Gillitrut
Reading from the golden book under bright red stars
No bio...
User ID: 863
Gillitrut
Reading from the golden book under bright red stars
11hr ago
My hot take is that too many programmers use garbage collection as a crutch. GCs free you from some very specific work having to do with allocating and freeing memory but they are not a "get out of jail free" card for ever thinking about memory management or object lifetime again. Can think of a lot of examples of my own work in C# where people write inefficient code in hot paths without worrying about it because they let the garbage collector clean up after them.
Gillitrut
Reading from the golden book under bright red stars
17hr ago
This is getting off topic, but I thoroughly enjoy reading Raymond Chen's blog Old New Thing for the many stories of Windows bugs or implementation details or programmer misuses that later became compatibility constraints. When you upgrade your operating system and your Favorite Program stops working people rarely blame their Favorite Program even if it is the thing that was doing something unsupported!
Gillitrut
Reading from the golden book under bright red stars
19hr ago
I don't know. I find this a topic that it's pretty easy to be nuanced about. Different languages attempt to provide different guarantees to the programmer during their operation. To provide those guarantees they have to be able to understand the code and prove the code satisfies those guarantees. Most such languages provide ways to disable checking those guarantees for particular code sections on the assumption that you, the programmer, have information the compiler lacks that things will work without the compiler having to check. If you, the programmer, tell the compiler you know better and then turn out to be wrong I think it's fine to blame the programmer.
I think everyone has, in their mind, a different idea about the extent to which buggy code should be caught by the compiler and these ideas are what inform what side of the blame the programmer/blame the compiler distinction you fall on. As an example: In college a friend and I had to write some networking libraries in C. At the time we didn't use any fancy editors or anything, just good old gedit and gcc. My friend was writing a function that was supposed to perform an arithmetic operation and return the output but every time he ran it he got a different (implausible) result, even with the same inputs. What was happening is that he had accidentally omitted the return statement for his function, so he was getting back some random garbage from memory on every run. Should the C compiler let you declare a function that returns a value and then let you omit the return statement? Is that mistake your fault or the language's fault? Formally doing this is undefined behavior but that does not always mean crash!
Well that's technically not true, they did. It's just that calling .unwrap(), a function which will immediately abort the application on error, counts as "handling" the error. In other words, the path of least resistance is not to actually handle the error, but to crash. I argue that this isn't a better outcome than what would have happened in C, which would also be to crash. Sure, the crash won't be a segfault in Rust, but that doesn't matter if half the Internet dies.
In this case I find the behavior of Option<T>.unwrap() unintuitive, but I am also coming from the perspective of exception-based error handling. As an analogy, C#'s Nullable<T>.Value will throw an exception if the nullable is actually null. That option obviously isn't available in a no-exception world. Maybe the default behavior should be more like the behavior with the try trait such that it returns the error instead of panic? Then let the programmer panic if the value is error, although that introduces another layer of error checking!
This month, a CVE was filed in the Rust part of the Linux kernel, and it turned out to be a memory corruption vulnerability, ironically enough. "But how could this happen?" Rust has these things called unsafe blocks that let you do unsafe memory operations, closer to what you would be allowed to do in C (though granted, I have heard convincing arguments that unsafe Rust is still generally safer than C). So the path of least resistance is not to do things the safest way, but to just surround everything in unsafe if you get tired of fighting the borrow checker.
I'm a little unsure of the criticism here of Rust as a language. Is it that unsafe exists? Presumably all the code that is not in an unsafe block has guarantees that equivalent C code would not. Is that not a benefit? Is the worst case here you wrap all your Rust code in unsafe and then you end up... as good as C?
To be clear, I'm not saying that these incidents alone mean Rust is a bad choice for anything, ever. I'm not saying Cloudflare or Linux shouldn't use Rust. I'm not telling people what they should or shouldn't use. I'm just pointing out the double standards. Rust people can attack C all day using one set of (IMO, entirely justified) standards, but when they are confronted with these incidents, they suddenly switch to another set of standards. Or to put it more clearly, they have a motte and bailey. Motte: "Rust can't prevent shitty programmers from writing shitty code." Bailey: "C is unsafe, because of all the memory unsafe code people have written, and we should rewrite everything in Rust to fix all of it!"
