10 ms·

This is the entire problem. Swap “Google” for “AWS” in the article and you’d have legions of folks calling foul on the article. Instead, most folks are nodding

This is the entire problem. Swap “Google” for “AWS” in the article and you’d have legions of folks calling foul on the article. Instead, most folks are nodding along to the article going “yep, seems like something Google would do all right…” Google’s problem remains its reputation for turning things off.

james_s_tayler 7y ago
Also AWS is just the best at marketing their services. They did what Google did with search. We always say we "Google" things. Developers these days say "Chuck it in S3" or "Run it on EC2". I feel they have a strong advantage because they've kind of come to own the language of the cloud.
- JimboOmega 7y ago
  Actually, most of the time we say "in the cloud" rather than naming a specific hosting provider. Definitely Amazon has the first mover advantage and accumulated reputation first, but it's hardly unassailable. Still... I think the bigger problem isn't whether or not Google plans topull the plug or not, but that it has a reputation of pulling plugs to such an extent that people factor that into a decision of using their services. When it's something as big a choice as a cloud provider that can cast a long shadow over their offerings.
  - sdan 7y ago
    In addition, I remain cloud agnostic as well. As a student, I stick to whichever platforms can give some free credits. At the moment I have some cloud infra on google and some on aws.
    - dragonsh 7y ago
      That’s a good thing not to tie in to proprietary api’s, indeed every cloud provider is doing it and after Google’s recent management shuffle to bring in management from Oracle making them alike. Try using apache-libcloud to make your deployment and programs cloud agnostic. Also try using smaller alternatives like hetzner or scaleway. Once you go down this cloud rabbit hole you will be hold hostage to whims of these providers, where they can shutdown any service at their will or ask to pay an arm and leg. Also try to have non USA cloud providers as another backup option in the long run.
- mindcrime 7y ago
  Developers these days say "Chuck it in S3" or "Run it on EC2". I feel they have a strong advantage because they've kind of come to own the language of the cloud. After many years of working with AWS almost exclusively, I recently started a new $DAYJOB where our primary public cloud provider is Azure. A couple of days ago I was chatting with a co-worker about something, and he literally says "I think we should just chuck it in S3 and call it good" (paraphrased slightly, due to my imperfect memory). It struck me as odd for a few seconds, and then I realized that exactly what you are talking about has taken place. S3 is so dominant that it's become the default label for object storage in the cloud.
  - kevinventullo 7y ago
    https://en.wikipedia.org/wiki/Generic_trademark https://en.wikipedia.org/wiki/Generic_trademark
    - mindcrime 7y ago
      Exactly.
- sidlls 7y ago
  I don't think I've ever heard AWS-specific terms applied to GCP or Azure infrastructure.
rhizome 7y ago
>Google’s problem remains its reputation for turning things off. I'd put it this way: Google has a reputation for turning things off, but its _problem_ is an inability to stop doing that, to commit to its users/customers. In other contexts this could be written in terms of addiction. If I were Sergey and Larry, this would be the thing I'd be retiring over, and I'm not sure Suchar can change it.
- adventured 7y ago
  If Satya Nadella could get Microsoft to move beyond its Windows-all-or-nothing pathology, someone in charge at Google (including the two founders that vote-control the entire company directly - they can de facto hand Sundar the power to act as he needs to) can trivially put a stop to the relationship commitment issues at Google.
  - throwaway2048 7y ago
    Its far from trivial to change an entire corporate culture on a dime. From everything we have seen of Google's culture, working on new projects is exciting and is absolutely the way to get promoted. Doing maintenance work, especially on products that aren't wildly successful gets you no attention, no staff, no funding, and probably more than a little tongue waging. The CEO cant just change all that with directives, even if they do something extreme like a "No Depreciations" policy, products are liable to be depreciated in all but name (Joe the intern takes care of it every 5th Tuesday, technically not depreciated) by sheer corporate momentum alone, nobody wants to be stuck with a dead fish, even if its now illegal to openly comment on how much it stinks. It would require change at every level of Google to place value into fixing/improving existing projects, rather than all the glory going to greenfield development. Considering the kind of "kingmaker" culture Google's mentality breeds, ie what gets me ahead is gathering resources to focus on my project, changing things to a setup where contributing to other peoples silos is welcome and common is going to be an extremely difficult prospect because virtually all the people with power inside Google got that way by carefully tending their fiefdom, they aren't just going to throw that all a way, no matter what any executive says.
sundvor 7y ago
