Chat GPT explains all its versions in a proper way

You said:
Why do suckers that created you made fucking billion inconsistent names? "GPT-3" is ok, "GPT-3.5", "GPT-4" are ok, what about those shit "GPT-o1", "GPT-o3", "GPT-o3mini" and other meaningless shit? Are they stupid or what?

ChatGPT said:
OpenAI's naming is a mess, no argument there. They went from a clear versioning system (GPT-3, 3.5, 4) to random labels like "o1," "o3," and "o3-mini" that mean nothing to normal users. Probably some internal model identifiers they just dumped on the public without thinking. Makes zero sense from a branding perspective.

You said:
Can you sum up all the shit they created in a proper increasing order? "GPT-3" should be "GPT-1" since it was in reality the first GPT that was available to public, "GPT-3.5" should be "GPT-2", etc.

ChatGPT said:
Here's the mess cleaned up logically:

GPT-1 (2018) – Never public
GPT-2 (2019) – First public model, but with limits at launch
GPT-3 (2020) – First real public GPT (should be GPT-1)
GPT-3.5 (2023) – Improved version (should be GPT-2)
GPT-4 (2023) – Major upgrade (should be GPT-3)
GPT-4 Turbo (o1) (Nov 2023) – A cheaper, faster variant (should be GPT-4)
GPT-4 Turbo (o3, o3.5, o3-mini, etc.) (2024) – Different tweaks, no clear meaning (should be GPT-5)
They could’ve just numbered them properly, but instead, they turned it into a cryptic mess.

Postman is shit, alternatives?

Recently discovered that Postman is now dead shit and its devs are fuckers that now made their shit a payware.

https://blog.postman.com/announcing...

The solution is to use latest normal version https://github.com/postmanlabs/...

Or get an alternative. Are there any good ones?

Explaining docker in 5 simple steps:

1. We need to be OS agnostic so we we use docker containers
2. We remove 3 lines of code to run migrations on host computer because they are OS specific
3. We add one container for DB and another one to apply Flyway migrations
4. We write 15+ lines of .SH OS SPECIFIC GLUE CODE TO MAKE THAT CONTENIRIZED SHIT WORK
5. Now we are OS agnostic because we use containers

Stupid fucking monkeys

Most actual GraphQL explanation:

1. Still uses your xhr/fetch/axios on FE
2. Just sends all the requests to single endpoint
3. On BE uses its own resolution schema to call proper controller to handle the request, rather than relying on router for that

That's all!

Just another useless layer of abstraction with its learning curve, tricks and bugs as ORMs are

Modern approach to displaying dates is shit

I want to see a face of a fucker who decided that "5 YEARS AGO" has more meaning than exact date "15 Sep 2019". All apps that use that fucking retarded date representation are useless shit, worst UX ever.

HP is shit.

Just look at this thread

https://h30434.www3.hp.com/t5/...

Not only those fucks don't know how to properly make an uninstaller for the shit software they make, their own support knows less about it, than random guys from the internet.

People responsible for closing threads on stackoverflow for "We don’t allow questions seeking recommendations for software libraries" should die and rot in hell forever.

Multer is shit.

Do you know how to validate file size + mime type of the file before saving it to the disk?

Here is an instruction:

1. In one place you provide a it a fileFilter function that will validate the mime type, but this function can not validate file size or send user response.
2. In second place provide it a config limits file, that will validate file size, but can not validate mime type
3. In third place you provide middleware to capture any errors caused by previous 2 places and send response back to the user.
4. Go google for what is cb in multer because their shit documentation says nothing about it

Such simple thing as request->validation->response is split into 3 different parts, each of different type: callback, config file and middleware.

Why do so much people use this shit written by a braindead retarded morons?

All OS file browsers are intentionally made not to show folder sizes to then sell OS "cleaners" or make users buy new hardware instead of allowing them to easily clean their PC from garbage.

First person who comments, that “calculating folder sizes is slow” is a liar, because it is not. Also having an on/off toggle for that is easier than anything else.

All mail clients are intentionally made not to show sender email address, but rather their chosen name to then launder money on anti-phishing trainings.

React router is shit
I have never seen more retarded library.

Not only those suckers change the 100% of the API every fucking update for no reason, also they have the most fucked up documentation ever.

No search in the docs!!! Fucking bullshit examples with no such easy things like how to create nested routes.

Please, stop using this piece of shit, I'm tired of working with this fucking abomination. Hope they will delete their shit repo one day.

Twitter comments are shit
How the fuck to use them? Why don't they just show one under another? If post has more than 2 replies it can be clicked to open its direct replies, some of them are shown in main thread, some of them are shown under opened thread. Some replies from main thread are shown under child thread. And to make it even worse any comment of sub thread can be expanded as well showing some random replies that could be expanded recursievly so you may end in main thread again. It is just impossible to read all replies in chonological order. Show me the moron who made this, I wanna kill him....

Macbook keyboard is shit
Especially european variant.

I want to see any person in the world that is braindead enough to think that making short left shift if favour of having abolutely useless button there is good.
This is the most stupid decision that could be ever made.