I think there is a more productive discussion here about how language features and guarantees can help protect against writing buggy code and potentially making it easier to review code for bugs. I suppose I think of it by analogy to Typescript and Javascript. All Javascript is valid Typescript but Typescript needs to be compiled to Javascript. That compilation, in my experience, helps avoid whole classes of errors due to the lack of typing in Javascript. Sure you can write Javascript that just doesn't have those errors, and most people do, but Typescript renders them inexpressible. Similarly so for C and (non-unsafe) Rust.
Gillitrut
Reading from the golden book under bright red stars
1d ago
I cannot speak for all managers but this description does not sound like any manager I have worked for. My managers have done key work in prioritizing the work for me and other team members. Coordinating work across teams. Translating high level strategy shifts from higher level executives into concrete terms for people like me. Their role has been very, obviously, valuable.
Previously, if someone took the time out of their day to physically travel to you and tell you something, you could reasonably expect it to be important.
It is kind of funny to read this in a world where Office Space exists. It's a satire but I am under the impression the phenomenon it satirizes was real. Was it important to put the cover sheets on the TPS reports?
Gillitrut
Reading from the golden book under bright red stars
2d ago
Sure, but what fraction of jobs does that describe? The examples in the wikipedia article are things like store greeters, lobbyists, academic administrators, and managers. I think it would be pretty hard to characterize those jobs, depending on the specifics, as being like digging a hole and filling it back in!
Gillitrut
Reading from the golden book under bright red stars
2d ago
Perhaps I am too economics pilled but at a sufficiently high level your outcomes (2) and (3) seem like the same thing to me. Or, to the extent they aren't, (3) seems like it contains a contradiction. On the one hand there are still going to be unmet human wants and desires. On the other hand I am supposed to believe there is no scalable use human labor could be put to in order to satisfy those desires. I am skeptical that both these facts can obtain.
I guess I'm also skeptical of the concept of "bullshit jobs" more generally. I have not read Graeber's book but browsing the wikipedia article for some examples does not give me confidence. For basically all the listed jobs it does not seem difficult to me to describe how the people doing the jobs provide value for the people who are paying them. Maybe "bullshit" is supposed to mean in some broader societal sense but then you are just saying you value things other than what market participants value. That's fine, but you shouldn't expect the market to produce outcomes as if it valued something else!
Gillitrut
Reading from the golden book under bright red stars
4d ago
Important context that OP did not include and relevant for this comment: Cline's product is an AI coding assistant, similar to Microsoft's Github Copilot. My understanding is they don't develop any AI models themselves, their bot is a layer on top of various other AI models. I'm not entirely sure what "head of AI" means at a company like Cline but I doubt it is a trivial role, given the product.
Gillitrut
Reading from the golden book under bright red stars
7d ago
All the way back in 2019 Cory Doctorow wrote a short story about people radicalized into doing violence against health insurance companies due to claim denials. In his story it was a bombing rather than assassination, but still.
Gillitrut
Reading from the golden book under bright red stars
7d ago
I guess it's not clear to me why the memetic change being driven by genetic change is different from the general memetic change. To concretize a bit, I imagine a church founded by, say, Catholics. Consider two evolutions. In one case the descendants of the original founders (for whatever reason) convert to Mormonism and convert the theology of the church the same. In the second case the descendants of the original founders gradually move away or stop attending, but newcomers move in and gradually convert the church toward a Mormon theology. It seems to me that the church is no longer the same church in either case, whether the members are descended from the founders or not.
I also think the degree to which continuity is thenetic varies by institution. Consider, for example, an institution like "The Supreme Court of the United States." Or "The United States Congress." To the extent these institutions are continuous through time I think it is in primarily a thenetic way. In their form or structure.
Gillitrut
Reading from the golden book under bright red stars
8d ago
To my mind there's a conspicuous absence of a fifth church:
- Church Five is a multi-generational congregation, with limited influence from newcomers. Perhaps it's in a remote place where the population doesn't change all that much. However over time the subsequent generations find themselves drawn to other beliefs and practices. So much so that after some number of generations there is little in it that would be recognizable to prior generations.
Is my hypothetical Church Five more like a Church Three or more like a Church Two? Is it identifiably the "same" church? It seems like a theory that permits continuity to be established by either genetics or memes would be constrained to say Church Five is the "same" church in the relevant sense.
