Me: PHP, please load the images.
PHP: No.
Me: Come on PHP, the deadline is tonight.
PHP: Haha nope.
Me: What if I swap the libraries?
PHP: Still no.
Me: What if I add a symlink?
PHP: No. And also I hate you.
Me: *gives up and sends an email to the client saying we have to delay the release*
PHP: lololol jk here's the images haha
Me: :/

A few years ago:

In the process of transferring MySQL data to a new disk, I accidentally rm'ed the actual MySQL directory, instead of the symlink that I had previously set up for it.

My guts felt like dropping through to the floor.

In a panic, I asked my colleague: "What did those databases contain?"

C: "Raw data of load tests that were made last week."

Me: "Oh.. does that mean that they aren't needed anymore?"

C: "They already got the results, but might need to refer to the raw data later... why?"

Me: "Uh, I accidentally deleted all the MySQL files... I'm in Big Trouble, aren't I?"

C: "Hmm... with any luck, they might forget that the data even exists. I got your back on this one, just in case."

Luck was indeed on my side, as nobody ever asked about the data again.

Today was hell on earth as for user support. Phone going non stop, tickets coming in faster than we were able to process.

At the end of the day I had to make a symlink for a customer which is fine. But, the day was so busy that I just couldn't focus anymore.
I've made 1K+ symlinks in my life probably but I couldn't remember if the source or destination comes first with a symlink.... The day has been hell and I just couldn't bring up a single second of focus anymore..

Fuck it, I'll do it tomorrow. I know I can do this but I don't trust myself with this right now in case of a huge webshop (swap the source/destination: webshop gone).

I think I'll thank myself for this tomorrow.

While trying to symlink my i3 config, I accidentally deleted it
So frustrated, and now I'm back to gnome cause I can't be bothered to redo the >5 hours I spent customising it :(

2 Things:. Never symlink the root directory and don't try to remove a symlink with rm -rf

Nearly shit my pants today.

Fuck android studio updates, fuck OpenGL libraries, fuck nvidia graphics on Linux, fuck it all to hell.

These bullshit updates keep breaking stuff in apps, and then the emulator itself gets broken somehow, then I need to run a bullshit shell script I found on some esoteric Arch Linux forum to symlink OpenGL .so libraries on my Linux machine every time I update because fucking OpenGL and nvidia crappy drivers on Linux, which also broke my xrandr configs and segfaulted the xfce4-display app.

Seriously that's horseshit. Linux will never ever ever break through the glass ceiling and become idiot-proof non-tech user friendly like the botnets Winblows and macPrison.

There are users that copy shortcut from their desktop somewhere to make a backup. We laugh at them. I just copied symlink to my flash drive and realized it only when I copied it back to different computer and target didn't existed.

I am a Windows person. I always argue how great it is.

Well, not today.

I was today years old when I learned that you CANNOT uninstall store app via store ;p You need to go to settings / apps and functionality / your app / uninstall

The photo app (Yes the one bundled with win10) doesn't work if you use Hard drive compression AND it is a symlink for OneDrive (So you don't need to keep all photos on the drive). Fucking Paint works without problems.

Email client : If you alt+tab too fast after hitting 'Send email", there is 50% chances that email won't be send. Basiclly you need to hit "send" and wait until you see it in "sent" folder.

Well, as i'm ranting, here for Linux too :
I have a small ubuntu server VM, worked very well for last 6 months. Now "System in read only mode". Fucking apt-get upgrade fucked with something. I don't want to look, so I'll just rebuild a fresh vm.

And macOS should take sometyhing too : Who the fuck decided "enter" is for editing the name of file ?! really !

Well, ALL os are shit, all have downsides, I need my own OS. But I still want AA games... So windows for me.

OMG you fucking little cunt!

Previous issue with this co-worker we hate eat other but can maintain minimal contact due to covid. Last interaction was actually nice, we joked a bit.

He teaches me how to do the build and ‘updated’ the confluence page. By update he striked through one paragraph.

Been doing these for week and now others what builds done all the time and since I am not an asshole they can approach me to do this but now I spend all
day long doing builds.

Work on a classified app so it has to
be burned on a disk, taken to a ‘secret machine’ and deployed. Takes about an hour and people are like. Can you rebuild it? I forgot to commit something?

So I updated the page to flush out the directions. Did not remove one thing only added things simple things like do a ls -lah in the dist folder to make sure the are built correctly. Things like check to make the symlink works, bolded words.

He was not at standup so I figured he was out of the office today and was going to ask him to review tomorrow.

Fucker goes in to make changes while I am making changes and doesn’t think to msg me telling me?

He is removing things and moving things which is fine just let me know! What a dick!!!!!

Screenshot of all the activity today, I am
in blue. I will spend all day watching the page to make sure I get the last fucking edit!