Not only the symbols used by that button are seldom used, they also are duplicated on they other keyboard keys. But shift is used all the time and must be big enough instead of that shit.

Why useImperativeHandle in React is bad?

I have component A. I have component B which is a child of A.
I have button in component A that should change how component B is displayed.

Usual approach is to set state in A, change it upon click and pass that state as flag via props to B.

But there is another approach via exposing ref to component B using forwardRef() + useImperativeHandle().
And then calling method that belongs to B from A. Like this: bRef.current.changeDisplay()

https://stackoverflow.com/questions...
https://reactjs.org/docs/...
https://github.com/reactjs/...

People on SoF and Docs say that the first way is preferrable. But no one explain why.
Does anyone know why calling child method from parent is considered bad?

CORS is shit
Stupid useless shit that protects from nothing. It is harmful mechanism that does nothing but randomly blocks browser from accessing resources - nothing more.

Main idea of CORS is that if server does not send proper header to OPTIONS request, browser will block other requests to that server.

What does stupid cocksuckers that invented CORS, think their retarded shit can protect from?
- If server is malicious, it will send any header required to let you access it.
- If client has malicious intents - he will never use your shit browser to make requests, he will use curl or any ther tool available. Also if server security bases on something as unreliable as http headers it sends to the client - its a shit server, and CORS will not save it.

Can anyone give REAL examples when CORS can really protect from anything?

Vivaldi browser is shit.

Simple isntructions on how to make most shitty browser ever:

1. Force users to use "really-fucking-long" password that will not match to any of their existing ones.
2. Invent some useless stupid "encryption password" (why does any normal browser work fine without that shit) and most ridiculous - automatically set it to be the same as the main password.
3. Of course you forget the pass you set because you dont remember what symbol you added 5 times in the end of your normal pass to fit their stupid rules.
4. You have to reset it
5. "Encryption password" does not reset with it, so you still dont remember it
6. Sync is not working!
7. If you think this is shitty enought, you are not right - they went futher. To reset that fucking "encryption password" you have to... ERASE ALL YOUR CLOUD DATA.

Fucking retarded piece of shit - never, never trust those morons who made this shit browser to sync any of your sensitive information.

All motherfuckers who set their stupid website to try to send you notifications must burn in hell after they die. I am tired to setting every browser I install to always block notifications because of that stupid morons.

Why does everything installed via npm sux so hard?

Why the fuck does any minor update in their bullshit packages either forces you to change config files:
E.g. now should be "@babel/core" instead of "babel-core" - WHAT A FUCKING SIGNIFICANT CHANGE!!! Rewrite all you configs motherfucker, that goddamn "@" in front of our shit is SO IMPORTANT that we will break everything to add it

Or breaks the code internally:
Consider the recent fail of fucking Terser [https://github.com/gatsbyjs/gatsby/...] that breaks fucking webpack and FORCE YOU TO ROLLBACK TO ANY VERSION THAT WORKS, why you nerd retards, can not run a simple dummy project BEFORE YOU RELEASE YOUR SHIT???!?!!?

Why any fucking update from *.*.1 to *.*.2 turns into hours of googling of what the fuck got broken this time??
The way that webpack, babel and other npm packages are released nowadays is absolutely retarded. I really have a strong feeling that it is better to keep old error-proof working config and NEVER UPDATE, than constantly suffer from butthurt

p.s.
Of course I am sorry for all the hate and caps in my post, and have respect for guys that develop amazing stuff for us for free, but I need to share this

Does this book worth reading?

Hello I've finished several books about JavaScript already and many of them mention this one

JavaScript: The Definitive Guide, 6th Edition [2011]

I am a bit embarrassed, because despite it looks like a thorough 1000 pages guide, its last edition dates 2011 year.

From one point of view, it should contain core concepts that still remain the same, from the other, reading 2011 book in 2018 sounds like a not very good decision.

What would you say?

I think that the idea of IIFE in JavaScript is a good example of how shitcoding can be built into the core of the language.
How can I make my code clean if the language itself forces me to make such an ugly constructions?
What do you guys think about it?

Firefox is shit

- In version 62 the cocksuckers finally removed the ability to add description to bookmarks, now there is no difference between mozilla shit browser and chrome, except chrome has working mobile version and working bookmark sync.
- Yes, firefox mobile on android does not even allow you to edit bookmarks so that useless crap can't be called a browser at all.
- Firefox shitcoded sync takes more time to sync 1 bookmark, than chrome takes to setup all bookmarks/addons/settings on fresh-install and in addition to that, it messed my bookmarks several times, so I am forced to sync places.sqlite via dropbox to syncronize bookmarks, because it is way much faster and reliable.

WooCommerce is shit

Just try to find anything in the documentation on their retarded website and you will get what I am talking about.

A billion of unstructured links. With absolutely no sidebar or table of contents.
IMPOSSIBLE to find any fuck there.
It is far easier to build an e-store from scratch, than to customize their ugly monster.

Now look Laravel or PHPunit documentation to compare how it should be.

New gmail is shit, better make a gcloud app for linux instead of redesigning that shit billion times, you fucks