Gillitrut
Reading from the golden book under bright red stars
10d ago
Imagine you meet a woman who must be in the 100th percentile for promiscuity (at least in terms of numbers of sexual partners); who's had sex with men who were cheating on their girlfriends with her; who's explicitly encouraged married men to cheat on their wives. Maybe she'll tell you that's it's just a persona she's playing and she's nothing like that in real life (or maybe not). Either way, are you going to take the risk of introducing her to your husband or boyfriend? Maybe you'll counter that you're extremely sex-positive, without so much as a single SWERF bone in your body, and that you'd never get into a relationship with a man unless you trusted him completely – but I would hazard a guess that that does not describe the average woman. And a woman you don't trust to leave alone with your husband or boyfriend (or even your potential husband or boyfriend) is not your friend, no matter how you slice it.
I guess two things that come to mind.
1. I notice the shift in goalposts from "she doesn't have any friends" to "the average woman probably wouldn't be her friend." I'll agree to the latter, but the former doesn't follow from that.
2. This also seems to ignore the existence of both happily single and lesbian women, for whom the potential partner stealing is presumably not an issue.
I agree that she is above average. The point I was making about the average number of films a female performer stars in before leaving the industry is that a lengthy career is not the norm. IAFD has an "active from–to" field listing a performer's period of activity: if one were to scrape this data it should be trivial to find the average duration of a female performer's career. Given what I've read about the industry and what I know about the relationship between a woman's age and her perceived attractiveness (her value on the sexual marketplace), I would be astonished if the average female performer's career lasts for ten years or more. I'll do some digging and see if I can find a definitive answer to this question.
I don't disagree with anything in this paragraph, I just question the accuracy of extrapolating Bonnie Blue's career longevity from the average porn star's career longevity, given the many other ways in which she is not average.
Gillitrut
Reading from the golden book under bright red stars
10d ago
I intended to convey something like (2), appreciate the clarification.
I am highly sceptical that Bonnie Blue has friends of any kind, at least as you and I would understand them.
Why?
It's a well-established finding that a woman's sexual desirability tends to decline over time, which has obvious implications for a sex worker's expected earnings and career longevity. Of course there are women who can keep it up well into their forties, but such people are the exception. This deep dive into the stats of the Internet Adult Film Database found that 47% of female performers leave the industry after filming fewer than three films.
Ok. But I think we have already established Bonnie Blue is hardly average. I am not sure how to compare traditional films to OnlyFans but I'm confident she has done more than the equivalent of three.
Gillitrut
Reading from the golden book under bright red stars
10d ago
I think you overestimate the attention to detail people have. There's a bluesky thread here full of mistakes in AI-generated captions on anime from Crunchyroll. I cannot believe a human fluent in English read all of these along the lines they are supposed to caption and signed off on them. You would think a very simple and fundamental step in review would be "and then a human reads the caption alongside the line it's captioning" but no! Clearly not!
To take another example, in a legal filing by Anthropic their AI messed up a citation to a journal article. The AI got the journal, page number, and link all correct but got the title and authors wrong. The attorney's declaration was very clear that if you went to the link the citation provided it was the correct article (with correct title and authors). You might think a step in verifying an AI generated citation includes "do the cited title and authors match the actual ones" but, again, apparently not!
Given the rank incompetence with which companies deploy AI I am happy to believe HBO just put the un-edited originals in some AI upscaler and hit publish on the result, without any human sitting down and watching it all the way through.
Gillitrut
Reading from the golden book under bright red stars
10d ago
Assuming that 800,000 figure is correct in the first place (there’s probably room for doubt but that is beside the point) I think the simple explanation is that society generally condones or at least tolerates porn “actresses” making large amounts of money because people generally understand that such women are condemning themselves to social damnation with assumptions about their reputations that may very easily turn out to be naïve and thus deserve to be at least financially well-compensated by simps whom society considers to be loser chumps anyway.
This paragraph inverts the justificatory burden, to my mind. If society wants to prohibit some profession they need a good reason for it. Thing are permitted by default, not forbidden. In the United States, at least, it's not like no one ever tried! They were consistently prevented by courts ruling that the first amendment protected the production and distribution of pornography. This in a sense just moves the discussion "up" a level, why not amend the constitution to permit restriction on pornography? But prohibition of pornography has never enjoyed that widespread degree of support.
