Conversation

Notices

1. Embed this notice
   Julia Evans (b0rk@social.jvns.ca)'s status on Friday, 15-Aug-2025 15:45:47 JST Julia Evans

   are there any games that teach C pointers? like flexbox froggy for pointers

   • Embed this notice
     Alfred M. Szmidt (amszmidt@mastodon.social)'s status on Friday, 15-Aug-2025 15:45:40 JST Alfred M. Szmidt

     in reply to

     • Mark Levison
     • fcalva

     @mlevison >In an ideal world, only the device drivers are written in C.

     Why not something higher level with actual abstractions and semantics that make sense and allows you to write safe code? Why not Lisp?

     @fcalva @b0rk

   • Embed this notice
     Mark Levison (mlevison@agilealliance.social)'s status on Friday, 15-Aug-2025 15:45:41 JST Mark Levison

     in reply to

     • fcalva

     @fcalva @b0rk

     In an ideal world, only the device drivers are written in C.

     I've not done Operating System work for decades, so I'm a little rusty.

     I wonder how much of a modern OS could be written with languages that avoid pointer bugs.

   • Embed this notice
     fcalva (fcalva@cyberplace.social)'s status on Friday, 15-Aug-2025 15:45:43 JST fcalva

     in reply to

     • Mark Levison

     @mlevison @b0rk Rust has "castrated" pointers. Go isn't a systems language and isn't fast enough. And what other modern systems language ? Zig ? Odin ? "C2" ? All have pointers
     You just can't avoid pointers if you want to get close to the machine (aka fast).
     PS : I am not saying all code needs to or should be written in C, but saying "speed doesn't matter" is how you get the JS tower of babel
     Compiled Lisps for example, are only a few times slower than C, but are very nice to work with.

   • Embed this notice
     Mark Levison (mlevison@agilealliance.social)'s status on Friday, 15-Aug-2025 15:45:44 JST Mark Levison

     in reply to

     • fcalva

     @fcalva @b0rk

     Rust? Go? Or other modern systems language.

     In addition, 95+% of the code that people tell me needs to be "fast" isn't in a place where speed matters.

   • Embed this notice
     Mark Levison (mlevison@agilealliance.social)'s status on Friday, 15-Aug-2025 15:45:45 JST Mark Levison

     in reply to

     @b0rk

     I’m curious as to why? Isn’t it time we killed off pointers and all of related bugs or security flaws they created?

     I stopped using pointers in 1999. My mental health improved immediately.

   • Embed this notice
     fcalva (fcalva@cyberplace.social)'s status on Friday, 15-Aug-2025 15:45:45 JST fcalva

     in reply to

     • Mark Levison

     @mlevison @b0rk If you want to write fast code you are going to have to deal with pointers. Simple as that.

   • Embed this notice
     Alfred M. Szmidt (amszmidt@mastodon.social)'s status on Friday, 15-Aug-2025 15:54:42 JST Alfred M. Szmidt

     in reply to

     • Mark Levison
     • fcalva

     @fcalva >You just can't avoid pointers if you want to get close to the machine (aka fast).

     Absolute nonsense; in Lisp we avoid pointers completely and yet can out perform C (C is such a low benchmark of comparison it is silly).

     @mlevison @b0rk

   • Embed this notice
     Alfred M. Szmidt (amszmidt@mastodon.social)'s status on Friday, 15-Aug-2025 22:55:35 JST Alfred M. Szmidt

     in reply to

     • Mark Levison
     • fcalva

     @fcalva your making two fundamental mistakes, one cannot compare languages, and pointers have nothing to do with writing fast programs. Hence why what you say is total absurd nonsense.

     @mlevison @b0rk

   • Embed this notice
     fcalva (fcalva@cyberplace.social)'s status on Friday, 15-Aug-2025 22:55:37 JST fcalva

     in reply to

     • Mark Levison
     • Alfred M. Szmidt

     @amszmidt @mlevison @b0rk I mentionned compiled Lisps in the thread. And afaik they are still a few times slower than C when not calling any C libraries.

   • Embed this notice
     Alfred M. Szmidt (amszmidt@mastodon.social)'s status on Saturday, 16-Aug-2025 03:50:18 JST Alfred M. Szmidt

     in reply to

     • Mark Levison
     • fcalva

     @mlevison calling things as they are is essential to understanding. It is absurd bullshit to say that pointers are needed or required for a fast language, and then also in the same breath quote a language that doesn’t have pointers as a equal comparison,

     @fcalva

   • Embed this notice
     Mark Levison (mlevison@agilealliance.social)'s status on Saturday, 16-Aug-2025 03:50:20 JST Mark Levison

     in reply to

     • Alfred M. Szmidt
     • fcalva

     @amszmidt @fcalva

     Can I suggest that we both disagree with @fcalva

     Saying someone's position is absurd nonsense falls below the level of good debate.

     My position, is that we can build fast resilient software with languages that don't have C's pointers.