Posts by zwol@masto.hackers.town
(DIR) Post #AtFyfIqbNmcuSJ5wEi by zwol@masto.hackers.town
2025-04-19T13:58:14Z
1 likes, 0 repeats
@lanodan @andrewrk @cadey I know part of the story: at some point even longer ago, they ran out of space in the original `struct _IO_FILE` and had to come up with a horrible hack to enlarge it without breaking old binaries. (See libio/old*.c in the glibc source code.) So, in the new enlarged version, they put a bunch of padding at the end to make room for future enlargements. The weird size calculation is meant to cause a compile error if they ever run out of space again.
(DIR) Post #Au2E08i7bog6DVn3JI by zwol@masto.hackers.town
2025-05-13T00:45:44Z
0 likes, 0 repeats
@foone if you're gonna have an index, it should be at the beginning. Putting the index at the end is an example of the very common design error of optimizing the format for ease of *writing* at the expense of ease of *reading*.
(DIR) Post #AuE6RfVz9pRYgg3BIm by zwol@masto.hackers.town
2025-05-18T17:07:30Z
0 likes, 1 repeats
May as well throw this query out to the fediverse.I am looking for *concrete examples* of code that works correctly when interpreted by Python 2.7 but *silently produces incorrect output* when interpreted by Python 3.x. I encountered such a thing about 10 years ago but didn't save it and have been unable to reconstruct it.All examples are good, but examples that produce output from which it's difficult or impossible to recover the correct output are better.#python #pycon
(DIR) Post #AuG3oxP0uIPLHL7eJE by zwol@masto.hackers.town
2025-05-18T00:23:00Z
0 likes, 1 repeats
Have you ever been annoyed by some Python code not stopping immediately when you press control-C? It's a common problem in scientific computing because of bugs in many of the compiled-code "extension modules" that accelerate this use of Python.I gave a talk today at #PyCon about what needs to happen to get those bugs fixed. I'm told a video will be available within a few days, but already you can see my slides and detailed notes at <https://research.owlfolio.org/pubs/2025-pyext-ctrlc-talk/>.
(DIR) Post #AuG3p6FQ1L52iBhc6S by zwol@masto.hackers.town
2025-05-18T00:30:40Z
0 likes, 0 repeats
If you can't see any "speaker notes" (approximate transcript of what I said) try a larger screen, the slide deck library's screen-size responsiveness is ... imperfect.
(DIR) Post #AvLJYHe18WRtto61B2 by zwol@masto.hackers.town
2025-06-21T03:14:56Z
0 likes, 1 repeats
This is on bus stop benches all over North Hollywood. I don't remember enough Hebrew to decode it. Does anyone know what it says?
(DIR) Post #Avcvd1ezjJwD5jNnhQ by zwol@masto.hackers.town
2025-06-29T02:29:36Z
0 likes, 0 repeats
@Mae @ariadne couple years ago I asked one of the Xorg core devs why they hadn't done this to provide a seamless migration path away from the old DDX drivers, and they said it was impossible and refused to explain why
(DIR) Post #AwIu7PeYoMTSclJuvw by zwol@masto.hackers.town
2025-07-19T21:12:05Z
0 likes, 0 repeats
Suppose you encountered a piece of software named "service-domowik", what would you think it did?If you don't know what "domowik" means, you may choose to answer either before or after looking it up.
(DIR) Post #AxXMSQ9Hv8mCyOGu2a by zwol@masto.hackers.town
2025-08-25T15:24:13Z
1 likes, 0 repeats
@jonny The issue is more basic than that; it's SQL injection but worse. _There is no quotation mechanism in the input to a language model_. All the prompts and input data get fed into the same statistical blender. So _even if you could_ filter out all the "conspiratorial behavior against the protagonist", you still could not stop the "data" inputs from interfering with the "control" inputs.And this is fundamental to how LLMs work. We _can't_ make there be more than one blender.
(DIR) Post #AxXMSYN3LqXaTOm0bg by zwol@masto.hackers.town
2025-08-25T15:30:28Z
0 likes, 0 repeats
@jonny (To be clear, you are also correct, and the effect you describe is probably more significant for LLMs as actually deployed today. I just don't want anyone getting ideas about hiring a whole bunch of humans to manually filter down the training set so that the robot will, they hope, behave like the perfect servant. Not only would that be an shitty thing to make people do for a living *and* incredibly expensive, it wouldn't work!)
