rurban

34 ms·

331.
by rurban 6mo ago
... and most of this work is done by other distros and maintainers. Starting with binutils
332.
by rurban 6mo ago
England gave Pakistan the nuclear bomb. Churchill's "greatest" idea
333.
by rurban 6mo ago
If politicians would be effectively controlled by the government and not by some independent party those mysterious, oops, accidentally deleted it problems would increase.
334.
by rurban 6mo ago
That's what pony did also. Operator preceding rules are too arcane, such as the need for manual memory management.
335.
by rurban 6mo ago
But those are updated automatically. It's unlike Windows or Linux, where the user decides when to update. cPanel updates are decided by cPanel
336.
by rurban 6mo ago
I am in an European AI lab (first a startup, then bought up by some big military company) and I never heard of regulations nor hard copy folders. This shit is for the old folks at cars or airplanes. Your problems might be your European mana
337.
by rurban 6mo ago
Normal engineering practices as thought since the 70ies. Break the problems up into manageable pieces. Make a plan, have tests to verify the outcome, implement that part. Rinse and repeat. Have integration tests.
338.
by rurban 6mo ago
Well, also Unicode identifiers, a C11 spec bug, nobody cares to fix. Still in C26, because "users expect Unicode stability", esp. it's bugs.
339.
by rurban 6mo ago
An AI badge would be easier and shorter
340.
by rurban 6mo ago
I'm working with Deepseek for a few weeks with opencode, and there are no desires left. claude remote maybe, but opencode web is good enough.
341.
by rurban 6mo ago
Sure, another Rupert Sheldrake
342.
by rurban 6mo ago
They are no way as good as Opus yet. But Sonnet, yes. Using all in real life.
343.
by rurban 6mo ago
So his compiler cannot detect the unsigned overflows and instead chooses to call it a user mistake! Sizes and indices of course need to be unsigned, and any self respecting compiler should warn about dangerous usage.
344.
by rurban 6mo ago
Entitled maintainers are much worse. One such maintainer will stagnate the whole project, one such user is just blocked or ignored. Nobody cares
345.
by rurban 6mo ago
Unfortunately not. I'm using plain kimi, opencode (with deepseek, gpt, minmax, whatever) and claude. claude is the best, but only for some hours. The trick is to get a good AGENTS.md file, good test cases and test runner to repro, like
346.
by rurban 6mo ago
Well, I'm using all the top models extensively on the very same codebase, my new compiler. I use deepseek for it's cheap API costs, when kimi, claude and codex are in their overbudget phase. I asked deepseek V4 Pro for an estimate
347.
by rurban 6mo ago
I think I got the smallest: $ podman image list | grep cross docker.io/gotson/crossbuild latest d96ea9b7054b 3 years ago 6.71 GB used to cross-build to darwin.
348.
by rurban 6mo ago
Good luck begging Larry to fix this :) but he is still active from time to time. But patch? Andreas Gruenbacher is active currently. Or git can change to differentiate the message prefix better
349.
by rurban 6mo ago
Sure, but he still came up with a 2005 blog post, and attached a 2026 to it. No optimizing C compiler cares for the 2nd arg, when it's a register anyway. And if the 1st is constant, the dead branch is folded away. So the 2nd arg is dea
350.
by rurban 6mo ago
Of which decade is this post? I cannot think of any modern architecture which still passes args on the stack. Itanium? Stone age
351.
by rurban 6mo ago
Because I didnt benchmark -O1 yet, the opt pass. It has to do much less work on pure functions, so even compile time is faster then. Still in work...
352.
by rurban 6mo ago
It still can be done much cheaper. I just fixed my excessively cheap register allocator for my c compiler rcc. The register allocator is a simple first-fit bitmask with no spilling to stack except for the two predefined spill slots. Only if
353.
by rurban 6mo ago
As long as GitHub Actions is free for Open Source, I'll keep using it. There's nothing comparable for free yet. Of course Github using now Azure as a testbed is problematic. Since Azure is mismanaged and doomed https://
354.
by rurban 6mo ago
Syncing is trivial, the CI is the deal. GH actions are the best option still. Neither the FSF nor any other OSS lab didn't come up with a proper CI for us open source maintainers. The CI load also increased massively since.
355.
by rurban 6mo ago
It's a testbed for Azure since they started GitHub Actions. Since Azure is mismanaged and doomed https://news.ycombinator.com/item?id=47616242 such outages are normal. I think they hired cheap staff to manually fix ove
356.
by rurban 6mo ago
Do that will only be possible with something like better 3D NAND flash memory, needs a new hardware. People are already trying to bring that the market. Contemplated taking a compiler position in such a company.
357.
by rurban 6mo ago
tcc also supports a binary float extension. https://github.com/TinyCC/tinycc/blob/9b8765d8baaeb2a16112d6... long double la0 = 0B.110101100P12L;
358.
by rurban 6mo ago
It's a Sydney company. But parts are now manifactured in China. So who added this backdoor to listen to its customers? The CIA has similar companies doing it over popular loudspeakers. Could also be the Chinese.
359.
by rurban 6mo ago
April 2. Was this an April 1 joke?
360.
by rurban 6mo ago
Sucks to need to add unsafe new constructs the language authors didn't foresee, and still call it safe. Should have been forseen.

More ›