I mean, "serverless computing" has always only meant that it's "serverless" for the customer who buys the compute power in the sense that they don't have to bother the slightest with the architecture or managing it. Not really anything to reveal there...
I think the "Cat Looks Inside" meme would've been more appropriate, because the "Let's See Who This Really Is" (a.k.a. "Scooby Doo Reveal") meme is more about revealing something that is actually different, while CLI is sarcastic. Like "Wireless device. Look inside. Wires" isn't revealing anything serious but makes fun of the misleading nomenclature. A good SDR example would be pulling the mask off a KKK member to reveal a cop, while they are supposed to be on the opposite sides, they are one and the same.
On the meme spectrum, SDR sits somewhere between CLI and "They are the Same Picture".
MySQL: you set it up, if the server fails, you have to fix it. You set up replication, replication fails, you have to fix it. It's your alarms, you get up at 2:00 a.m., you set up backups. You deal with IP changes. You deal with your two+ boxes and their patches. Those servers are your responsibility. If their hypervisor needs an update you're stuck with the boxes going down.
Aurora serverless: you don't deal with any of that.
Saying they're the same as like saying that a self-driving taxi is the same as leasing your own car. In both cases there are servers involved, But in one of the two cases you don't have anything to do with the server.
Serverless in cloud computing typically refers to ephemeral processes...things like lambdas and message handlers.
Outside of that it's just a buzzword anyway (like "low code/no code" which is similar) so I guess any managed software is serverless by your definition?
From the point of view of the customer it is serverless. Maybe it's being done on a server, but maybe it's a magical genie in a bottle. You don't have to care because from your point of view you upload code and that code magically runs.
This fits perfectly in with other "-less" words. Like many "priceless" museum artifacts were bought and sold before they showed up in the museum. To the visitor and maybe to the museum they're priceless, but to the dealers who found it for the museum it had a price.
Maybe it's being done on a server, but maybe it's a magical genie in a bottle. You don't have to care because from your point of view you upload code and that code magically runs.
Hard disagree. As someone who wrote several AWS lambdas, I know you have to care that it's being run on a server and you have to adjust to your code to work within that very-specific server system.
If anything it should be called "poly-server" because you cannot write your code without considering that it can be executed from several servers around the same time. I don't buy what you're selling here, other -less examples don't seem to betray their terminology at all to me but "serverless" will always sound wrong to me.
"Function-based", "image-based" would have been slightly more accurate terms.
Wireless devices aren't actually "free of wires", it's that you don't have to deal with wires (or significantly less, since you still have to charge them etc., save for wireless charging). So that's not really new either.
The first time I heard the term wireless, I was a little kid and I understood very quickly. When I first heard the term "serverless" I was an adult who had been programming a couple years. I remember genuinely being confused as strings of unparseable buzzwords bounced off my brain. A minute or two into the explanation, I'm pretty sure I said "oh, so it actually does run on a server". The ops person was forced to say yes. It was a genuinely confusing and imo pointless conversation that we shouldn't have needed to have.
Were you doing any serious "devops" at the time? I didn't struggle with "serverless" knowing that otherwise I had to manually provision servers, virtual or bare metal.
Yeah, and the big selling point for serverless is that you only deal with the code you want to run, none of that "server management" stuff. It's a perfectly reasonable name based on what's appealing about it.
As someone in the ISP/hosting business, i can tell you that there are plenty of companies incapable of sufficiently managing actual servers. For their own safety it's probably better to let someone else manage it for them (despite getting ripped off then)
I'd like to take the stance that: If you can't manage your own data, don't start a business. But that seems like a shaky foundation to plant a flag, so I will instead say, "I hate Oracle."
If they cannot manage their own infrastructure, they also don't know what infrastructure is needed for their services. And they won't even have the opportunity to learn anymore.
Secondly, if you buy external services, you need to consider improving connectivity.
I mean, you can still work on your on-premises servers, if your internet connection fails. You cannot, if you outsourced essentials parts.
People used "function as a service", "managed *" or "compute as a service" for a bit, but serverless actually seemed to capture the gist of it for customers better. It may be marketing speak, but it does seem to be an effective shorthand for the value it provides.