I didn’t see this coming and I think it’s funny, so I decided to post it here.
You must log in or register to comment.
I was going to write that every function should be a service as sarcasm, then I realized that’s exactly what this article is proposing. Now I’m not even sure how to make a more ridiculous proposal than this.
Cant wait to set up a docker container for a service which takes a string input and transforms it into a number as the output. Full logging, its own certificate for encryption of course, 5 page config options and of course documentation. Now, you want to add two numbers together? You got the addition service set up right?
It’s a modern day enterprise fizzbuzz: https://github.com/EnterpriseQualityCoding/FizzBuzzEnterpriseEdition
Oh yeah, this is the stuff business dreams are made of.
I’m trying to understand how this is different than a concept I learned in computer science in the late 80s/early 90s called RPCs (remote procedure calls). My senior project in college used these. Yes I’m old and this was 35 years ago.
Neovimservices ftw
This is just distributed functions, right? This has been a thing for years. AWS Lambda, Azure Functions, GCP Cloud Functions, and so on. Not everything that uses these is built on a distributed functions model but a fuck ton of enterprises have been doing this for years.
someone at my work didn’t get the message.
2000+ line function with 3157 node packages.
is that nano enough?
Tech moved in cycles. We come back to the same half-baked ideas every so on, imagine we just discovered the idea and then build more and more technologies on top to try to fix the foundational problems with the concept until something else shiny comes along. A lot of tech work is “there was an old lady who swallowed a fly”.
I always keep saying " You cannot plan your way out of a system built on broken fundamentals." Microservices has it’s use case, but not every web app needs to be one. Too many buzzwords floating around in tech, that promise things that cannot be delivered.
Yep micro services are great, but monoliths are just as great and don’t let anyone tell you otherwise. It all depends on what the system requirements are.
Nano services are microservices after your company realizes monoliths are much easier to maintain and relabels their monoliths as microservices.
Unironically. I’d put a significant wager down on that being the source of this term.
quantum services
take your source code and put each character in its own docker container
this gives you the absolute peak of scalability and agility as every quantum of your application is decoupled from the others and can be deployed or scaled independently
implementing, operating and debugging this architecture is left as an exercise for the reader
that will be $250,000 kthx
implementing, operating and debugging this architecture is left as an exercise for the reader
Challenge accepted by a reader using AI, what could go wrong? xD
We already have nanoservices, they’re called functions. If you want a function run on another box, that’s called RPC.
What’s next? Femtofunctions
You only need two of them, one for
1and one for0
This “article” was written by AI, wasn’t it? This is just throwing vague buzzwords around
Planck services
My services are so small that it is impossible to know just how fast they are running!
I am now offering Planck services for sale, at US$0.0001 per bit.
For an extra fee, you can even choose the value of the bit.
I feel like this name addresses the problem of services claiming to be microservices when they’re not.
Does that even happen? cat is micro, sed is micro, systemd isn’t and doesn’t claim to be
You know what they say: micro services, macro outages.
At least you can accurately point the finger at who’s responsible
Well during the never sev0 I’m sure the shareholders will be satisfied with that.
This is just the Linux way but with… Rest? Seems slow.
