I'll continue beating the dead horse: I don't trust Google.
At best GCP is slightly better than AWS (debatable), which is not enough to offset the risk because of how much of a pain it is to switch cloud providers.
Even if GCP shutting down is one in a million, aws shutting down is 1 in ten million.
Also, I keep getting the feeling that Google keeps trying to automate their customer service through bots. I have use quite a few google services from youtube, to google workspaces (a paid service) - I have never talked to a Google representative. I have talked to AWS service reps multiple times, I have talked to Amazon reps multiple times. Amazon as a company has real people you can talk to.
I don't trust google to not shut down my gcp account because my gmail was connected to some project where I accidentally sent a bunch of spam emails to a test account. And if that did I happen, I feel like I would just get an automated message and a bunch of hoops to jump through before I could talk to a real person.
> The group [Pichai, Page, and Porat] even talked about—and eventually dismissed—the idea of leaving the market entirely, this person said. [1]
That they’ve actually discussed shutting it down in the recent past lends plausibility to this risk. In 2019, they set 2023 as a deadline for moving into the #2 spot ahead of Azure. Obviously, they’ve not gotten anywhere near that.
Wow. Imagine top level execs at Ford discussing, but eventually dismissing--the idea of leaving the automobile market entirely. Their hearts really aren't in it as much as with AWS.
Google is an advertising company. They're planning an exit in a non-core market. It shouldn't be a surprise. The weight of "traditional google" has held zero value to me in measuring their cloud product.
Google's cloud products are built around the needs of advertisers. A large and slow backend that's meant to churn ungodly amounts of data.
Amazon is a retail company their cloud products are build around this need. They have a highly segmented and mostly nimble backend that's mean to build unwieldy amounts of API.
Ford, ironically, is an example of a company that has recently exited a huge business segment - they no longer make sedans. They now exclusively make SUVs and trucks. It is largely thought of as a pretty good move on their part.
Google probably should spin off GCP rather than killing it, but it's not a terrible idea to divest from a failing arm of a business. Google (like Intel) just does it too early.
This is a US/NA focused comment. Ford makes sedans in EU and other markets. They also make sedans for the US market in China and other manufacturing locations.
So it's true that Ford US manufacturing have exited sedan production, Ford itself continues to build and sell sedans.
I think this is an unfortunate conflation of free Gmail/etc. consumer products with GCP.
All of what you say is pretty accurate for free-tier Gmail/etc. No customer service, risk of an account getting blocked and nothing you can do, new apps not finding success and getting cancelled.
But none of it is the case with GCP. Customer service is great and you can reach real people easily, paid accounts aren't getting shut down without recourse unless you really are being abusive, and Google's not cancelling the services companies are paying for. (The same goes for the paid tier of Google Workspace.) It's a normal paid B2B relationship with all that usually entails.
It's unfortunate that people take their experiences and the stories they hear about the free consumer side, and extrapolate them to assume they're also true about the paid business side. It's understandable, but it's just not the case.
And neither GCP nor AWS is getting shut down. Even using the numbers you give, the answer is that the risk of either shutting down is zero for all intents and purposes. There's no reason to split hairs over which of two infinitesimally unlikely events is more likely, or use that as a justification for choosing one or the other.
> conflation of free Gmail/etc. consumer products with GCP.
* Google Domains
* Google Cloud IoT Core
* Jamboard
Also they increased Google Maps pricing by like 10x at one point, and they increased Cloud Storage prices pretty substantially last year too. Price increases can be just as impactful as killing a product. I don't think AWS has ever increased any prices or shut down any service (though quite a few are essentially zombies - SimpleDB, OpsWorks).
Then on the consumer side, Google have killed quite a few paid products. One that bit me was they shut down Google Cloud Print which killed the ability for me to print to my (cheap) printer from my phone. Besides that, there was Nest Secure, Stadia, etc. etc.
Google's reputation problem isn't just coming from their free consumer apps.
AWS just KILLED EC2!!! Well, EC2-classic. And they did so with two years of notice and support to transition all customers. Really a class act of migration. I'm not sure Google won't just send out an email on a Friday before a long weekend and then kill the servers the following Monday.
