"Why are you working on C?"
I wrote this in an e-mail to people, but I figured I'd copy part of the response here since people keep asking.
…
We have people writing critical software. They are not migrating to new software anytime soon (modulo regulation-based incentives). But they have serious problems. Everything from vulnerabilities that are used by nation-state actors to quell dissidents, to not being able to change a typedef like intmax_t because the functions tied to it are baked into specific named symbols in an invisible way (ABI), to constantly seeing people's names getting butchered by Airlines, Databases, and Governments because they're using software that relies on the C locale and mangles names.
These are C problems. Not C++ problems. Not Java problems. Not Rust problems.
C problems.
My job is to solve C problems. That's the motivation. That's the coherent plan. When we stop having long-term, 20-to-40+ year problems, with 30+-year implemented existing practice that we never standardize despite it solving a wide variety of problems, that's when I'll stop writing C proposals.
ThePhD /
npub19p…u862l
2023-12-14 17:49:02
Author Public Key
npub19pz4w2q5uvs9jz8aydmgk9pgtgpmdy6380aekuy8gdvjfvztcx0qnu862lPublished at
2023-12-14 17:49:02Event JSON
{
"id": "7b23a521a93a096db1336b69fa98b7b9df238f5521cdc486625cb725d92809eb",
"pubkey": "2845572814e3205908fd23768b14285a03b693513bfb9b7087435924b04bc19e",
"created_at": 1702576142,
"kind": 1,
"tags": [
[
"proxy",
"https://pony.social/users/thephd/statuses/111580030097863666",
"activitypub"
]
],
"content": "\"Why are you working on C?\"\n\nI wrote this in an e-mail to people, but I figured I'd copy part of the response here since people keep asking.\n\n…\n\n We have people writing critical software. They are not migrating to new software anytime soon (modulo regulation-based incentives). But they have serious problems. Everything from vulnerabilities that are used by nation-state actors to quell dissidents, to not being able to change a typedef like intmax_t because the functions tied to it are baked into specific named symbols in an invisible way (ABI), to constantly seeing people's names getting butchered by Airlines, Databases, and Governments because they're using software that relies on the C locale and mangles names.\n\n These are C problems. Not C++ problems. Not Java problems. Not Rust problems.\n\n C problems.\n\n My job is to solve C problems. That's the motivation. That's the coherent plan. When we stop having long-term, 20-to-40+ year problems, with 30+-year implemented existing practice that we never standardize despite it solving a wide variety of problems, that's when I'll stop writing C proposals.",
"sig": "c8574bf5e96748c277f088680c6ee83decb657be737e6ace713c5b0fca7dd88e40794f0c19002ca8e0198f111343e6fe237bf9e45800ebc3902bd5202e4c6453"
}