> All the pro-cloud talking points are just that - talking points that don't persuade anyone with any real technical understanding, but serve to introduce doubt to non-technical people and to trick people who don't examine what they're told.
This feels like "no true scotsman" to me. I've been building software for close to two decades, but I guess I don't have "any real technical understanding" because I think there's a compelling case for using "cloud" services for many (honestly I would say most) businesses.
Nobody is "afraid to openly discuss how cloud isn't right for many things". This is extremely commonly discussed. We're discussing it right now! I truly cannot stand this modern innovation in discourse of yelling "nobody can talk about XYZ thing!" while noisily talking about XYZ thing on the lowest-friction publishing platforms ever devised by humanity. Nobody is afraid to talk about your thing! People just disagree with you about it! That's ok, differing opinions are normal!
Your comment focuses a lot on cost. But that's just not really what this is all about. Everyone knows that on a long enough timescale with a relatively stable business, the total cost of having your own infrastructure is usually lower than cloud hosting.
But cost is simply not the only thing businesses care about. Many businesses, especially new ones, care more about time to market and flexibility. Questions like "how many servers do we need? with what specs? and where should we put them?" are a giant distraction for a startup, or even for a new product inside a mature firm.
Cloud providers provide the service of "don't worry about all that, figure it out after you have customers and know what you actually need".
It is also true that this (purposefully) creates lock-in that is expensive either to leave in place or unwind later, and it definitely behooves every company to keep that in mind when making architecture decisions, but lots of products never make it to that point, and very few of those teams regret the time they didn't spend building up their own infrastructure in order to save money later.
This feels like "no true scotsman" to me. I've been building software for close to two decades, but I guess I don't have "any real technical understanding" because I think there's a compelling case for using "cloud" services for many (honestly I would say most) businesses.
Nobody is "afraid to openly discuss how cloud isn't right for many things". This is extremely commonly discussed. We're discussing it right now! I truly cannot stand this modern innovation in discourse of yelling "nobody can talk about XYZ thing!" while noisily talking about XYZ thing on the lowest-friction publishing platforms ever devised by humanity. Nobody is afraid to talk about your thing! People just disagree with you about it! That's ok, differing opinions are normal!
Your comment focuses a lot on cost. But that's just not really what this is all about. Everyone knows that on a long enough timescale with a relatively stable business, the total cost of having your own infrastructure is usually lower than cloud hosting.
But cost is simply not the only thing businesses care about. Many businesses, especially new ones, care more about time to market and flexibility. Questions like "how many servers do we need? with what specs? and where should we put them?" are a giant distraction for a startup, or even for a new product inside a mature firm.
Cloud providers provide the service of "don't worry about all that, figure it out after you have customers and know what you actually need".
It is also true that this (purposefully) creates lock-in that is expensive either to leave in place or unwind later, and it definitely behooves every company to keep that in mind when making architecture decisions, but lots of products never make it to that point, and very few of those teams regret the time they didn't spend building up their own infrastructure in order to save money later.