That one was such a long time ago when they were struggling to find PMF. I do my best to avoid Google bc I don’t trust them or their quality, but I think I’d give them a pass on the search appliance.
> I think this is an unfortunate conflation of free Gmail/etc.
I would also say that this is a byproduct of how google treats its customers, even on paid platforms. I used to manage Google Workspace few years ago (2016 -> 2020) and GCP (2018-2019) and I remember the difficulty I faced when attempting to reach a human. Even if I were to able to reach someone, I got someone who clearly was a tier 1 support, who wanted me to troubleshoot things by "turnings things off and back on again".
I've also seen blog posts (some even on hacker news) mirroring my experiences, so I know it's not just a "me thing".
But I am not a Google customer. I am a Google user. I use all sorts of Google properties, but i dont give them any money.
What most Gmail users call "customer support" is really "user support". Your free account has the durability of what you pay - zero.
So should we judge customer support based on user support? Technically no. Do we? Of course we do. We absolutely judge the unseen experience on other things we can see.
Quite literally, we judge a book by its cover.
Google has a reputation for bad support. It has earned this reputation from endless free users. It may not be fair, it may not be true, but it exists.
Equally it has a reputation for shutting down services. Popular services. Gmail is likely safe. GCP is likely safe. Am I gonna bet my business existence on "likely"?
Getting rid of https://domains.google is a good indication that nothing is off-limits and everything is a bean counter report away from being abandoned.
What's worse is that their loss-leading free products kill competing products before they finally abandon it themselves, leaving the space much worse than if they had ever entered it.
This really resonates with me – yeah, the supposed API-driven engineering company, makes it no longer possible to automatically provision domain names for its "valuable" cloud computing customers?
What possible end could that serve? I am baffled by the idea that succeeding via open access to their infrastructure is not a serious long-term goal for Google.
For a company who ostensibly is existentially interested in developing at least one meaningful complement to it advertising business, of all the things https://killedbygoogle.com/ over the years, why the domain registrar? The cost to developer trust vs. the cost of ongoing operations – did they not even have that conversation?
Now I imagine how I would've felt to have my registration transferred to – I can't even recall who bought their customers – and I try to imagine putting faith in trusting GCP products with a startup idea, where ostensibly their second-mover advantage could let me move faster in greenfield development ... and I throw my hands up. This is a company acting like it places no value on goodwill, not even from a bare bean counter perspective.
I guess I'll stick to AWS (and more bespoke cloud offerings).
> I think this is an unfortunate conflation of free Gmail/etc. consumer products with GCP.
Parent did point to Workspace where they also didn't get customer service.
To lay it first: GCP has excelent support if you're in the right bucket. For instance startup support programs will have dedicated engineers, sometimes on site, ready to meet and answer questions.
You'll also get that at the higher tiers (around where you start to negociate rebates in echange for usage volumes) with recurring slots with a support engineer to discuss your issues and growth inside the services.
But outside of the blessed situations, if you're just a mere paying user/business there will probably close to nothing for you.
And yes, education/corporate targeted products also get killed. Last month Jamboard's name was etched into its tombstone, along with Google Domains and so many others.
Even within GCP many products will get merged with more or less changes to adapt to. Stardriver becoming a different service with an unsearchable name comes to mind. Or the GA4 transition with the breaking changes.
I remember losing an argument with my company (that I cofounded) about using Google's awesome recruiting app that was part of GSuite. I said I don't trust Google not to kill it if there are less than a billion users on it. Everyone, and I mean everyone else said, nah. They'd be crazy to kill it. It's part of GSuite. They offer that to businesses and it's too core of a suite for them to kill it.
THAT was the name, ty. Google Hire. Yes, it was an excellent product. Leadership at the top needs to understand the cost in reputation and change the culture. Large corps aren't good at changing culture.
We had our main GCP account suspended because we were running a Lightning node, and some Google automata flagged us as mining.
We couldn’t get hold of any actual person at Google, and were told by our Google reseller to buy a fairly expensive support package to have our issue expedited after raising multiple appeals/objections. Suffice it to say, we run out of AWS now. I’ve heard GCP support (in terms of reaching an actual human support person/engineer has only gotten worse since then, and our experience occurred a good few years back).