Warehouse workers and information security officers have a certain level of respectable standing within their social circles. The likes of Bonnie Blue don’t. Women understand that she condemned herself to the equivalent of crack whore Hell.
Do they? I am highly skeptical the people who Bonnie Blue is friends with in real life regard her this way.
It’s very obvious that she’ll never find any sort of respectable job. She’ll never be a secretary, a nurse, a teacher, an HR manager, an accountant etc.
Why would she want any of these jobs? At 800k/month She will make the lifetime salary of many of these professions in a few years. Comparing to the warehouse worker, she made the equivalent of ~30 years doing that work in one month! It's also kind of funny if you read the first paragraph of her wiki page:
Blue was born in 1999[1] in Stapleford, Nottinghamshire. Before beginning her pornographic film career, she worked in finance recruitment for the National Health Service (NHS) and was married. In 2021, her marriage ended and she moved to Australia, although she told Cosmopolitan UK in 2024 that her ex-husband still worked with her "behind-the-scenes".
She had one of those respectable jobs and gave it up!
She’ll very likely stay in the porn business or become a “sex worker” or be unemployed. Maybe she’ll become a porn director and people will pretend like she has talent for it. Either way, everybody knows she’ll age out rapidly.
If you had collectively starred in/produced dozens or hundreds of porn videos that made millions of pounds, wouldn't you be good at it? Why would people have to pretend you were good? As far as longevity Alexis Texas and Angela White have been doing it for over 20 years. I don't know what their earnings look like over that time but it's clearly an industry you can stay in if you have the talent and desire.
Now you might make the argument that she brought it all upon herself and thus should not be getting any sympathy and deserves poverty. But society doesn’t apply such norms to young women because they are seen as possessing innate biological value and also as naïve and easily misled. We’re aware that most young women who get drawn to porning probably don’t fully understand the long-term consequences of their actions, with the explanation being that they were fed modern feminism their entire lives and thus assume that women no longer live in sexual shame and that selling access to your orifices in camera is empowering. We’re also aware that this is a lie but modern feminism benefits well-off middle-class women so we’re not prepared to just jettison it for this reason.
She clearly has a talent that means she doesn't "deserve" poverty. Even before she was getting rich from OnlyFans she seems to have had a fine career. I'm also skeptical she wants or needs my sympathy. I suspect things are going pretty well, from her perspective. There are plenty of things about the current pornography industry I think are bad but few, if any, seem to apply to Bonnie Blue.
Gillitrut
Reading from the golden book under bright red stars
14d ago
Without saying too much I'll say I've been part of an effort by my employer to use LLMs to identify security issues. They do a good job analyzing pieces of code in isolation for particular issues but a limited context window prevents them from finding end to end issues. For example, the LLM might flag that there's no input validation for function XYZ, but that's because the input validation happened much earlier in the scenario. Thinking about this the reverse way, generating exploits, probably means assuming that you've gotten the payload you want in the place where it will be parsed how you want which can often be the hard part.
As an aside Jesus Christ this is ugly code. I am very glad the brief time I spent working with Javascript was with Typescript.
Gillitrut
Reading from the golden book under bright red stars
14d ago
Here is the FBI agent affidavit in support of probable cause for his arrest. The evidence breaks down like:
1. Based on Cole's credit card transaction history he purchased all the parts that were themselves part of the bombs as well as other safety tools one might use to make a bomb across 2019/2020. Sample paragraph (not gonna quote all of them):
Both pipe bombs were manufactured using a 1” x 8” galvanized pipe with markings consistent with a particular manufacturer’s (the “Pipe Manufacturer”) product labeling. COLE purchased a total of six galvanized pipes of this size and shape on or about June 1, June 8, and November 16, 2020. The purchases were made at two different Home Depot location in northern Virginia. According to the Pipe Manufacturer, approximately 26,000 of these items from Pipe Manufacturer were sold in 2020, and over 22,000 of these items were sold to Home Depot.
(continue for the end caps, wiring, steel wool, kitchen timers, etc.)
