Why the fuck why are tslint and prettier are always sucking their dicks and fuck each others asses.

Do you know this moment when you think:.... 🤔 They have millions of downloads, why do I bother formatting my code so much. Mabye all this people are smarter and saving so much time.

Then you set up eslint and prettier adjusting 10000k rules just to find eslint and prettier five minutes later in one file fucking their asses again how to indent on save😠😱

I want to rant here but I have a lot of backlog. They’re all pretty deep for refinement. So I want to put them all in one sprint. Because I don’t want a marathon of ranting.
But I do realise in IT sprints never end, and eventually turn out to be a marathon of sprints. I came to a point to deliver over 100 story points in one sprint. Maybe I shall write a book instead of exploding this app.
I could have never imagined the dark side of this occupation.
However I love coding so much I’ll brush the backlog under the carpet😏

But you know there are always saviour heros wanted for refactoring. This honour goes to:

The arrogant guys who think they’re a genius when their code compiles.
The insecure guys who want to overpower the next available when their code doesn’t compile.
The egoists who like to underestimate and show of, where their faulty biased googles display a little girl instead of a developer.
The aggressives when they are invited to the reality and kindly offered to sit back on their place.

I hope this rant wouldn’t ouch anyone. If it does, not sorry, the message is delivered.

If there’s an offence or reaction to this refactoring job, the 4th one would be offered to clean the mess.

I finally got the lstm to a training and validation loss of < 0.05 for predicting the digits of a semiprime's factors.

I used selu activation with lecun normal initialization on a dense decoder, and compiled the model with Adam as the optimizer using mean squared error.

Selu is self-normalizing, meaning it tends to mean 0 and preserves a standard deviation of one, so it eliminates the exploding/vanishing gradient problem. And I can get away with this specifically because selu *only* works on dense layers.

I chose Adam, even though this isn't a spare problem, because Adam excels on noisy problems and non-stationary objectives (definitely this), and because adam typically doesn't require a lot of hyperparameter tuning its ideal here, especially considering because I don't know what the hyperparameters should be to begin with.

I did work out some general guidelines on training quantity vs validation, etc.
The initial set wasn't huge or anything, roughly 110k pairs for training.

It converged pretty quick all things considered, and to the low loss like I mentioned, but even then the system always outputs the same result, regardless of the input, so obviously I'm doing something incorrectly.

The effectiveness of this approach for training and validation makes me question if I haven't got something wildly wrong. Still exploring though and figuring out how to get my answers back out. I'm hoping I just fucked up the output, and not the input as well.

How can a novel emerging challenger software (written in Rust) take me 4 hours to install (still ongoing)?

Today I have decided to give Pijul a go. Pijul describes itself as a theory-sound alternative to Git, which I have wanted to get away from for a while now, due to various reasons -- many of which I saw Pijul advertise to have solved on design level.

So I set away a day to learn Pijul, today. Well, 4 hours after I sat down -- after a number of hilariously wonky failures of "Rust ecosystem" to do the right thing as I had to install Rust with some shell one-liners those insane wizards recommend for installation process (all in the name of "stability but not stagnation") -- Pijul has now been installing with the blasted `cargo` for an hour now (that's after 3 hours of getting to the point where `cargo install pijul` stopped exploding in my face) -- telling me I only have 40 crates more to install. Are they throttling me, perhaps? I don't care -- I should have been installing Pijul from a repository in accordance with my Linux distribution, or -- at worst -- download a BLOODY COMPILED PROGRAM IMAGE.

What is it with the hipster developers today? Everything they get of tools, they subsume and churn out intricate complexities the likes of which we hadn't seen yesterday. Tell me fellow developers who think installation of your software has to require three and a half novel "installation solutions" to which I can't be arsed to be made privy -- do you think your life today is easier than, I don't know -- wrangling with a Makefile and a C compiler (which today thankfully can do rather good job of standards compliance)?

I mean I wouldn't mind Pijul being written in Rust -- but it turns out Rust's advertised elegancy in practice is wrapped in so much "giftwrap" I feel like what desire I had to learn Rust myself, I'll stear well clear.

Here's an advice for developers in general -- an advice continiously ignored for decades -- stop blowing your original scope of delivery in auxilary packages you think you need to reinvent just because you can or because your mom is out of town! For programming languages like Rust this most certainly entails NOT writing your own package manager, with its own package delivery mechanism that has its own configuration file format and virtual machine to configure dependency resolution or what have you!

You wanted to write a programming language that has novel features you think we need? Fine -- write one and stop there. Watch it grow, and watch people who are busy working on other parts (scopes) of software to integrate your offer.

What a shitshow. Stop smuggling alternative package managers, installers, and discombulators with your actual product -- I only want the latter, I don't want the rest of your damn piping, walls, roof and a cathedral on top of it!

Don't be that guy starting with a pin, and ending up with a fucking diorama miniature of a pig farm in Netherlands. Jesus.