideas they wouldnt have slowly died out.
Or even maybe if they’d make privacy respecting software instead of spreading their legs and whoring out all the data they collected on you they’d be doing much better too. Google has become a bloated beast that needs to be put down. Anti competitive and monopolistic behaviour with no innovation, it’s a wonder anyone uses their shit apps and services anymore
Google targets ads they deliver based on the data they collect. They don’t “whore out” the data. That’s just a lie that’s been repeated so often people take it as gospel. I’ve never seen a shred of evidence to support it, and when I worked there, the employee training everyone took annually was very, very clear on respecting user privacy and getting everything reviewed by privacy experts before it could be released.
Ex-IBMer here too. I only heard the motto twice - once when Ginni visited our location, and once when local departments were closed.
Coincidentally there was never any money for me to work on interesting ideas…
I get what you’re saying, but IBM is not a poster child for what Google or anyone else should be doing. IBM has been an also ran for a very long time; huge in size and very profitable with occasional innovations but a fraction of what it was when it was in its heyday.
IBM stayed at the forefront for so long because the barriers of entry into the computer sector were too high. But in terms of innovation Microsoft demolished IBM, and Google demolished Microsoft. Now Google is just another big incumbent, very profitable but unable to innovate like it used to and instead beholden to shareholders short-termism just like Microsoft and IBM.
The bigger problem is these companies could stifle innovation through sheer scale and market clout, perhaps to a detrimental degree in the US in particular going forward. They try and mop up all the talent and then put them to work in dull areas. For example Google is largely just an advertising company that dabbles in other things now.
While I agree that this is a risk, I sincerely don’t believe this happens often when interviewing at Google.
For one, employees are continuously reminded to avoid bias in anything that they do, from the way in which interviews are conducted to the design of products.
Googlers are reminded to avoid this on a nonstop basis through annual trainings or even artwork and signage throughout Google offices that target specific bias-awareness programs. In the restrooms, posters with detailed recommendations, often tailored towards engineering, make for an educational read while you’re doing your business. Screens at the cafeteria show prompts challenging you to rethink assumptions. Dedicated teams are involved in performing reviews of proposals and code solely from the perspective of inclusivity.
I’ve never seen anything regarding “desirable traits” as part of a job listing. Hiring managers provide a job description that is reviewed to avoid bias, and pass along specific requirements for education and professional experience to recruitment teams. Recruiters take a first pass at CVs for those, and I’m honestly not sure how some kind of personality trait could even be distilled from a CV. Once a candidate that fulfills the minimum requirements is matched, they are set up to discuss other requirements with the recruiter, like relocation and timelines. I don’t recall from my own experience ever being asked anything aside from these practicalities.
For interviewing specifically, there are multiple steps needed to qualify as an interviewer, each of which puts a heavy emphasis on avoiding bias. The interview question itself needs to be vetted by a dedicated team and interviewers usually select their questions from the pre-vetted ones. Prior to performing your first interview you need to be doubly shadowed with topics like avoiding bias in mind. When asked to perform an interview, the details about the role that the candidate is applying to are provided and the interviewer is required to review the CV themselves ahead of time. As evidence of this, you’ll see that the interviewer will often match items from the CV against the listing to give the candidate the opportunity to expand on it and offer more detailed insights.
Rating the interview is performed within explicit rubrics, each of which with detailed descriptions. There’s not an option to simply reject a candidate—interviewers need to select options from these rubrics and provide evidence. This is in part why you will see interviewers vigorously taking notes during an interview.
The first phone screen has more relaxed requirements as a general confirmation that the candidate exhibits the skill level expected at the listing’s minimum requirements.
There are at least four in person interviews that then follow, performed by different interviewers. These results are reviewed by a hiring committee who makes a final decision solely based on the evidence with no insights into the associated candidate.
I have personally never worked at a company that is so meticulous in avoiding confirmation bias. In one smaller company that I worked at, I was the only interviewer and the sole decision maker for a candidate. Honestly, I cannot envision how Google can do better than they currently are with hiring.
