merc

From what I understand, it's using an LLM for coding, but taken to an extreme. Like, a regular programmer might use an LLM to help them with something, but they'll read through the code the LLM produces, make sure they understand it, tweak it wherever it's necessary, etc. A vibe coder might not even be a programmer, they just get the LLM to generate some code and they run the code to see if it does what they want. If it doesn't, they talk to the LLM some more and generate some more code. At no point do they actually read through the code and try to understand it. They just run the program and see if it does what they want.

Tests are probably both the best and worst things to use LLMs for.

They're the best because of all the boilerplate. Unit tests tend to have so much of that, setting things up and tearing it down. You want that to be as consistent as possible so that someone looking at it immediately understands what they're seeing.

OTOH, tests are also where you figure out how to attack your code from multiple angles. You really need to understand your code to think of all the ways it could fail. LLMs don't understand anything, so I'd never trust one to come up with a good set of things to test.

Also, LLMs are essentially designed to produce code that will pass a code review. It's output that is designed to look as realistic as possible. So, not only do you have to look through the code for flaws, any error is basically "camouflaged".

With a junior dev, sometimes their lack of experience is visible in the code. You can tell what to look at more closely based on where it looks like they're out of their comfort zone. Whereas an LLM is always 100% in its comfort zone, but has no clue what it's actually doing.

I think storyboards is a great example of how it could be used properly.

Storyboards are a great way for someone to communicate "this is how I want it to look" in a rough way. But, a storyboard will never show up in the final movie (except maybe fun clips during the credits or something). It's something that helps you on your way, but along the way 100% of it is replaced.

Similarly, the way I think of generative AI is that it's basically a really good props department.

In the past, if a props / graphics / FX department had to generate some text on a computer screen that looked like someone was Hacking the Planet they'd need to come up with something that looked completely realistic. But, it would either be something hand-crafted, or they'd just go grab some open-source file and spew it out on the screen. What generative AI does is that it digests vast amounts of data to be able to come up with something that looks realistic for the prompt it was given. For something like a hacking scene, an LLM can probably generate something that's actually much better than what the humans would make given the time and effort required. A hacking scene that a computer security professional would think is realistic is normally way beyond the required scope. But, an LLM can probably do one that is actually plausible for a computer security professional because of what that LLM has been trained on. But, it's still a prop. If there are any IP addresses or email addresses in the LLM-generated output they may or may not work. And, for a movie prop, it might actually be worse if they do work.

When you're asking an AI something like "What does a selection sort algorithm look like in Rust?", what you're really doing is asking "What does a realistic answer to that question look like?" You're basically asking for a prop.

Now, some props can be extremely realistic looking. Think of the cockpit of an airplane in a serious aviation drama. The props people will probably either build a very realistic cockpit, or maybe even buy one from a junkyard and fix it up. The prop will be realistic enough that even a pilot will look at it and say that it's correctly laid out and accurate. Similarly, if you ask an LLM to produce code for you, sometimes it will give you something that is realistic enough that it actually works.

Having said that, fundamentally, there's a difference between "What is the answer to this question?" and "What would a realistic answer to this question look like?" And that's the fundamental flaw of LLMs. Answering a question requires understanding the question. Simulating an answer just requires pattern matching.

I don't think most people understood what the "status quo" was until Trump started destroying it.

Yes, "status quo" is billionaires, and monopolies, and low-level corruption, and so-on.

But, "status quo" is also civil liberties, a somewhat functional department of education, not having the national guard chasing down brown people in the street, a bureau of labour statistics that didn't invent numbers to please the dear leader, cabinet secretaries that weren't actively working to destroy the thing they were assigned to lead, low-level crony and nepo corruption and not over-the-top obvious bribery, and so-on.

Heh, I bet if you're the CEO of a megacorp, you might not even carry your own electronics. You just have a gaggle of assistants around you who you bark orders at, and then they use their electronics to do something.

Which do you use most often?

A CEO might have a nice desktop, but is always out playing golf and so mostly uses his phone.

This is true up until a point, and then the pattern starts to reverse. Like, the receptionist isn't going to get 2 monitors. They're likely to get one monitor and a very old desktop, or an old laptop.

Edit: Also an intern / co-op student / work experience student, etc. is probably as low as you can go on the totem pole of office work. I bet in many cases they're not even assigned a permanent office / cubicle since they're expected to shadow / be mentored by a variety of people. As a result, they probably get a second-hand, used laptop.

And, if the company has retail sales, techs who do installations, etc. they're often very low on the totem pole, and they're often not getting a computer at all. Maybe in some cases they'd get a "work phone", so they'd have the same kind of equipment as the CEO, but effectively be at the opposite end of the pole from them.

Where I live, I still see people in a horse-drawn wagon. So, I guess horse-drawn wagons never died? It's only used for tourists and weddings, but that counts, right?

According to Tiobe, PHP was the programming language of the year in 2004. In 2010 it was number 3 in the top 10 programming languages. It's now out of the top 10 entirely. There really isn't a language that has completely disappeared. Mainframes are still programmed using COBOL, Scientists are still using FORTRAN, even Lisp, which has been around since the 1950s, is still going strong.

Maybe Actionscript counts as truly dead, since it was tied to Adobe Flash, and Flash is truly dead?

I have a lot of bad memories of PHP. It was, for a brief time, the main language I used, but it was so ugly and inconsistent. The only thing I loved about it, at the time, was that it wasn't Visual Basic. As bad as PHP was, at least I wasn't making web pages in that pile of hot garbage. But, I never felt joy writing something in PHP. At best it was a slog. At worst it was like pulling teeth.

Just about every other language has given me moments of fun. Original Javascript was a mess, but it already contained scheme-like features. It was sold as being an interpreted version of Java, but it had features that Java wouldn't have for at least a decade. C is a brutal and unforgiving language, but as long as you're not working with strings, it's great to have such low-level control over everything.

Maybe PHP has evolved like other languages, but I still am not interested in trying it out. Everything it was good at can be done better by other languages, and those are languages that give me joy, not pain. I hope it keeps dropping in the rankings so that people aren't exposed to it as one of their first languages.