(DIR) Post #AxfaGGJvacPpdf1ZEe by zwol@masto.hackers.town
2025-08-29T16:21:22Z
2 likes, 0 repeats
I have to rebuild a server again and so I've got a buncha heterodox hot takes about server configuration best practices rattling around in my head again. Who wants to hear them? Each like = 1 take, until I run out.
(DIR) Post #Axi5OiQwR3CA2ejfSy by zwol@masto.hackers.town
2025-08-30T00:41:35Z
0 likes, 0 repeats
Resuming the thread. Can I think of 36 more hot takes about server administration? We'll see.I probably should've said back at the beginning that I'm a strictly small-time sysadmin. I've got two cloud servers, one for public-facing stuff and one for private stuff, and there's only one other person who has write access to anything on them. My problems are simple, and my goal is to spend as little time on maintenance as possible. If your situation is different, you might not want to do like me.
(DIR) Post #Axi5OjVwPwmxOS5Atk by zwol@masto.hackers.town
2025-08-30T00:45:59Z
0 likes, 0 repeats
14. You should definitely have an automated process scanning your logfiles and notifying you when unusual things happen. But you need to spend time tuning it so it _only_ notifies you about the unusual stuff.For example, out of the box, logcheck will (or used to, anyway) tell you about all the _failed_ attempts to guess ssh passwords. You don't care. They failed. Filter that out.
(DIR) Post #Ay3GsodV7NGbmWK9DM by zwol@masto.hackers.town
2025-09-09T15:46:31Z
1 likes, 0 repeats
@GossiTheDog @microsoft I feel like this might be a good time for the "you should know that someone is signing your name to stupid letters" style of response (see https://www.snopes.com/fact-check/cleveland-browns-letters/)
(DIR) Post #Ay4sY895NUwtckcYuu by zwol@masto.hackers.town
2025-09-10T22:58:21Z
0 likes, 0 repeats
@foone @eniko yeah I was just going to say. The _original_ original is a series of edutainment games where you're the detective trying to catch her, so she _had_ to be the villain, but even there, she's got all the style. And I feel like each installment of spinoff ever since has moved her further in the direction of antihero or even straight up robin hood analogue.
(DIR) Post #AyEdAlfjxfm1PyWy0m by zwol@masto.hackers.town
2025-09-15T14:11:58Z
2 likes, 0 repeats
A tiny hill I will die on: It is *wrong* to print out numeric errno codes. If you feel it is necessary to print a mysterious code in addition to the human-readable message you get from strerror(), print the Exxx name. Yeah, C doesn't give you any good way to do that. Maybe you don't need to print a mysterious code at all?You know what you *should* print, though? The name! of the thing! you couldn't do something with! (usually but not always a filename)
(DIR) Post #AyEdAtVOq0FtiDkp0K by zwol@masto.hackers.town
2025-09-15T14:17:20Z
0 likes, 0 repeats
this grump brought to you this morning bylet mut msg = ioerr.to_string();if let Some(cut) = msg.rfind(" (os error ") { msg.truncate(cut);}
(DIR) Post #AyHZdcD6Gh5txW3ZqK by zwol@masto.hackers.town
2025-09-17T01:44:38Z
1 likes, 0 repeats
@lanodan @alanc @jmc Yeah, my brief time teaching was much later and all the students were on regular Linux (glibc, not musl). Though, if the Sun clade has had "Arithmetic exception" forever and musl uses it too, that's all the excuse I need to go change glibc myself :)
(DIR) Post #AyVImgcwZY3NBVcwlM by zwol@masto.hackers.town
2025-09-23T16:55:00Z
0 likes, 0 repeats
@neauoire It is also important to recognize that simplicity itself can become a hindrance. There's a really basic result in dynamic control theory that says any system that interacts with its environment must have at least as much internal complexity as the complexity of all its inputs put together, or else it won't be able to handle every possible input. This is why "let's start over and keep it simple this time" efforts can never succeed.