This is kind of a problem with non-paid support I think. I had an AWS account with some credits that were going to expire and decided to use them to spin up a bare metal instance. My account was immediately flagged and I got locked out. I finally managed to reach someone who made me jump through hoops changing my password and rotating my API keys. It still took a few days before I was able to log in again, and by then my credits were expired. Luckily I wasn't doing anything important in my account, but someone running a business would have been screwed.
> none of it is the case with GCP. Customer service is great and you can reach real people easily
I'm curious how recent your experience was with GCP support? Sadly several contacts of mine were not long ago laid off from working there. Many of them had 10+ years industry experience. Presumably their roles were off-shored but from my experiences with other companies who have off-shored technical support, I don't imagine you'd get the same customer experience as you did before.
I worked at an ad agency for a time. We were spending millions a year with Google on AdWords.
We onboarded a new client who had many different locations serving different brands. We set up these accounts, set up Google accounts for the locations to own/access the AdWords accounts, etc.
Out of about a dozen accounts, one got suspended.
There is no general Google Support for that. So I contacted AdWords support. They put in a ticket for me with the accounts team. They also told me it's unlikely I'd ever hear back about it and advised me that my best course of action was just to register a new account and hope that wasn't suspended too.
Three years later when I left that company, I was still getting occasional emails to remind me that my ticket was still open but there were no updates to provide.
Meanwhile, in AWS land...
With my personal AWS account (~$3k/yr spend) I've never had an issue getting in touch with support that can resolve any issue I raise. I accidentally bought a bunch of reservations under the wrong account and put a ticket in and they swapped them to the intended account right away. I raised a ticket about trying to use a new product and running into some issues that appeared to be on their end and a day later got an email from the lead engineer for the product.
At a company spending ~$250k/yr, not only could I access _the exact same support_, but we also had an account rep that would swing by the office once a month just to check in. We had his direct email and phone. He was actually empowered to do things. We had an AWS key leak and ran up over $100k in additional costs before we caught it and our payment didn't go through. Sent him an email and almost immediately got back to not worry about it--he'd put a note on the account that no action was to be taken without his approval. Shot off a ticket to support about the spend due to a security breach, and they asked us some basic questions to demonstrate that we knew how to prevent this happening again and then cancelled over $100k in charges. (Current company I'm at spends ~$360k/yr on GCP and we have no contact at Google.)
This isn't "conflating google's free product support with their B2B support". This is "google's culture does not leave room for providing proper product support".
That doesn't put my mind at ease. Under who's definition? Do I even get told what I did wrong? You try to draw a distinction between the free and paid products but that doesn't address the perception people have of the company in general. If I was a small startup I wouldn't want to risk it.
I remember hearing a story about how one engineer had worked for something nefarious (spam? gambling? straight-up malware?) and got hired by a different company. Eventually something triggered on that engineer's account that caused Google to spread a blacklist to everyone's account at the new company.
Potentially apocryphal, but it rings true enough that you have no idea what might anger the Google automation.
It's not just "free" customer products. See: Pixel Pass being canceled right before it was supposed to pay off with a phone upgrade. Those people paid significant amounts of money.
I don't think GCP is going to shut down as a whole, but that doesn't mean I'd trust Google to keep any specific service I might depend on up and running.
> See: Pixel Pass being canceled right before it was supposed to pay off with a phone upgrade
Did you first try to understand what Pixel Pass was? No, it wasn't supposed to "pay off" with a phone upgrade. To get a new phone, users would have had to renew their Pixel Pass subscription. It wasn't - pay monthly for 2 years and get 2 phones: one at the start and end of the 2 years.
The basic thrust of your argument is equivalent to "Google shut down a product we were paying for at the end of a normal billing cycle, so how could you be mad?" Fair enough, but acknowledging Google will honor basic contractual requirements before shutting a service down doesn't really mitigate the argument that "Google might unceremoniously kill a paid product with minimal warning." Props to Google for not committing actual fraud, for whatever that's worth.
Having spent the past week dealing with GCP support, which I pay for, I can assure you, it sucks. It’s miserable and my least favorite part of GCP by far.
I used to share that sentiment about GCP until they shut down a service I was using (Cloud IoT). Now I try to avoid GCP even though I really do like a lot of their products more than AWS.
