The Impossible Optimization, and the Metaprogramming to Achieve It | Discussion
The Impossible Optimization, and the Metaprogramming To Achieve It
verdagon.dev
November 1, 2025 at 10:20 AM
The Impossible Optimization, and the Metaprogramming to Achieve It | Discussion
Here's a fun bit of Ruby metaprogramming: get the name of the method that called the current method.
`caller_locations(1, 1).first.label.split(" ").last`
docs.ruby-lang.org/en/master/Ke...
🧵 1/3
`caller_locations(1, 1).first.label.split(" ").last`
docs.ruby-lang.org/en/master/Ke...
🧵 1/3
module Kernel - Documentation for Ruby 3.5
module Kernel: The Kernel module is included by class Object, so its methods are available in every Ruby object. The Kernel instance methods are documented in class
docs.ruby-lang.org
October 29, 2025 at 4:24 PM
Here's a fun bit of Ruby metaprogramming: get the name of the method that called the current method.
`caller_locations(1, 1).first.label.split(" ").last`
docs.ruby-lang.org/en/master/Ke...
🧵 1/3
`caller_locations(1, 1).first.label.split(" ").last`
docs.ruby-lang.org/en/master/Ke...
🧵 1/3
Yeah, metaprogramming in both Rust and C++ gets pretty hairy.
I can't remember what it was, but I recall being very impressed at Zig's comptime but then I read something that sort of soured it for me... I'm frustrated that I don't remember what it was. It's a while ago.
I can't remember what it was, but I recall being very impressed at Zig's comptime but then I read something that sort of soured it for me... I'm frustrated that I don't remember what it was. It's a while ago.
October 29, 2025 at 3:51 PM
Yeah, metaprogramming in both Rust and C++ gets pretty hairy.
I can't remember what it was, but I recall being very impressed at Zig's comptime but then I read something that sort of soured it for me... I'm frustrated that I don't remember what it was. It's a while ago.
I can't remember what it was, but I recall being very impressed at Zig's comptime but then I read something that sort of soured it for me... I'm frustrated that I don't remember what it was. It's a while ago.
Also "template metaprogramming" is one thing, I'm not saying templates and metaprogramming separately.
October 23, 2025 at 11:11 AM
Also "template metaprogramming" is one thing, I'm not saying templates and metaprogramming separately.
Ok, you ask a question. I then respond at the level I operate professionally. Just started throwing polymorphic multiple-inheritance template-metaprogramming examples your way, because it's what my brain is spitting out
October 23, 2025 at 3:09 AM
Ok, you ask a question. I then respond at the level I operate professionally. Just started throwing polymorphic multiple-inheritance template-metaprogramming examples your way, because it's what my brain is spitting out
Lisp descendants are awesome because they give you incredibly powerful tools like metaprogramming in order to solve all the intractable problems caused by the use of metaprogramming
October 16, 2025 at 9:52 PM
Lisp descendants are awesome because they give you incredibly powerful tools like metaprogramming in order to solve all the intractable problems caused by the use of metaprogramming
I gave a presentation tonight about MetaProgramming at JS Montreal. Got good feedback.
arnaudbuchholz.github.io/decks/MetaPr...
arnaudbuchholz.github.io/decks/MetaPr...
MetaProgramming
arnaudbuchholz.github.io
October 15, 2025 at 12:47 AM
I gave a presentation tonight about MetaProgramming at JS Montreal. Got good feedback.
arnaudbuchholz.github.io/decks/MetaPr...
arnaudbuchholz.github.io/decks/MetaPr...
Working in C++ one of the few big things I miss for metaprogramming compared to Rust is rustdoc. Headers don't work well as your lib docs when your interfaces are generated by a ton of magic, but big interfaces generated by Rust macros can still result in readable docs.
October 12, 2025 at 9:32 PM
Working in C++ one of the few big things I miss for metaprogramming compared to Rust is rustdoc. Headers don't work well as your lib docs when your interfaces are generated by a ton of magic, but big interfaces generated by Rust macros can still result in readable docs.
“Effective LLM use is metaprogramming for wordcels and that’s why almost nobody is doing it: because they either don’t know metaprogramming or they are not wordcels”
a sentence I just uttered at work, help
a sentence I just uttered at work, help
September 26, 2025 at 6:15 PM
“Effective LLM use is metaprogramming for wordcels and that’s why almost nobody is doing it: because they either don’t know metaprogramming or they are not wordcels”
a sentence I just uttered at work, help
a sentence I just uttered at work, help
Metaprogramming? You mean working at Facebook?
September 26, 2025 at 6:40 PM
Metaprogramming? You mean working at Facebook?
to be fair empty structs are much less useful in C where you don't have tag dispatching, "functors", or template metaprogramming
September 24, 2025 at 8:34 PM
to be fair empty structs are much less useful in C where you don't have tag dispatching, "functors", or template metaprogramming
We really loved this series of tutorials on #metaprogramming in #LeanLang by Heather Macbeth. It's a great intro to a complex topic for novice users of #LeanProver!