I believe that the frustrating thing about getting hired there is simply the high bar and disparity between the high supply of candidates and the relatively low offering of positions. When you’re prematurely rejected after submitting your CV or you’re rejected after interviewing, remind yourself that you aren’t necessarily unqualified or that the interview was unfair, but that many qualified candidates might have already applied, or the head count may have been removed, or an internal transfer took place or some other reason unrelated to your skillset.
Holy shit, TIL IBM has been around since before WWI.
what’s kept them in business for literally a century
- find a new adverb
- no, it was buying other companies and sucking them dry. Go read some Cringely.
What can you expect from a company that in 2024 does not yet have a Linux client for Google Drive?
https://abevoelker.github.io/how-long-since-google-said-a-google-drive-linux-client-is-coming/
To be fair, they barely have a Windows client. I’m constantly having issues with it
Just use rclone. Its bidirectional sync is kinda meh last I tested it, so I do manual syncs in each direction. Otherwise its awesome. Can even encrypt your stuff with your own key.
Supports a bunch of backends. There is an androind client called Round Sync with cron-like scheduling to keep my phone backed up.
Do you need it though? i feel like the linux userbase is already fairly low, and the intersection of people who cant do a RW mount with rclone ans uses linux is even lower.
You would be pouring a bunch of money into a development for the 0.001% Userbase
You can tell how passionate a company is with their products by their Linux support. That means no one there cares enough to push hard for Linux support. Even Dropbox has a Linux client.
Valve’s commitment to Linux is why I’ve consistently bought games virtually exclusively through Steam.
I think at best, what you can tell is how many developers Maining linux are on that product’s team.
Even Microsoft OneDrive…
There is an official Onedrive client for Linux?
Last time I checked there wasn’t and you had to rely on a third-party application, based on MS API (͡•_ ͡• )
You’re absolutely right, official version doesn’t exist. The closest thing would be this: https://github.com/abraunegg/onedrive/ My brain was confused with Mega’s excellent client. SORRY
OneDrive
We call it OneChive
I recently did this, and it was fucking annoying to create the app in Google’s Cloud. Incredibly laggy (5-10 seconds until clicks register), loading times of up to a minute between navigations.
This was on a very beefy PC, I suspect the issue is that I used Firefox.
i feel like the linux userbase is already fairly low
Linux desktops now outnumber mac desktops, apparently.
Only if you don’t count mac laptops as “desktop”
“desktop” in general is a much smaller number than it once was, since laptops can do so many of the things people once used desktops for.
Even if what you said was true, i think the better choice would be to go in the direction of simplicity, not the direction that favours segregation.
I mean, excepting that google isn’t even really the main offering when it comes to institutional compute (that’s Microsoft/ azure).
Like IBM had mainframes and legacy infrastructure on lock.
The only thing google really has on lock still is gmail, but honestly, take it or leave it.
They had search, but I get better answers asking a space heater to hallucinate a couple hundred characters for me these days.
:Android stares intently: “Am I a joke to you?”
Considering Android hasn’t fully implemented DHCPv6? Yes
i left it, tyvm
youtube is the only product of theirs i use regularly.
and android if you count lineageos as “theirs”
Google has YouTube
ChatGPT is nowhere near being able to replace search, and even if it was remotely similar, many people don’t even really know what it is, whereas Google is ubiquitous with search.
To say they only have a lock on Gmail is doing them a huge disservice. They own a huge part of online advertising and search.
For Gmail if you mean a lock on consumers who generally don’t pay for the product I would agree, but I have done more g-suite to Office 365 in the past 12 months than I have in the past 5 years. It is too bad because we could really use some competition and different ideas in the office productivity space.
I blame Sundar… Wayyy too many duplicated projects (e.g. Allo) and projects terminated too early (e.g., Stadia) under him.
Stadia’s shutdown reallly pissed me off. The problem it had was the monitisation, not the actual thing it did. Stadia worked in places with crappy wifi, like the 2.4ghz only I had at my Mam’s house, when GeForce Now, XBox Cloud and Amazon’s Luna, all shit the bed. Really well optimised, it also worked at higher quality than everything else when you actually had a good connection.