> And neither GCP nor AWS is getting shut down. Even using the numbers you give, the answer is that the risk of either shutting down is zero for all intents and purposes. There's no reason to split hairs over which of two infinitesimally unlikely events is more likely.
That's what I thought for a long time about google domains. Yet here we are.
Ugh. So it was first reported that Google Domains was being abandoned, but Google Cloud Domains was still just fine. I thought -- exactly, you can count on their paid enterprise cloud services.
I looked it up now to reconfirm, and it turns out a few days later it came out that Google Cloud Domains was also being shut down? [1]
WTF. OK, well that's idiotic. Maybe I have to take back my parent comment then. That's one of the dumbest corporate own-goals I've ever heard of.
That's not something I can even begin to explain or understand at all.
Google Domains is not being shut down - it's being sold to Squarespace. I would assume Cloud Domains either went with that package, or is being removed as part of the deal.
Selling to another company != shut down in my book. Although I'm already in the process of moving domains to another registrar...
Google, and all other domain registrars, are obliged by ICANN to allow domain migrations in case that the original registrar is shutting down. Getting money from Squarespace is just icing on the cake. You're sounding that this is a decision that Google is doing voluntarily, but it's literally not an option here (for once).
Also if they're still keeping the original Google system then it's believable that they are indeed just selling Google Domains, but the fact that all domains must be migrated into Squarespace's system tells a different story.
They are forced to sell it by their contracts with ICANN - this is not a decision that they have done voluntarily. If they can legally shut down Domains they would have done it.
Sure, in any other contexts, selling Stadia customers to Microsoft or Nvidia (for example) would be a voluntary decision, however they are barred from just shutting down this one. They can even shut down Gmail or YouTube with no legal repercussions (sure their reputation will suffer but legally it's in the clear), but Domains is in a different legal standing.
While chump change to the likes of Google, it demonstrates Squarespace views this business as highly profitable. Migrating an acquired business into your systems makes a lot of sense.
Unless you have some evidence it would have been just simply shut down otherwise, I think you're inferring things that do not exist. Few if any of the other "shut down by google" operations could have sold for anywhere near this amount, as they almost universally lost money and/or were niche.
And they're also shutt- I mean deprecated their enterprise version? (https://cloud.google.com/domains/docs/deprecations/feature-d...) That doesn't sound like "oh, we're selling this because it's valuable money" but more of "we don't have the energy to maintain this".
First, selling it to another company is just shutting it down but doing you the courtesy of migrating the data to another vendor. It still results in all of the management UIs, integrations, and APIs that you rely on going offline. It just avoids creating an immediate outage (but it will create one eventually, unless you start interacting with this other vendor).
Second, the point has repeatedly been made that GCP is fundamentally different from Google's other offerings and essentially immune from being shut down (even in the responses to this article). That may be true for GCP as a whole, but this proves that it's not true of the individual sub-components of GCP, even ones that seem extremely core.
I actually agree with the article that GCP is just good in ways that competitors aren't but... Running on their platform does continue to be a bit nerve-wracking.
As pointed out multiple times previously, Amazon and Microsoft provide domain registration services even if it's a solved problem. It's not that Google needs to be the best domain registration platform, but providing a normal domain registration service is miles better than not providing one at all (for example simplified billing and administration for smaller businesses, and automation for bigger ones).
Google, and all other domain registrars, are obliged by ICANN to allow domain migrations in case that the original registrar is shutting down. Getting money from Squarespace is just icing on the cake.
Google is selling Domains to Squarespace. Yes, many techies will migrate their domains, but majority of the people on Google Domains probably wont. That's why Squarespace thought it was worth paying anything for, after all.
Personally, I'm moving my domains already. But, saying Google Domains was shut down is inaccurate - it was sold, as business units/subsidiaries tend to be...
They are forced to sell it by their contracts with ICANN - this is not a decision that they have done voluntarily. If they can legally shut down Domains they would have done it.
Sure, in any other contexts, selling Stadia customers to Microsoft or Nvidia (for example) would be a voluntary decision, however they are barred from just shutting down this one. They can even shut down Gmail or YouTube with no legal repercussions (sure their reputation will suffer but legally it's in the clear), but Domains is in a different legal standing.
