The closest to vibecoding in the 90's was to open Borland's Turbo C help in any page, copy and paste the example and modify it until you understand it or until it did what you wanted.
You know, that's exactly how I learned to program.
I started up QBasic knowing nothing other than that it seemed like a thing for programming computers and programming seemed like a cool thing to do.
I typed in random words, and eventually I typed "screen". When I pushed enter, QBasic capitalized it, so it seemed important. I hit F1 and read the help. It made no sense, but the example ran and had other capitalized words so I could repeat the process.
Eventually I started making really terrible text-based Final Fantasy knock-offs.
Wow, that’s exactly my memory. As far as my family was concerned I was spending day and night in front of the “blue screen”. I got as far as programming a GUI by copying windows 95 pixel by pixel, text editor, fonts, cd player, minesweeper. I wish I had the code.
The closest to vibecoding in the 10's was to keep searching Stack Overflow for questions vaguely asking similar things as what you wanted and copy-paste things until it kind of worked.
Turbo pascal help for me! The polynomial example taught me how to use pointers. Before that, I could only use static arrays up to a certain length.
Learning about heap allocation was euphoric. I kept beaming because I had unlocked infinite memory, and people around me didn't get why I was such a happy teenager.
To be fair, I already knew about memory regions from PEEK/POKEing on a commodore as a child, but it was always static and pre-populated.
> Anything more complex than a few lines, you can just copy it from lib\ folder of the CD-ROM. There's a component for everything. You want to left-pad a string?
You know, honestly: thinking back to 1992 as a 13 year old, and downloading ircii source code and hacking on it (commenting out the 3 lines with build errors on AIX and then seeing what happened)... trying to add a function here or there or wire in a slash command...
This was a -great- experience. Inheriting code and not knowing what to do with it and trying to forensically triangulate what is going on and learning to read code in the process: this was the best way to learn. The argument that vibe coding is something like that is maybe one of the more hopeful arguments i've heard about it.
had a stint as a programmer for a dark hat org back in the day and hacked my way around rdesktop to make it async. When i say hack I literally hacked down the entire codebase until there was not much left except the login flow which consisted of.. a LOT of functions that needed to be made async. I did not have the slightest idea what I was looking at, looked like arcane magic to me but I eventually managed to make rdesktop into what was probably the fastest RDP bruteforcer there was thanks to boost.asio and chopping it up months on end. I remember the bruteforcer that was circling the forums made a thread for each client, ate up a lot of ram and CPU and it crashed a lot too. Mine wasn't even keeping the machine at 20% CPU, couple gigs of RAM but was topping the bandwidth of the server.
I'm not proud of creating a malevolent tool but am proud of the technical achievement of it considering I just finished high school.
I remember stumbling into coding by way of CSS and HTML. For advanced functionality well there was "scripts" that you copy and paste in and magically worked. The ultimate boss came when stumbling onto "php scripts" that you needed some particular cgi-bin thing to run. Follow the instructions, paste in the php script and it worked! This PHP stuff led to Wordpress. Follow those instructions and WOW mind blown, full end-to-end site with admin, login, database, CRUD, and it looked beautiful.
The problem was I couldn't actually understand any of the code. I learned CSS and HTML!
After many trials, I used what I knew and realized well whatever this PHP code is doing, I'm going to put it in a div with style="border:5px solid red" and see what it does that way.
Fast forward 2, 5, 10, 12 years and that's how I learned to be a programmer.
The problem is with vibe coding a beginner just gives a new prompt to edit it. In the past, the code you got was static so you had no choice but to learn it little by little if you wanted to change it.
> In the past, the code you got was static so you had no choice but to learn it little by little if you wanted to change it.
Seeing a lot of programmers being raised in the age of Stack Overflow, there was a ton of web developers who definitly didn't understand the code they copy-pasted from Stack Overflow, and some of them literally had a spray-and-pray approach to programming that makes you question how it worked sometimes when they had almost no understanding of what the code actually did.
But they managed to produce stuff that made webpages do things, so management was happy.
Fair point. There is much laziness in the package approach to software development, and unless you're doing pure assembly, you're no less guilty for vibe coding than the guy who plays npm like lego.
I think the most interesting point in the post is this one:
> I can create anything. Let me just take a look at that CD.
I like the idea of shifting the discussion from "how it is done" to "what are we doing".
Therefore, the point here is that we should do things the CD can't.
In that sense, the package thingy is better than LLMs. It gives you a directory that you can explore and the choice of not wasting time doing things that are already on the CD.
But then, you can say that directory is very large today. So large, we might need an index. And LLMs are just that. But if they're that, then there's some value in finding novel ways to glue things together.
At first glance, it seems to be a satire of vibe coding, but after reading it a second time, I find that the author is more sarcastic towards those who think AI will replace programmers. Did I get it right?
Both. Pointing out that "any old non-programmer will be able to program!" Has been a tech industry promise since at least the 90s, and despite the eternal lament of "every js dev is just gluing libraries together blindly" the reality continues to look like programmers are in no danger of being replaced
Oh man, remember CPAN? And COM and ActiveX and OOP? The attempts to make reusable things was definitely juiced by open source repositories of libraries out there for the using.
Readit News
Microsoft Quick Basic help was also gold.
I started up QBasic knowing nothing other than that it seemed like a thing for programming computers and programming seemed like a cool thing to do.
I typed in random words, and eventually I typed "screen". When I pushed enter, QBasic capitalized it, so it seemed important. I hit F1 and read the help. It made no sense, but the example ran and had other capitalized words so I could repeat the process.
Eventually I started making really terrible text-based Final Fantasy knock-offs.
Learning about heap allocation was euphoric. I kept beaming because I had unlocked infinite memory, and people around me didn't get why I was such a happy teenager.
To be fair, I already knew about memory regions from PEEK/POKEing on a commodore as a child, but it was always static and pre-populated.
This got me.
This was a -great- experience. Inheriting code and not knowing what to do with it and trying to forensically triangulate what is going on and learning to read code in the process: this was the best way to learn. The argument that vibe coding is something like that is maybe one of the more hopeful arguments i've heard about it.
I'm not proud of creating a malevolent tool but am proud of the technical achievement of it considering I just finished high school.
The problem was I couldn't actually understand any of the code. I learned CSS and HTML!
After many trials, I used what I knew and realized well whatever this PHP code is doing, I'm going to put it in a div with style="border:5px solid red" and see what it does that way.
Fast forward 2, 5, 10, 12 years and that's how I learned to be a programmer.
Seeing a lot of programmers being raised in the age of Stack Overflow, there was a ton of web developers who definitly didn't understand the code they copy-pasted from Stack Overflow, and some of them literally had a spray-and-pray approach to programming that makes you question how it worked sometimes when they had almost no understanding of what the code actually did.
But they managed to produce stuff that made webpages do things, so management was happy.
I think the most interesting point in the post is this one:
> I can create anything. Let me just take a look at that CD.
I like the idea of shifting the discussion from "how it is done" to "what are we doing".
Therefore, the point here is that we should do things the CD can't.
In that sense, the package thingy is better than LLMs. It gives you a directory that you can explore and the choice of not wasting time doing things that are already on the CD.
But then, you can say that directory is very large today. So large, we might need an index. And LLMs are just that. But if they're that, then there's some value in finding novel ways to glue things together.
And round and round we go.