If they’d actually built out the infrastructure properly, had all the features, like being able to play a game via youtube after watching a video on it, and the quasi split screen thing, it would’ve done a lot better. It also needed a bit of time, which Google seems hesitant to actually give any of it’s projects.
The main reason stadia failed is because they have cancelled so many projects before stadia that people were taking bets on when stadia would close before it even started.
No one wanted to buy into a service that was going to shut down and they created a self fulfilling prophecy.
Essentially all new Google projects wikl forever be doomed to this fate.
It failed because they didn’t give it enough time to succeed. Google had enough money to invest in Stadia at a loss for many years until it eventually succeeds, but instead they just confirmed their image of randomly cancelling products.
All true. The monetization wasn’t even that bad, it was more so the marketing. Lots of people didn’t know about Stadia or were against it because of the bad launch they had.
I think the service would’ve done far better had Google made some guarantees like “all your purchases will be refunded if we shutdown in the next 10 years” and then ran a new ad campaign for it.
I ended up getting a Stadia Premiere Edition (Controller and a Chromecast Ultra) for £20 down from £70 and tried it for essentially free (Stadia Pro 12 months free).
The got £20 refunded when they wound down - announced 3/4 of the way through my Stadia trial.
So all free and got a long cabled USB-C charger, a newer Chromecast than my old one, which ended up on the 2nd TV, and a controller I can repurpose.
Wild.
I got a Chromecast on sale for like $15 last year when I read that they could be rooted.
This was not he case. I have get to use it.
Stadia was neither open enough to excite nerds and build a tech community nor polished or straightforward enough to go straight to pure consumers (notpowersuserss).
That’s my take at least
I read some days ago that google fosters a sort of innovation culture, that encourages their developers to create new stuff instead of putting effort into existing things. Which is also why so many of their products seem to be just rebrands of older ideas they had.
Google was like that before Sundar.
Does that make OpenAI the new Google?
And does that make Mistral the new OpenAI?
Didn’t their exhaust fans used to catch fire?
No, that was Sabre and their printers.
I was partly wrong. It was a pedestal fan not an exhaust fan.
Mistral are the new M$.
I think it makes open source the new Google.
Fight me.
The only thing I can give you for that comment is rough sex.
I’ll take it.
Uh ibm’s not gone…
Nope, that would make it a very stupid analogy.
But IBM is just a tiny shadow of their former glory, of almost complete dominance of business computers of any size.
According to the article Google/Alphabet is losing their leadership position too, in much the same way.Might as well be, it’s not really IBM anymore.
They have a stranglehold on enterprise computing after they ate redhat, and they still make insane mainframes, they’ve just left the consumer world (which makes sense given their name i guess)
I wouldn’t go that far, there’s been a mass exodus from Redhat and they’re hardly the only game in town when it comes to mainframes. You do have a point though.
The mass exodus from RH has been massively overstated. It’s mostly a bunch of Redditors and Lemmings saying “omg I’ll never use RHEL now, even though I never have”, but in reality, they’ve not seen an exodus.
I do think their actions have a good chance of causing damage in the long term, though.
Maybe, but our data center people decided to switched to SUSE so some places definitely ditched RHEL.
How did that work out? We used sles in the past (moved to rhel6). Management of larger environments has been easier with rhel, but we’ve slowly been decoupling from redhat-isms. Satellite is just doing drm -the only thing that gives us grief- and repos now.
Mainframes are still the shit in the world of financial transactions.
Cries in COBOL.
Why cry?
That’s what the C in COBOL stands for.
Cry-inducing Old Business-Oriented Language.
As someone who runs a fleet of series i machines running as400, they’re doing just as well today as they were 40 years ago
Norways it’s mostly a software company. IBM people keep talking about cloud, AI and all that stuff.
It’s a trend-hopping company. Always 1-2 years behind actually valuable topics, but on the forefront on any useless bullshit you can think of.
