Also from that thread, a Google engineer saying (out of context:) "We want our users be able to depend on our features and services, and if you can't do that, we're letting you down."
How long before Google realizes that these casual shut-downs of side projects are harming their image tremendously?
The Google i18n font repo (https://github.com/googlei18n) went private all of a sudden in the last few weeks and it broke my Yocto build (and hey boot2Qt team, if you're listening, you're broken too).
I've been steadfast in not adopting any of the Google toy side-projects (e.g. Brillo, Weave, Things, etc etc) but now I'm expanding that rule to anything and everything from them at all, including AOSP. It's all quicksand.
Aha, I see, the original URL in the HN comment was the org URL, which didn't redirect. https://github.com/googlei18n
I don't know why original poster was linking to that!
I misunderstood, indeed. Kind of funny that a bunch of other HN discussion then took off about how it was evidence that Google as an entire giant company was now terrible! Ah, HN.
Talking to Googlers nowadays, I think the bar for who gets to be a Googler has dropped significantly, but assumptions of prestige from without have not changed over time. Googlers are just people and now that their hiring standards have shifted we can expect to see even Google making junior-level mistakes.
More likely that Google engineers are in a bubble because of all the available internal tools. Also the interview process nowadays select people that have time to train for the interviews rather than good engineers. The Google interview is still hard, but not for the same reasons than in early 2010.
This. Their interview process selects people who have hundreds of hours to blow to practice on stupid competitive coding, aka LeetCode. These aren't people who are innovative, they're ones that know how to grind.
Same process used by Amazon, Meta, etc. And then they wonder why there's no more innovation at any of these places.
Googles strategy is to find innovative people who can code well, but it's way easier and faster to determine the latter. So they hire a TON of people with these interviews and then wait for the sharp and innovative ones to float up while the not so innovative ones stay at L4.
Whether that is working or not is harder to say, but i can at least say that high level googlers have all been quite good in my experience
That’s not it. Googlers were good when Google was a startup with a high bar of entry trying to change things. Nowadays Google is a prestigious company. They are attractive to people who are first and foremost looking for a prestige job and don’t get me wrong they can be quite good but generally that’s not very interesting people.
"You can divide our industry into two kinds of people: those who want to go work for a company to make it successful, and those who want to go work for a successful company." —jwz
> I think the bar for who gets to be a Googler has dropped significantly, but assumptions of prestige from without have not changed over time.
The former is obviously true, if you go back far enough. Aiming for concentrating talent, the bar you can set if you need 10-20-30 engineers a year is wildly different than what you can do if you need thousands. Same applies at team level; it's possible to put together a top world-class team of 10 people, a struggle to put together 100, and beyond that order probably not going to happen, for a host of reasons.
The latter though, doesn't seem to be true at all. I don't think I know anyone in tech who holds google technically in the same regard they were held in 15 years ago as an organization. Individual teams, sure.
I think the problem is Google is significantly less productive with their thousands of incompetents than they used to be their their dozens of professionals.
Google does not need thousands of engineers, they simply don't have enough financial pressure to stop themselves from hiring dead weight.
Google's organization is often described as "thousands of startups under the same roof". So each team has ownership over their product(s) and each team has the room for advocacy for their product. But if their product is deemed unprofitable or whatever by higher-ups, I assume they start experiencing pressure about that. Ultimately though I'm sure budget concerns is king, and if a SVP needs to hit some quota for revenue or whatever and your product is relatively niche then they may be able to make the argument that your time is better spent working on something more "productive".
CJK. Like a sibling comment said, a redirect would have worked but it looks like there are other issues.
Right there in the About panel: Noto fonts, except for CJK and emoji
If there's some technical/licensing/political issue about the font sure, whatever, I get it. I'm just saying it's all brittle and obviously a lesson that I need to internally fork every. single. thing. when working with Yocto.
It sucks that Google did that, but if you're relying on a third party lib for something important/in production, why wouldn't you have your own mirror?
In Yocto development I typically rely on external OSS repos to achieve the first builds but, yes, in production one can lock down the source in a local mirror.
Google fonts were already ruled a GDPR violation by a court (at least when loaded remotely), at this point it's probably a good idea to avoid those too. And probably anything else from Google with a reasonable alternative.
This is a very astute observation, and one that I've seen in practice all too often. Thank you for this remarkable insight, I will definitely put it to use.
They don't care because they're not incentivized to care. It will remain like that until maintaining projects well is at least as good a path to a promotion as creating new projects.
> It will remain like that until maintaining projects well is at least as good a path to a promotion as creating new projects.
But then there's equally a danger of becoming IBM, which would be worse. (In this case, if Google creates something that people need then pull it, then other open-source people can replicate it).