Just gonna leave this here because I am too lazy to write a proper article for my website:
If anyone is trying to create a Vue.js website with Node.js backend do NOT use express-vue, it is unnecessarily complicated and broken. Instead use this method I found.
You will need:
- IntelliJ IDEA / WebStorm / other IDE supporting multiple modules per project and tasks
- Nodejs and npm
- vue-cli

Step by step:
1. Create new empty project
2. Add your frontend module using vue-cli generator
3. Add your backend module using Express generator
4. Run npm build in your frontend module once
5. Move or remove public folder in your backend module
6. Create a symlink from your backend module root called public pointing to dist folder in your frontend module root
7. Make sure to add "Run npm build" from frontend module to your "bin/www" task (default task for Express module)
8. Enjoy developing your REST API in Node/Express and your frontend in Vue.js with single-file components and it being served by the same server that is providing the backend.
(Since they are separate modules and you are not mixing webpack and Node/Express you can add ts-loader, stylus-loader, pug-loader or any other loaders without screwing anything up)

For deployment you just need to copy the contents of dist into public on the server. (and not upload the symlink)

This semester, we have a lecture called IT Security by a guy, who absolutely know his subject.
Nevertheless, he wanted to show us that sha256 is broken by an existing collision. (Google that, fellow ranters!)
There are two pdf files by google researchers, that show the caption „SHAttered“ both on different backgrounds, although they give the same SHA-hash.
He then tried to share us these two files by moodle and wondered, why he uploaded the same file twice.
Guess what happened? The moodle backend checks new uploaded files for their ... hash ... and then decides, weather to upload or the file is already existing. So, it did just a new symlink to the old file.
Ironic, that an exercise, that should show us sha collision failures on sha collision 😃

A word of advice to any new Arch user: don't use lxdm for your display manager. It's utter shit. The package hijacks the symlink of any other DM when you install it, and it refocuses the DM on a 5 second loop while it's running, making it a pain in the ass to use a VT. LightDM was broken temporarily so I had to install it, but frankly, I don't ever want to touch thiis piece of software again.

How to bring your zsh start-up time from 7s to 0.2s on macOS:
1. Don't call "brew info", piped to grep, piped to awk
2. Don't dynamically detect the current version of brew-installed packages
3. Don't call java_home
4. Actually don't do anything dynamically. Just symlink shit as they get updated

There you go. Don't be like me. Use the "brew --prefix" command and put its output in your .zshrc, instead of running it every time

What a delight to have to work on macOS. Not.
Took me two days to notarize our app bundle. The ultimate issue was a dead symlink inside the bundle which would make the codesign verify (with strict option!) fail, while verification of signing operation itself passed.
Notarization would just give generic error: not properly signes.

Why, Google? WHY?
My wife was annoyed, that her android image gallery showed the images she has sent via telegram, but not the ones, that she had received.

Stupidity no. 1: telegram puts received pictures into Pictures/Telegram on the internal memory. It seems like the default gallery apps don't take nested image-containing directories. As Pictures only contained the default Sony dummy images I moved them away.

Stupidity no. 2: both the receiving and sending image directory of Telegram is named "Telegram" and guess what... Android does not like that. Only the first ist shown (sent images).

Stupidity no. 3: to work around that, I installed the emulated shell to make a symlink named "Telegram-Received". Aaaand that requires root access.

Goddammit Google! She just wants to see our couple selfies that I sent her in her gallery!

Writing a library is so much nicer than writing an app. You can concentrate on the core thing you want to achieve and need to handle just a few files.

And you can still test it live even if it's a node module: just symlink it into the node_modules folder of your app.

So on my new position I get to work on Spark jobs. Never had to work with the infamous big data technologies. I never thought this would get SO frustrating for all the wrong reasons.

I'm currently trying to introduce integration tests for some Spark job I wrote. This isn't trivial though, as the data comes from several HBase tables. Mocking everything simply isn't feasible. So why not use the integrated HBaseTestingUtility? With it you can start a mini cluster that runs all nessecary services in the scope of your test.

Sounds great, eh? WRONG. Firstly the used mapr dependencies get in the way. The baked in configuration tries to automatically authenticate with your local cluster through Kerberos. Of course this doesn't work. And of course there is no way to reconfigure this as it happens IN A FUCKING STATIC BLOCK. AHHHH.

Ok. So after calming down I "simply" had to exclude all mapr dependencies and replace them with vanilla ones. After two days of dependency hell it FINALLY works!

...or does it? Well now we need test data. For that we got a map reduce algorithm that can import dumps. Sounds again, great, eh? WROOOONNNG.

The fucking map reduce mini cluster can't start, as it tries to write a symlink. Now take a wild guess what the sys admin here blocked. Yepp. TWO DAYS OF WORK RENDERED USELESS, BECAUSE OF SOME FUCKING SECURITY SETTING.

This is fine.