This message was deleted.

google killing any of their products would not be surprising at all imo

lol. I’m fairly sure I’d be quite impressed at how much effort and cash would be required just to spin down a cloud offering.

Yeah but does that mean they wouldn't do it? I mean, Google is big into self-immolation right now. 🤷🏻‍♀️ @Adrian Kuepker

I have found the discussions and speculation on the topics of those twitter threads you linked @Christina Harker extremely interesting. I haven’t used GCP in any major capacity, so I won’t be mad if ya’ll disregard me completely… …but the fact that in a few months time, it won’t be straight forward to write Terraform or other IaC that can spin up a full stack with domain and certificates handled without reaching another service provider (and thereby using more than one terraform provider for example). Likely splintering the monitoring and billing control planes etc….it baffles me that this was considered a good idea… Unless….indeed the plan is to kill/sell GCP.

Yeah, @Heiðar Eldberg Eiríksson, I think that's the general direction behind the questions. You nailed it. The move is so illogical that people are searching for a rational explanation. And the most logical reason appears to be killing/selling GCP. Then again, Hanlon's razor could be in play here: "Never attribute to malice that which is adequately explained by stupidity." The way you describe it though, it's going to be messy and cause a lot of turbulence for all kinds of use cases.

I think the aspect of this that troubles me the most is the apparent or perceived callousness of the decision and the timeline. I work in pharmaceutical logistics. We create tech, to monitor medicine. The data we generate is used to determine whether the medicine has been stored and transported under the correct environmental conditions to be safe for human consumption. Any major change in our interface or functionality of safety critical system components, needs to be tested and validated for several weeks (with the current process and technology in the industry). Operating under conditions where my provider can pull the rug from under me in 3 months is just not acceptable. So at the end of the day….whether Google intend to kill or sell GCP doesn’t even matter for me….I and my employer…can basically never use them regardless.

That's a really incredible perspective. And you can't be alone in it, either. SMDH at Google.

Same guy’s pinned tweet.

That’s kind of my point though Adrian. Of course it’s technically possible for example for me to operate the infrastructure of my current employer on GCP. Even with these kinds of decisions. It’s just my opinion that it would be a worse experience than my current provider, due to a history of decisions like these. That opinion of mine is worth somewhere between 2.5 and 5 million dollars annually in cloud provider spend and growing….but it’s just my opinion.

Yep. Just as selling off or shutting down a software group is a vastly different proposition from just shutting down and walking away from a US$20B or more physical infrastructure investment. It won’t just “shut down”, but someone else could end up running it. 22 compute regions is a LOT of investment for which the value is only retained by staying up and available. And it’s far more likely that Google will instead continue to fragment under the Alphabet banner and GCP will be forced to sell as advertising revenue declines or gets de-coupled from the rest of the operation, and those impacts will be long-term rather than short-term once Ad revenue is no longer there to cover the shortfalls.

Reasonable assessment. The overall value of course is also tied to usage and continued trust…That’s why this is so puzzling. Why create speculation around the future of your $20bn investment? It’s just so head scratch inducing right?