I think the sane middleground is for them to clearly identify what's experimental, what may be discontinued/deprecated/pulled, and to communicate those expectations and timeframes, in advance. Even Yahoo did this.
> I think the sane middleground is for them to clearly identify what's experimental, what may be discontinued/deprecated/pulled, and to communicate those expectations and timeframes, in advance.
That might not work well for Google specifically, thinking about how long GMail was branded as "beta". Also, much of their consumer-facing things need network effects (think their piles of messengers)…
I think the people making this sort of decision ("Should we kill X minor feature of Y product?") are quite a long way down from the top, and probably were engineers themselves in the past. This isn't a "pointy haired boss" level decision. It's a trivial feature in a core product. It's much more likely it just needs someone to own it and no one wants to.
How long until people get it through their think skull? You are not the customer. Google's customers are corporations advertising and spying on you for the US government.
Stop using their services, stop giving them free data. Stop giving them access to your life to sell to the highest bidder.
There's no incentive for people to do that: things like maps, search, and gmail are fantastic products, and on the other side of the equation it doesn't negatively impact my life in the slightest to have google profiting from having access to private information about my life. That's the case for the vast majority of people; their private lives aren't interesting enough for there to be any reason to avoid selling them to tech companies in exchange for excellent products. So it's quite deliberate; it's not a stupid mistake that people are making due to lack of thought processes in their thick skulls. On the contrary, it's the population of people voicing your opinion that is making the mistake in their risk analysis: you are both wildly overestimating the probability that something genuinely bad will happen and also overestimating the extent to which your private life is interesting to anyone else.
That's pretty much how I've always seen it.
Though I suspect a very large % of Google product users don't even give it a second thought that if they're getting something for "free" they must be paying in some sort of hidden/indirect manner.
I'm actually a fan of targeting advertising too - when it's done well it can genuinely be a useful sort of information that can assist in deciding what products to buy. What I don't get is why it's so often not very smart (e.g. getting tonnes of ads for a product that I literally just bought and couldn't even have received yet given shipping times).
When they have lost revenue enough quarters in a row that they have to shuffle and start doing business differently.
Until something big changes they reap record profits off our data and people continue to be the product while their customers continue to be the advertisers.
Even if that happens, it's not going to change the core problem.
They'll just what microsoft did, and say "look, we are good people now". The world will forgive and forget, and use their product hapilly. But because the underlying cause hasn't change, only the context, as soon as the context will change again, we will be back to square one.
One one hand, sucking the life out of the web ecosystem and then killing their vampire implementation is really bad.
On the other hand, the alternative to shipping a DOOMED implementation is Google not shipping these in the first place, which people also don't want.
The least the company could do to not destroy the ecosystem would be to open source or at least write up these implementations for someone to rebuild black-box.
Realistically, any developer could write a blog with black-box reimplementations.
An alternative is to keep a team explicitly for life support of these projects.
Brave search still works as well. Honestly, non search engine alternatives really are not equivalent. A timer is not important enouhgt for a bookmark or desktop icon.
The DDG one works okay but does not update in the page title like the Online Stopwatch one does, so it's a bit less convenient to quickly check how much time is left (or if it's expired and you didn't hear the sound) when the tab is somewhere in the background.
It does have a better UI and it's nice that it supports running multiple concurrent timers though.
Ideally it would support sending a browser push notification so you know the timer is done even if your volume happens to be on mute.
There's an iOS and macOS app called Due [1] that has all the bells and whistles if anyone else uses a lot of timers regularly.
I’ve already stopped relying on Google. It’s too hard to tell the difference between a product they are going to support and a product they aren't. I don't want to have to pull up some financial statements just to understand if google is going to cancel some product I'm interested in within 2 years, I'm going to save my time and just go with somebody else.
I honestly wonder if they might be able to solve the issue with branding.
> "We want our users be able to depend on our features and services, and if you can't do that, we're letting you down."
That kind of ethos used to exist at Google circa 2004 and was the way the majority of engineer and the few product manager behaved.
Very few of these folks are left and are a very tiny minority.
Most folks at Google these days don't give a rat's ass about that kind of thinking nor about the company itself, they're just here for the GSUs and the free food.
Some would argue that these "failures" due to "market forces" are simply an excuse, a ruse to cover up their monoplies elsewhere (i.e., search and ads).
Give how often they sunset things it often feels like a solid argument.
https://www.online-stopwatch.com/timer/1hour/
Also from that thread, a Google engineer saying (out of context:) "We want our users be able to depend on our features and services, and if you can't do that, we're letting you down."
How long before Google realizes that these casual shut-downs of side projects are harming their image tremendously?