Another perception is that of killing hierarchies of user accounts / pulling rugs on projects e.g. if links are established between adding someone who once broke the rules and whatever they might be working on now, with no way of reaching someone human for support... ... where one of the few ways of rectifying the situation is to get a top posting on HN in the hope that a Googler will see your plea for help. I mean, that's probably mainly related to Android dev but I've seen plenty of stories to support that feeling: E.g. (just a random example) https://news.ycombinator.com/item?id=17431609 https://news.ycombinator.com/item?id=17431609 My perception might be misguided / influenced by being in a perception bubble, however that is my reality and AWS or AZURE just seems a whole lot safer. So, why should I as a developer invest any time in GC? I did attend a Google dev day re machine learning++ in Melbourne a couple of years ago, and the products seem nice and all, but the competition isn't exactly lagging behind. My recommendation would be to really address this issue of human support and trust.
res0nat0r 7y ago
The issue is that folks here think GCP is the same thing as Google Reader. That's absurd. Google is currently building a massive $500 million datacenter outside of Reno as we speak, and has 10+ billion invested in their datacenter cloud offering buildout this year alone. GCP and the entire physical infrastructure isn't going to get "turned off" if it isn't #1 in the market in 3 years, that's just silly. https://9to5google.com/2019/07/01/google-data-center-nevada/ https://9to5google.com/2019/07/01/google-data-center-nevada/ https://www.rgj.com/story/money/business/2019/02/13/google-investing-13-b-data-centers-2019-including-nevada/2861826002/ https://www.rgj.com/story/money/business/2019/02/13/google-i...
- dana321 7y ago
  It doesn't matter, a datacenter can always be repurposed for their own internal product lines. To be perfectly honest, i would rather go with a company that provides cloud as one of their main products rather than a company that has it as one of their many products.
- jlj 7y ago
  > Google is currently building a massive $500 million datacenter.. Money talks. Culture talks louder.
- hvidgaard 7y ago
  On the other hand the, in comparison tiny, cost that Google Reader and other was in the budget, would have given Google so much good will. It was in other words cheap positive PR.
  - ageofwant 7y ago
    Killing Reader was massive strategic blunder, which in fairness to Google was probably not obvious to see at the time. But then to repeat that with Maps API,Google+, Sketchup and Wave (perhaps wave was first) just cemented Google's untrustworthy reputation.
    - Applejinx 7y ago
moksly 7y ago
Amazon and Microsoft are much better at support though. I doubt you’ll see a lot of Google cloud in the European public sector where I work for instance. Google simply doesn’t know how to work with us or give us the reassurances we need to trust them. I’m by no means a AWA or Azure fanboy, I’m on the team that hopes we get a solid European alternative, but until we do both Amazon and Microsoft are leagues ahead of Google.
- angry_octet 7y ago
  I remember vividly talking to a GCP saleswoman. I was asking her about technical things and process/pricing and she just kept saying things like "We're talking with $FIRSTNAME about that." And I was like, 'who is this $FIRSTNAME guy?' and then I realized it was the CTO. This guy is so far divorced from anything technical or policy it's absurd, he just produces content for his 1- and 2- up bosses to look good in press releases, and he has no input to the way anything is done. But GCP had identified him as the way into the organization, and was totally uninterested in engaging with anyone else other than giving away swag. Contrast this with AWS. They have expert engineers, pre- and post- sales, helping you find good ways to solve your problem. I could ring a senior guy on his cell right now if needed. The only problem with AWS is cost and complexity, it's the fast path to developing a $20k/day cocaine habit. But the AWS dealer will definitely come to your house to deliver at 11pm. (It is a bit of a change from 10 years ago though, when everything had to be paid by credit card. Gave the accountants a heart attack. Gave me an ulcer from thinking about everything halting because the fraud control might reject the bill without warning...) Azure definitely isn't in that league, still want to sell to the whole enterprise, getting promo stuff requires them to sign off at five levels, lawyers are involved in everything. In short, their dev relations are shite.
  - EmberQuill 7y ago
    To illustrate your point about AWS, at my own workplace there are a few AWS people who are actually in our office once a week, sitting nearby just so they can be easily available to answer questions and work with us to design our architecture or assist in troubleshooting issues. I honestly can't imagine Google providing that level of dedicated support.

james_s_tayler 7y ago

Also AWS is just the best at marketing their services. They did what Google did with search. We always say we "Google" things. Developers these days say "Chuck it in S3" or "Run it on EC2". I feel they have a strong advantage because they've kind of come to own the language of the cloud.

