Ranter
Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Comments
-
It's a terrible, terrible name for a really useful product.
No, I don't think it's going to "take over the world", but the ability to scale near instantly, have zero maintenance on the underlying server, and in particular scale to zero cost on "glue code" and similar is *really* handy. -
It's a weird term and yeah it deserves some ridicule.
On the other hand you can't use any slightly technical sounding marketing speak around technical folks without someone getting pedantic too.
There's no winning if you're playing in both worlds. -
@N00bPancakes I've always maintained "function as a service" is a way better name.
-
@gronostaj it's a reference from an absolutely stupid, but funny series called New Kids Turbo. Has been made by some netherlands dudes.
The translation simply says "crazy Mongo" -
bioDan56224y@AlmondSauce FaaS already exists
But I agree with your point about it being a terrible name, I'd even prefer SeaaS or SeraaS (Server as a Service)
Best option: Managed Servers -
In that case I have some bad news regarding the name of Democratic People's Republic of Korea
-
Anything that is not your machine but is accessible is a server, it's called serverless because you don't have to deal with implementation, just write code
-
@AlmondSauce It's super useful, but also has awful bottlenecks and overhead in its own right.
Microservice-like architectures and modular code makes it easier to scale (and replace!) parts independently, but communication between parts becomes more costly, both in terms of latency and financial cost.
I mean, both HTTP REST calls and message bus systems can be fast, but it's never as fast as calling one method from another method within a monolithic binary.
Serverless is great for things like adapters on top of third party APIs. It works well for relatively static pages and simple APIs as well.
It doesn't work too well for performance-critical, complicated backends. Those can still be broken up into "mini services", but when you go "too small", adding too much infra crap will turn into a performance & optimizing nightmare.
Some companies have succeeded at it (Netflix for example), but I would argue that it's not the ideal solution for them either, that it was an uphill battle which they won through perseverance, despite all the drawbacks. -
@bittersweet Sure, precisely. It's another tool in the toolbox, and there will be cases where it's really useful and completely inadequate (as well as corner cases where it may or may not be a good fit.)
The problem is with tech like this so many people tend to think you have to go for the extremes - either thinking it's the best thing going and everything will be using that and nothing else in the future, or hating it and saying anyone who uses it is a moron.
The truth is rarely that black and white, of course. -
@AlmondSauce Exactly.
Many devs ask too often "how did <successful company x> solve this" instead of "what works for our problems".
Facebook makes billions because of abusive advertisement tracking pixels, not because their JavaScript is compiled from Flow script.
Of course you can learn from other companies and their tech stacks, but you have to take their success stories with a grain of salt and view it through the lens of their own survivorship bias. -
@bittersweet
"but you have to take their success stories with a grain of salt"
Their sucess stories aren't even survivors bias any more.
They're intentional marketing pieces full of myths and sold dreams.
Like facebook. Does anyone not half asleep still believe some geek dropped out of college and simply started a company that went on to make billions?
Yeah maybe but not really.
Or was it handed to him under a different project name? Oh, I don't know, something along the lines of "lifelog"? -
@Wisecrack Well regardless of the ethics, the main tenets of economics and markets apply just as well to tech companies.
Is the product something with a large demand, and is it more marketable than competing products? MySpace didn't innovate (read: steal ideas) fast enough, so in came Facebook with a fell swoop.
Microsoft stole most of their IP, Apple stole, Facebook stole... It's always a game of 4X (explore, expand, exploit, exterminate).
The success stories are rarely the first or even best, they are just the ones which survive because they juggle their balls just in the right way (while elbowing more ethically inclined competitors and even many customers in the face).
But tech stack is rarely the deciding factor.
Reddit for example became big on a monolithic PHP/MySql backend, with nearly zero JavaScript, or even any attention to proper design. They just offered a simple but unique service. They usurped the previous king of news aggregation, Digg, which ironically dug(g) it's own grave with a technical migration which screwed over end users. -
NoSQL is Not Only SQL, not no sql.
Blackbox testing isn't done on a machine with black painted hardware.
Penetration testing doesn't mean inserting screw drivers in the processor.
Cloud Functions is a framework, not just a function.
BigQuery isn't a big query, the query is still small, the data is big.
Can someone please continue. I'm out of ideas...
Related Rants
Someone has to fucking say it.
rant
server
serverless