Pt 1: youtube.com/watch?v=cKvg...
Pt 2: youtube.com/watch?v=5er4...
Pt 3: youtube.com/watch?v=TJ8T...
Pt 1: youtube.com/watch?v=cKvg...
Pt 2: youtube.com/watch?v=5er4...
Pt 3: youtube.com/watch?v=TJ8T...
September 24, 2025 at 7:11 PM
We really loved this series of tutorials on #metaprogramming in #LeanLang by Heather Macbeth. It's a great intro to a complex topic for novice users of #LeanProver!
Pt 1: youtube.com/watch?v=cKvg...
Pt 2: youtube.com/watch?v=5er4...
Pt 3: youtube.com/watch?v=TJ8T...
Pt 1: youtube.com/watch?v=cKvg...
Pt 2: youtube.com/watch?v=5er4...
Pt 3: youtube.com/watch?v=TJ8T...
I think any language has to accept that you either implement strong metaprogramming or people will hack the worst metaprogramming possible on top of your language
September 6, 2024 at 5:14 PM
I think any language has to accept that you either implement strong metaprogramming or people will hack the worst metaprogramming possible on top of your language
About the template metaprogramming example — referencing concatenated constexpr string literals is possible since C++11, but there it requires a much more sophisticated approach: github.com/xamidi/pmGen...
April 2, 2024 at 3:49 AM
About the template metaprogramming example — referencing concatenated constexpr string literals is possible since C++11, but there it requires a much more sophisticated approach: github.com/xamidi/pmGen...
metaprogramming is peak. don't let anyone tell you otherwise
February 19, 2025 at 9:52 AM
metaprogramming is peak. don't let anyone tell you otherwise
Cognition: A New Antisyntax Language Redefining Metaprogramming Discussion
Cognition
Other languages are inflexible and broken. Let's fix that.
ret2pop.nullring.xyz
May 2, 2024 at 2:00 AM
Cognition: A New Antisyntax Language Redefining Metaprogramming Discussion
I was just thinking about Ruby (the programming language). I quite liked it, and there were some really neat metaprogramming stuff it could do. But I think what did it in is people drank too much of the metaprogramming Kool-Aid and started fetishizing some misplaced concept of elegance...
December 13, 2023 at 3:09 AM
I was just thinking about Ruby (the programming language). I quite liked it, and there were some really neat metaprogramming stuff it could do. But I think what did it in is people drank too much of the metaprogramming Kool-Aid and started fetishizing some misplaced concept of elegance...
statically typed metaprogramming is cool as fuck
February 5, 2025 at 9:12 PM
statically typed metaprogramming is cool as fuck
Metaprogramming in ES6: All about Symbols and why they're awesome http://blog.keithcirkel.co.uk/metaprogramming-in-es6-symbols/ #es6 #javascript
November 21, 2024 at 11:56 PM
Metaprogramming in ES6: All about Symbols and why they're awesome http://blog.keithcirkel.co.uk/metaprogramming-in-es6-symbols/ #es6 #javascript
Okay what we actually need is to make executives shut up, surprising no one. Excellent read this-a-way.
And it only reinforces my parallel with Metaprogramming For Madmen: the approach is not viable for the goal Phil Spencer claims it could have. It is for other things!
bsky.app/profile/mtrc...
And it only reinforces my parallel with Metaprogramming For Madmen: the approach is not viable for the goal Phil Spencer claims it could have. It is for other things!
bsky.app/profile/mtrc...
New Post - Muse: Before You Tweet
Hello! Microsoft's new Muse system isn't trying to generate game ideas.
If you want a concise summary of what the paper tries to do, and a few notes and criticisms about the project, I wrote something about it.
www.possibilityspace.org/blog-before-...
Hello! Microsoft's new Muse system isn't trying to generate game ideas.
If you want a concise summary of what the paper tries to do, and a few notes and criticisms about the project, I wrote something about it.
www.possibilityspace.org/blog-before-...
February 19, 2025 at 9:17 PM
Okay what we actually need is to make executives shut up, surprising no one. Excellent read this-a-way.
And it only reinforces my parallel with Metaprogramming For Madmen: the approach is not viable for the goal Phil Spencer claims it could have. It is for other things!
bsky.app/profile/mtrc...
And it only reinforces my parallel with Metaprogramming For Madmen: the approach is not viable for the goal Phil Spencer claims it could have. It is for other things!
bsky.app/profile/mtrc...
Chelsea. Dear the whatever from high atop the thing.
I had a head full of template metaprogramming and now I have a head full of.... something else.
I had a head full of template metaprogramming and now I have a head full of.... something else.
June 3, 2024 at 5:16 PM
Chelsea. Dear the whatever from high atop the thing.
I had a head full of template metaprogramming and now I have a head full of.... something else.
I had a head full of template metaprogramming and now I have a head full of.... something else.