JimboOmega 7y ago
Actually, most of the time we say "in the cloud" rather than naming a specific hosting provider. Definitely Amazon has the first mover advantage and accumulated reputation first, but it's hardly unassailable. Still... I think the bigger problem isn't whether or not Google plans topull the plug or not, but that it has a reputation of pulling plugs to such an extent that people factor that into a decision of using their services. When it's something as big a choice as a cloud provider that can cast a long shadow over their offerings.
- sdan 7y ago
  In addition, I remain cloud agnostic as well. As a student, I stick to whichever platforms can give some free credits. At the moment I have some cloud infra on google and some on aws.
  - dragonsh 7y ago
    That’s a good thing not to tie in to proprietary api’s, indeed every cloud provider is doing it and after Google’s recent management shuffle to bring in management from Oracle making them alike. Try using apache-libcloud to make your deployment and programs cloud agnostic. Also try using smaller alternatives like hetzner or scaleway. Once you go down this cloud rabbit hole you will be hold hostage to whims of these providers, where they can shutdown any service at their will or ask to pay an arm and leg. Also try to have non USA cloud providers as another backup option in the long run.
mindcrime 7y ago
Developers these days say "Chuck it in S3" or "Run it on EC2". I feel they have a strong advantage because they've kind of come to own the language of the cloud. After many years of working with AWS almost exclusively, I recently started a new $DAYJOB where our primary public cloud provider is Azure. A couple of days ago I was chatting with a co-worker about something, and he literally says "I think we should just chuck it in S3 and call it good" (paraphrased slightly, due to my imperfect memory). It struck me as odd for a few seconds, and then I realized that exactly what you are talking about has taken place. S3 is so dominant that it's become the default label for object storage in the cloud.
- kevinventullo 7y ago
  https://en.wikipedia.org/wiki/Generic_trademark https://en.wikipedia.org/wiki/Generic_trademark
  - mindcrime 7y ago
    Exactly.
sidlls 7y ago
I don't think I've ever heard AWS-specific terms applied to GCP or Azure infrastructure.

rhizome 7y ago

>Google’s problem remains its reputation for turning things off. I'd put it this way: Google has a reputation for turning things off, but its _problem_ is an inability to stop doing that, to commit to its users/customers. In other contexts this could be written in terms of addiction. If I were Sergey and Larry, this would be the thing I'd be retiring over, and I'm not sure Suchar can change it.

adventured 7y ago
If Satya Nadella could get Microsoft to move beyond its Windows-all-or-nothing pathology, someone in charge at Google (including the two founders that vote-control the entire company directly - they can de facto hand Sundar the power to act as he needs to) can trivially put a stop to the relationship commitment issues at Google.
- throwaway2048 7y ago
  Its far from trivial to change an entire corporate culture on a dime. From everything we have seen of Google's culture, working on new projects is exciting and is absolutely the way to get promoted. Doing maintenance work, especially on products that aren't wildly successful gets you no attention, no staff, no funding, and probably more than a little tongue waging. The CEO cant just change all that with directives, even if they do something extreme like a "No Depreciations" policy, products are liable to be depreciated in all but name (Joe the intern takes care of it every 5th Tuesday, technically not depreciated) by sheer corporate momentum alone, nobody wants to be stuck with a dead fish, even if its now illegal to openly comment on how much it stinks. It would require change at every level of Google to place value into fixing/improving existing projects, rather than all the glory going to greenfield development. Considering the kind of "kingmaker" culture Google's mentality breeds, ie what gets me ahead is gathering resources to focus on my project, changing things to a setup where contributing to other peoples silos is welcome and common is going to be an extremely difficult prospect because virtually all the people with power inside Google got that way by carefully tending their fiefdom, they aren't just going to throw that all a way, no matter what any executive says.

