magic_lobster_party

I’ve seen things you people wouldn’t believe

I know

Job description: we’re looking for someone with experience in deploying cutting edge machine learning systems, preferably PhD.

Actual job: Excel spreadsheet

And after all that it is discovered that it was the wrong solution all along because the requirements were poorly specified, so the process must be started all over again

Product Manager: Make a step by step guide of how they think the lightbulb is going to be fixed without explicitly mentioning the broken lightbulb.

jQuery got popular because Internet Explorer, Firefox, Chrome and other browsers weren’t exactly cross compatible. Writing vanilla JS was risky business in that sense.

It also supported AJAX across all major browsers, which meant the website could make API requests without reloading the entire page. It was super revolutionary to press a button and it only changed a part of the page.

Then Angular and React took it a step forward and that’s where we are now.

JavaScript frameworks are invented because pure HTML and CSS suck for dynamically loaded pages, and vanilla JavaScript suck in general.

What’s not shown is that the car doesn’t have an engine. Management was really eager to release it to the customer. Don’t worry, it’s planned to get fixed later (spoiler: it’s never going to get fixed).

I was in on the crypto hate. I don’t really have a hate boner for AI.

Sure, there are things to dislike about AI, but it can be moderately useful. In contrast to crypto, AI is the hype because it’s widely used. Crypto was the hype because a few people hit the jackpot.

Rust and Cargo were built to be in a symbiosis with each other.

NPM is an afterthought of a rushed language.

It’s annoying when going through a list of multiple ”1 month ago” entries. Maybe I’m looking for an entry at a certain date. Aim with mouse, wait one second, repeat.

What I could easily visually identify in less than 1 second can take more than 10 seconds to find.

It also greatly increases the cognitive load of using the program. If there are many entries to look for, then it’s going to be difficult to keep all actual dates in memory.

”Where was the 14 April entry? I need to check again. Ah there it is! Now I need to compare it against the 30 April entry. Where was it again? It was just in front of me…”

Then mouse hover doesn’t work on mobile.

Then you haven’t seen bad documentation (or had that sex you regret).

Showing ”2 weeks ago” or ”1 month ago” instead of the actual date. ”1 month ago” can be anything between 30 days and 60 days ago.

There’s also ”we do machine learning”, which usually translates to ”someone trained an SVM model 10 years ago”.

The flaw also highlighted a social engineering exploit. It’s not the first time some vulnerability has entered open source software due to social pressure on the maintainer. Notably EventStream exploit.

This is difficult to account for. You can’t build automated tooling for social engineering exploits.

I don’t think the $500 million marketing budget would’ve worked if Java was introduced at a time other than the 90s.

The 80s would’ve been too early. It would just turn into a parenthesis in programming language history (next to smalltalk). The 00s would’ve been too late. It would’ve missed the dotcom bubble boat. Java came in the right time to become a dominant programming language.

I’m not saying the marketing didn’t have any influence. It probably had an big influence in which OOP language was selected for computer science education.

I don’t think OOP is as bad as many people make it out to be. It’s perfectly fine in moderation.

The problem is that it can lead to over engineered applications when abused.

Usually the teachers/professors/lecturers have no real world experience of software development besides the usually university projects

Adding to this: university projects are built on a relatively short timeframe compared to many industry projects. The growing pains that typically occur after a few years of continuous development is unlikely with the small scale of university projects.

I wouldn’t go to a university professor for advice on how to build a system that will last a decade of development.

AbstractionBubbleBuilderFactoryStrategyImplementation mind you

OOP was hype during the 90s. Schools adapted their curriculum to this trend. So they needed a programming language for this, and Java became the choice. C++ is too tricky as a first language.

The result is that a lot of people knew Java, which means it’s a good choice of language if you want to recruit programmers.

I believe most of Java’s success was luck. It released at the perfect time.