I’d compare them more with Microsoft from 10 years ago.
This is the best summary I could come up with:
But the past few years have introduced new troubles: lower tolerance for risk, crackdowns on innovation, layoffs, and a narrative that its famed products like search and Gmail are getting worse.
In 2012, Business Insider listed 10 reasons Google was “the greatest company in the world,” including that it “made sick perks standard for startups” and created “a little something we call ‘Google Glass.’”
“In the technology industry, where revolutionary ideas drive the next big growth areas, you need to be a bit uncomfortable to stay relevant,” Page wrote in a memo to employees at the time.
A Google spokesperson noted that employees were still encouraged to pursue other projects, and pointed to work like AlphaFold and quantum computing and products like Magic Eraser as examples of innovation.
Still, X — a moonshot lab that birthed Google’s self-driving-car unit and explored exoskeletons and space elevators — has also curtailed its ambitions as it faces increased pressure to reduce losses.
Rivals like Meta and Microsoft have turned their fortunes around in recent years with more-decisive, top-down leadership, forging a cultural compromise between unconstrained innovation and corporate efficiency.
The original article contains 2,475 words, the summary contains 186 words. Saved 92%. I’m a bot and I’m open source!
So does this make the old IBM the new Novell?
Ha.
There’s a bank in China that avoided total pownership because they are running novell.
Mostly because the malware wasn’t expecting it. who would?
Removed by mod
Whats a novell?
I’m glad to see that something I’ve been raising for about 18 months now seems to be coming true.
IMO this is happening all over the tech industry. I work for a competitor, and the shift over the last year or two has been seismic. Back when Google was the king, new engineers wanted to build “the next Google”, and startups were where the magic happened. The big tech companies saw this, and they hired the best and brightest en-masse to work on moonshot ideas. For the last 5-8 years, you could work for Google, Amazon, Apple, any of the big tech companies in any number of industries - if it worked out, your career would be solid, if it didn’t you’d move on to the next thing while keeping your job. All of this was secured with great salaries, freedom of movement to live/transfer wherever you want, and job security (assuming you’re not at Amazon).
Now, not only are the moonshots gone, but also the following:
- Scrutiny for all hiring, with established teams struggling to secure any headcount to improve, or even maintain due to previous layoffs.
- A push to do more with less, often resulting in obvious enshitification to boost metrics or make more profit at the expense of users.
- A focus on speed over completeness - with companies favouring technical debt, hiring externally to get people in seats fast, etc. You could be a strong performer internally, and you’d be overlooked for an internal role because someone with far fewer credentials can be brought in immediately from outside.
- Zero job security. Layoffs have been rolling for over a year now, and with companies that PIP, you’ve got the worry of both being laid off AND being one of the 5-15% of people that are fired to meet performance quotas.
- Leaders taking zero ownership, or skirting the culture and rules of the company, like with Amazon going on CEO “gut feeling” while expecting others to use data, or Google basically shifting towards being “ungoogley” and “evil”.
Outside of pay, the benefits of working for a big tech company are gone. The innovation is happening elsewhere, and these companies have purposely bled talent to appease shareholders. Students don’t want to prep for months for a job that’ll fire them months later, especially when opportunities are limited. Finally, no one wants to work on things that don’t generate profit - especially when whole orgs are laid off or shut down, all while the leaders that set the direction to fall off a cliff are parachuted into roles elsewhere in the business. That last one is key, because tech employees now look at VP+ level moves as a sign of them losing their job, another distraction from being able to do your job or caring about your output.
So yes, Google will be the new IBM, and I don’t think there’s a way back for any of the FAANG tech companies. They either course correct through new leadership and focusing on their talent/prouct again, or they become relics of the 2000’s and crumble when their share price inevitably drops.
Very interesting article, learn nothing new but it’s great to put things in perspective and try to understand why Google fcked up so badly.
🤦
I’ve noticed with android everything is a Google feature, android is essential a closed source operating system
MS the new IBM
The new who?
Yeah, his name is Ncuti Gatwa