sundvor 7y ago

Another perception is that of killing hierarchies of user accounts / pulling rugs on projects e.g. if links are established between adding someone who once broke the rules and whatever they might be working on now, with no way of reaching someone human for support... ... where one of the few ways of rectifying the situation is to get a top posting on HN in the hope that a Googler will see your plea for help. I mean, that's probably mainly related to Android dev but I've seen plenty of stories to support that feeling: E.g. (just a random example) https://news.ycombinator.com/item?id=17431609 https://news.ycombinator.com/item?id=17431609 My perception might be misguided / influenced by being in a perception bubble, however that is my reality and AWS or AZURE just seems a whole lot safer. So, why should I as a developer invest any time in GC? I did attend a Google dev day re machine learning++ in Melbourne a couple of years ago, and the products seem nice and all, but the competition isn't exactly lagging behind. My recommendation would be to really address this issue of human support and trust.

res0nat0r 7y ago

The issue is that folks here think GCP is the same thing as Google Reader. That's absurd. Google is currently building a massive $500 million datacenter outside of Reno as we speak, and has 10+ billion invested in their datacenter cloud offering buildout this year alone. GCP and the entire physical infrastructure isn't going to get "turned off" if it isn't #1 in the market in 3 years, that's just silly. https://9to5google.com/2019/07/01/google-data-center-nevada/ https://9to5google.com/2019/07/01/google-data-center-nevada/ https://www.rgj.com/story/money/business/2019/02/13/google-investing-13-b-data-centers-2019-including-nevada/2861826002/ https://www.rgj.com/story/money/business/2019/02/13/google-i...

dana321 7y ago
It doesn't matter, a datacenter can always be repurposed for their own internal product lines. To be perfectly honest, i would rather go with a company that provides cloud as one of their main products rather than a company that has it as one of their many products.
jlj 7y ago
> Google is currently building a massive $500 million datacenter.. Money talks. Culture talks louder.
hvidgaard 7y ago
On the other hand the, in comparison tiny, cost that Google Reader and other was in the budget, would have given Google so much good will. It was in other words cheap positive PR.
- ageofwant 7y ago
  Killing Reader was massive strategic blunder, which in fairness to Google was probably not obvious to see at the time. But then to repeat that with Maps API,Google+, Sketchup and Wave (perhaps wave was first) just cemented Google's untrustworthy reputation.
  - Applejinx 7y ago

moksly 7y ago

Amazon and Microsoft are much better at support though. I doubt you’ll see a lot of Google cloud in the European public sector where I work for instance. Google simply doesn’t know how to work with us or give us the reassurances we need to trust them. I’m by no means a AWA or Azure fanboy, I’m on the team that hopes we get a solid European alternative, but until we do both Amazon and Microsoft are leagues ahead of Google.

angry_octet 7y ago
I remember vividly talking to a GCP saleswoman. I was asking her about technical things and process/pricing and she just kept saying things like "We're talking with $FIRSTNAME about that." And I was like, 'who is this $FIRSTNAME guy?' and then I realized it was the CTO. This guy is so far divorced from anything technical or policy it's absurd, he just produces content for his 1- and 2- up bosses to look good in press releases, and he has no input to the way anything is done. But GCP had identified him as the way into the organization, and was totally uninterested in engaging with anyone else other than giving away swag. Contrast this with AWS. They have expert engineers, pre- and post- sales, helping you find good ways to solve your problem. I could ring a senior guy on his cell right now if needed. The only problem with AWS is cost and complexity, it's the fast path to developing a $20k/day cocaine habit. But the AWS dealer will definitely come to your house to deliver at 11pm. (It is a bit of a change from 10 years ago though, when everything had to be paid by credit card. Gave the accountants a heart attack. Gave me an ulcer from thinking about everything halting because the fraud control might reject the bill without warning...) Azure definitely isn't in that league, still want to sell to the whole enterprise, getting promo stuff requires them to sign off at five levels, lawyers are involved in everything. In short, their dev relations are shite.
- EmberQuill 7y ago
  To illustrate your point about AWS, at my own workplace there are a few AWS people who are actually in our office once a week, sitting nearby just so they can be easily available to answer questions and work with us to design our architecture or assist in troubleshooting issues. I honestly can't imagine Google providing that level of dedicated support.