2. Analysis of cellphone data shows that Cole's phone was connected to towers in the vicinity of where the bombs were placed at the same time surveillance footage shows the bomb planter in the area. Sample paragraph (5 or so of these, covering from 7:39 to 8:24):
At approximately 7:39:27 p.m., the COLE CELLPHONE interacted with a particular sector of Provider tower 59323, which faces southeast (approximately 120˚) from its location at 103 G Street, Southwest in Washington, D.C. (“Sector A”). Also at 7:39:27 p.m., the COLE CELLPHONE interacted with a particular sector of Provider tower 126187, which faces east (approximately 90˚) from its location at 200 Independence Avenue, Southwest in Washington, D.C. (“Sector B”). Video surveillance footage shows that at approximately 7:39:32 p.m., the individual who placed the pipe bombs walked westbound on D Street, Southeast and then turned southbound on South Capitol Street, Southeast. These locations are consistent with the coverage areas of Sector A and B.
3. A license plate reader caught Cole's vehicle in the area shortly before the first security camera footage captures the bomb planter. Cole's cell phone also starts communicating with towers in the area shortly after.
COLE is the registered owner of a 2017 Nissan Sentra with a Virginia license plate. On January 5, 2021, at approximately 7:10 p.m., COLE’s Nissan Sentra was observed driving past a License Plate Reader at the South Capitol Street exit from Interstate 395 South, which is less than one-half mile from the location where the individual who placed the devices was first observed on foot near North Carolina and New Jersey Avenues, Southeast at 7:34 p.m. Approximately 5 minutes later, at 7:39:27 p.m., the COLE CELLPHONE began to interact with Provider towers in the area.
Gillitrut
Reading from the golden book under bright red stars
15d ago
Sounds like your coworker needs to learn about pescetarianism.
I admit I'm struggling a little to understand what the actual policy change is here. Downthread @ToaKraka posted this USCIS Policy Alert. That Policy Alert points to this Presidential Proclamation as the source of the 19 countries. That Presidential Proclamation was issued all the way back in June. So the National Guard shooting was the impetus for USCIS to implement a 6-month-old Presidential Proclamation? I guess I'm also finding it a little hard to follow as, like, a matter of logic. An Afghan national who started working with the US in 2011 in Afghanistan and was brought to the United States in 2021 along with a bunch of other US-allied Afghans then was granted asylum in 2025 and shoots two national guardsmen later that same year, therefore we must restrict immigration from Burma. Huh?
Gillitrut
Reading from the golden book under bright red stars
16d ago
What was the legal justification for killing them?
Gillitrut
Reading from the golden book under bright red stars
16d ago
If it is not a "war", if the people the admin is blowing up are not lawful combatants nor achieving military objectives, then it is murder instead.
Gillitrut
Reading from the golden book under bright red stars
17d ago
I mean, the Post's reporting is that the order was to kill everybody. That doesn't sound like the killing of the two initial survivors was incidental. That may turn out to be wrong, of course, but if it's accurate I am pretty confident saying it's a war crime.
Gillitrut
Reading from the golden book under bright red stars
17d ago
Someone online pointed out that 18.3.2.1 of the Department of Defense Law of War Manual reads:
The requirement to refuse to comply with orders to commit law of war violations applies to orders to perform conduct that is clearly illegal or orders that the subordinate knows, in fact, are illegal. For example, orders to fire upon the shipwrecked would be clearly illegal. Similarly, orders to kill defenseless persons who have submitted to and are under effective physical control would also be clearly illegal. On the other hand, the duty not to comply with orders that are clearly illegal would be limited in its application when the subordinate is not competent to evaluate whether the rule has been violated.
That second strike, if it happened, is literally in the manual as an example of an illegal order that would be a violation of the laws of war. I am as-yet unclear on how involved Trump or Hegseth were in this operation but it sounds like, minimally, everyone in the chain of command between Admiral Frank Bradley and whomever actually executed the strike is, at least, a war criminal.
Gillitrut
Reading from the golden book under bright red stars
17d ago
I've worked with probably a dozen or so Indian coworkers over the years and this does not describe any interaction I've ever had with them. I am also deeply skeptical that a kiwi farms post from a thread entitled "The India Menace - Street shitting, unsanitary practices, scams, Hindu extremism & other things" which cites no evidence is going to contain accurate generalizations about Indians.
It's a little interesting to contrast this with my perception of Chen's attitude. He clearly was dedicated to making sure software that used to work would continue to work for users. It is basically never the software users fault that the program they bought did things wrong. On the other hand, he has palpable contempt for the developers of user-mode software that took a dependency on some undefined or non-contractual behavior and created these compatibility constraints. Ex: Application compatibility layers are there for the customer, not for the program
More options
Context Copy link