Dijkstra on foolishness of Natural Language Programming
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Related videos
What’s the best programming language for AI?
Theo - t3․gg
75.8k views
The Only Programming Languages Worth Learning in 2026
Tech With Tim
81.0k views
The Evolution of Programming | The Standup
ThePrimeTime
93.3k views
I'm switching languages again...
ThePrimeTime
165.4k views
Programming From Prison
ThePrimeTime
66.2k views
Progressive JSON
ThePrimeTime
239.5k views
Be A Great Programmer
ThePrimeTime
220.0k views
A Rant About Professional Programming - Prime Reacts
ThePrimeTime
151.2k views
The End Of Programming As We Know It
ThePrimeTime
447.2k views
Picking A Language In 2025
ThePrimeTime
253.3k views
Top Comments (10)
We already have an entire industry devoted to debugging natural-language code. They’re called lawyers.
For a second I though Dijkstra was alive.
at some point when AI prompt coding is normalized, someone's going to say "english is so verbose, wouldn't it be great if there was a shorthand way to describe things?" and then we'll be using code to generate the prompts.. to generate the code. full circle!
The Dijkstra are the friends we made along the path
People who have never coded in their life will look at the weird symbols and keywords and think "that's the hard thing about it, surely! What if we got rid of that?" ... but that's the easy part... the very tip of the iceberg.
"It sounds sensible provided you blade the obligation to use a formal symbolism as the source of your difficulties" is Dijkstra reaching through time to say "Skill Issue."
Say what? Dijkstra? I'm all ears! edit: one of his quotes has been stuck in my head for the past 30 years: "The question whether machines can think is as silly as asking if submarines can swim". #micdrop
Engineers once observed that people rarely use the last bit of a pencil and they decided to not put graphite in there. Years later, another group of engineers looked at this “useless” part with no graphite at the end and cut it off.
This is essentially an intellectual “no free lunch” rule. Same happens in other programming-adjacent disciplines like requirements. Folks are so reluctant to put in the intellectual effort to think through how something really ought to work. We have a LOT of cognitive/biological bias toward preserving energy, to not think too hard about stuff (see Khaneman), and yet we owe most of our progress as a society to our occasional willingness to do so.
I worked with a proprietary language called SanScript that started as a natural language syntax with the goal that "accountants should be able to write code". What we quickly found out was: 1. Accountants do not want to write code 2. You really do *not* want accountants writing code.
Unlock the Data Inside
Turn Videos into Knowledge
- Get FREE 10/day: transcripts, summaries, chats
- Chat with videos, export text & PDF
- $1 free API credit for RAG, chatbots & research
Free forever plan • All features unlocked
Top Comments (10)
We already have an entire industry devoted to debugging natural-language code. They’re called lawyers.
For a second I though Dijkstra was alive.
at some point when AI prompt coding is normalized, someone's going to say "english is so verbose, wouldn't it be great if there was a shorthand way to describe things?" and then we'll be using code to generate the prompts.. to generate the code. full circle!
The Dijkstra are the friends we made along the path
People who have never coded in their life will look at the weird symbols and keywords and think "that's the hard thing about it, surely! What if we got rid of that?" ... but that's the easy part... the very tip of the iceberg.
"It sounds sensible provided you blade the obligation to use a formal symbolism as the source of your difficulties" is Dijkstra reaching through time to say "Skill Issue."
Say what? Dijkstra? I'm all ears! edit: one of his quotes has been stuck in my head for the past 30 years: "The question whether machines can think is as silly as asking if submarines can swim". #micdrop
Engineers once observed that people rarely use the last bit of a pencil and they decided to not put graphite in there. Years later, another group of engineers looked at this “useless” part with no graphite at the end and cut it off.
This is essentially an intellectual “no free lunch” rule. Same happens in other programming-adjacent disciplines like requirements. Folks are so reluctant to put in the intellectual effort to think through how something really ought to work. We have a LOT of cognitive/biological bias toward preserving energy, to not think too hard about stuff (see Khaneman), and yet we owe most of our progress as a society to our occasional willingness to do so.
I worked with a proprietary language called SanScript that started as a natural language syntax with the goal that "accountants should be able to write code". What we quickly found out was: 1. Accountants do not want to write code 2. You really do *not* want accountants writing code.