I’m not seeing why you’re being down voted. Indeed, the point being made (that Google would simply aborted all their domains if it weren’t for ICANN) is strange.
Of course they’re likely to not outright get rid of them thanks to ICANN. They could have however just given them away rather than selling them off. I feel like there’s some precedent there too? Maybe from Wordpress?
In any case, they weren’t forced to sell them. They had other options including _not_ shutting down. But even if they did merely give them to another company or sell them, people would have the chance to move their domain to another provider.
Really I just don’t get what point is trying to be made. Google shut down domains in every good faith interpretation of the phrase “shut down”. All the ICANN argument does is try to conflate “shut down” with “going against ICANN rules” but no one was ever suggesting that.
> In any case, they weren’t forced to sell them. They had other options including _not_ shutting down.
Fine, sure, but this is a cop-out. Clearly they really want to clear the thing: if it was to be a GCP-only option with no more consumer-focused Domains, they have temporarily set a transition period with no renewals and new domains and then bow out of consumer space, but even the enterprise version is being shu- sorry, I'll be using Google's term here, deprecated (https://cloud.google.com/domains/docs/deprecations/feature-d...). The Squarespace buyout is irrelevant here because unlike GCP Domains there are no automation features in the Squarespace's system (and have no plans to implement one), so you are required to migrate to another provider like DNSimple or AWS. This is a clear sign that they really want to dispose of it as soon as feasible.
> Really I just don’t get what point is trying to be made. Google shut down domains in every good faith interpretation of the phrase “shut down”. All the ICANN argument does is try to conflate “shut down” with “going against ICANN rules” but no one was ever suggesting that.
It's a reasonable assumption when you're talking about their first time in closing down things, but there is a clear trend, even solely in the enterprise space. (https://news.ycombinator.com/item?id=38020254) At this point, it is clear that Google does shut down things when they feel it.
Really? I've got associates at less than that (10-20k/mo) that have access to real people. One associate just crossed the $1k/mo mark and their human contact-ability improved. My GCP spend is <$500/mo - I don't get shit.
I had a Google Workspace account and a bizarre issue with my account thinking I was on a Caribbean island. I was a paying customer, and could talk to a human, but it was an awful experience.
>All of what you say is pretty accurate for free-tier Gmail/etc. No customer service, risk of an account getting blocked and nothing you can do, new apps not finding success and getting cancelled.
Dunno, they've still taken the time to implement all this crazy auto-management software for their free tier, so I'm fairly confident they'll deploy it from time to time on their paid stuff.
I did mostly agree with you that GCP is not Google etc etc until a couple weeks ago when they suddenly announced that they will move most of Policy Intelligence under Security Command Center Premium, throwing most of their customers security teams under the bus.
GCP starter earning a bad reputation for reliability, transparency, and support on day one. Maybe that's changed? I feel like a see a "GCK is hosed, google support is AWOL, our production environment has been down for 48 hours with no ETA" post on HN once a year.
Again, I'm willing to be wrong about recent reality, but GCP does not have a good reputation with its customers.
I work for a fortune 100 company. Years ago we looked at all the cloud providers trying to determine who to go with. Google was thrown out on the first day for this very reason. They simply cannot be trusted not to kill their products at the drop of a hat.
I was expecting rigorous analysis in terms of cost of services and the complexity/simplicity of offering, reliability figures, etc. but this fortune 100 company went with feelings
It seems like all the OP is saying is that the company evaluated alternatives (Azure & AWS) whose cloud products are "first class" citizens of the business whereas with Google the first class product is search, advertising, Android (a couple other things but you get the idea).
It doesn't seem too far out of left field that Google would flounder with all it's cloud products being phased out in favor of some other new product they've made or acquired.
Yeah we’ve had one too. Actually no two. Wait three. Actually I’ve lost count. In the 6 years I’ve been using GCP we’ve had a new rep practically every quarter. It’s ridiculous.
AWS is stable. In that, the DX is predictable. I'm using AWS for over a decade and had huge gaps in between but coming back never was shocked or surprised by things totally reshuffled.
On GCP, not the same. The GCP console of around 2012 if memory serves me well, is barely recognizable. That just the UX and I'm sure APIs would have changed as well in the meantime.
The hip product management culture of Google is beyond me. Today's cloud console would be something else tomorrow and I'm sure some hip and smart PM is chalking that out on their Chromebook or something. Sorry for the stereotyping.
As of speaking to customers, I bet AWS (real people) directly, routinely reach out to even 2 person startups. I have experienced it myself. They really do care. Even I had multiple hour long calls with Azure representative for an account who's monthly spend was a joke.
On the other hand, good luck with getting anyone from Google on email let alone phone.
I concur. Google doesn't seem to believe in a tangible customer service methodology at all. They go out of their way to remove human interaction at seemingly every level. I don't know if it's a legal strategy or what, but it gives me bad vibes too.
GCP has people you can talk to, you just need to bring enough money to the table for them to bother. In other words, the fat cats get support, the rest can suck it.
A good part of GCP is part of Core google infrastructure.
Spanner is used ubiquitously inside of Google, supporting services such as; Ads, Gmail and Photos. ... Spanner on the other hand processes 3 billion queries per second at peak, which is more than 20x higher [than DynamoDB], and has more than 12 exabytes of data under management.
https://cloud.google.com/blog/products/databases/announcing-...
Monarch: Google's Planet-Scale In-Memory Time Series Database, provides metrics for Google's own services and GCP Operation suite.
https://research.google/pubs/pub50652/
Zanzibar: Google’s Consistent, Global Authorization System - underpins access control across Google's product and GCP IAM.
https://research.google/pubs/pub48190/
I would put the odds quite a lot closer to 1 in 5 than 1 in a million. Google exec shuts down lackluster business ? It would not even be news outside tech/business circles.
it does seem like any of Google's support options for paid services have the primary goal of reducing support load on Google, not to solve problems for paying customers
No interest on engaging on the core issues of trust, except to say my standard bit: HN is a site predicated on the whole idea of embracing risk. Except with this one particular vendor who everyone loves to distrust. Maybe... it's not about risk at all?
This is a deeply weird take IMO. Nobody "embraces risk" because they have a risk fetish. You take on risk because you believe you can be paid for it, and perhaps you are better equipped to mitigate the risk than the people paying you to take it.
There is no payoff for using a Google product versus one of the competitors, so a rational person would not accept the risk that Google pulls the plug on a product, or some contractor gets banned by Google and gets your whole account frozen, etc. There's no upside. It's not even gambling because there's no "win" possible - it's more like shooting an arrow straight up into the air and hoping it doesn't land on your head.
> There is no payoff for using a Google product versus one of the competitors
The linked article is literally about how Google is better than its competitor though! Obviously some people disagree.
If you want to write a nuanced take about risk and benefit, I'm there for it[1]. It's the stunningly absolutist take on something as clearly-not-a-big-deal as a cloud service provider, by people who would otherwise be willing to bet their careers on a job with a good vibe, that seems "deeply weird" to me.
[1] Actually I'm not. That's stuff's boring. But I respect that people do it.
There’s a stark difference between risk that is predicated on your own wits and risk that is mediated entirely through a capricious and aloof supplier with a bit of a reputation for the rug-pull.
Nonetheless the chances of a median HN poster's employer being shived by a rogue Angel or dying to a founder feud are vastly (like, vastly vastly) higher than having Google drop one of their dependencies. And I guess I'd expect the discourse to reflect that. But instead it's the boring tech giant (yes, also my employer, though I work nowhere near cloud) who everyone seems to think is the existential threat? Like I said, I don't think this is about risk at all.
At best GCP is slightly better than AWS (debatable), which is not enough to offset the risk because of how much of a pain it is to switch cloud providers.
Even if GCP shutting down is one in a million, aws shutting down is 1 in ten million.
Also, I keep getting the feeling that Google keeps trying to automate their customer service through bots. I have use quite a few google services from youtube, to google workspaces (a paid service) - I have never talked to a Google representative. I have talked to AWS service reps multiple times, I have talked to Amazon reps multiple times. Amazon as a company has real people you can talk to.
I don't trust google to not shut down my gcp account because my gmail was connected to some project where I accidentally sent a bunch of spam emails to a test account. And if that did I happen, I feel like I would just get an automated message and a bunch of hoops to jump through before I could talk to a real person.