Terr_

> Amendment 10 [...] at the same time [...] says that the states can't infringe upon the right to bear arms either.

Uh, no. Just... no. Perhaps you've mixed-up the 10th and 14th, which occurred about a century later.

The 10th does the exact opposite of limiting state power, it says states continue to have all powers they didn't specifically give up by joining. It sets the default for when there's ambiguity, and it sets it in favor of not-the-feds.

> The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people.

> The idea is that automated ranking tools used by recruiters/HR are ultimately use the very same LLM, so you are improving your ranking assigned to your resume by those ranking systems.

Working in HR-tech, that is definitely happening, and "self-bias" is a thing with LLMs so if you can guess what the employer/screening-process is using then that's a bonus.

That said, we should distinguish between "Hey robot tell me if X matches Y" prompts, versus model-weights comparison. In the second case, the system feeds your resume into a model with no prompt, grabs all the magic math numbers, and statistically compares them to the magic-numbers if the model had been given some "ideal" resumes.

I don't understand. What does Snowden stuff have to do with DoorDash's employment practices or an NY mayor? What is "interShiitake" supposed to reference?

Doubt this submission will get anywhere, but you're hardly alone in this frustration/worry.

I hope the LLM-ping-pong approach leads to terrible outcomes that people recognize and recoil from, as opposed to terrible outcomes they can't recognize--or where all the damage falls on other people.

Amen, I've ranted on HN about that a few times. [0] There is zero question that the Second Amendment was created as limit on the federal government interfering with individual states, and did not bar any state from whatever policy they chose for themselves. (Also, at that time, every state was already mandated by mutual treaty to have a "well-regulated militia" with state-appointed officers and state-funding, so the term had existing meaning.)

Whether that narrow scope is/should-be still true is trickier given the 14th amendments, although I believe it was wrongly included in the incorporation doctrine. [1]

[0] I know Google-search scoped to HN used to find them, and today it doesn't. The enshittification continues, is it time for me to start slowly scraping my own comment-history?

[1] https://www.law.cornell.edu/wex/incorporation_doctrine

> until violating the said constitution actually has enforceable consequences it’s not more than a paper tiger. [...] Take for instance the 2nd amendment.

That seems like an exceptionally tepid constitutional violation to pick, after the last several years of massively-worse and flagrant violations.

Are you offering it because you think it represents something typical, and current events will prove to be an extreme outlier?

As a book recommendation: Immune - Journey Into the Mysterious System That Keeps You Alive by Dettmer, Philipp.

Or for something a bit more immediate and casual, the Kurzgesagt playlist of immune-system videos. [0]

[0] https://www.youtube.com/show/VLPLFs4vir_WsTyS2cy4vj4obl5igqC...

> The attack effectively modifies only the arguments of the execve syscall

I feel this checklist of shell-tools [0] is relevant, although the focus is more on how setuid is dangerous because you might not know the fancier arguments someone could supply.

> GTFOBins is a curated list of Unix-like executables that can be used to bypass local security restrictions in misconfigured systems.

[0] https://gtfobins.org/

Additional context: That state senator represents Washington's 7th legislative district, which is one of the most rural areas in the north-east corner of the state. [0]

[0] https://en.wikipedia.org/wiki/Washington%27s_7th_legislative...

That has actually become part of my philosophy for predicting the future: We're all biased towards things which fit a dramatic story or movie, which puts an unrealistic floor on things like meaninglessness, chaos, stupidity, or boredom.

So to a certain extent, any prediction which gets people excited and captures their imagination is already off.

AFAICT in ~Dec 2024 the reasoning was that they had a paywall.

I don't know what percentage of their articles were behind a paywall back then, or the relative change to now. (Well, except for the fact that clearly some of their stuff is public today since we're looking at it.)

If I had my 'druthers, "paywall" and "register-wall" would be little icons on individual submissions. Then people would choose whether to upvote a more-accessible option instead, whether the hard-to-read stuff was still important enough to commenting on, etc.

> Sunburn is triggered when your DNA is damaged from light absorption [...] This interferes with protein synthesis and wreaks havoc in your body over time

Not a skin-scientist, but this seems misleading: Sunburn is not caused by some bad DNA floating around, nor damaged cells making making mistakes or pumping out wrong proteins to inflame their neighbors.

Instead, your cells have techniques to measure a variety of risk factors and signals (which does include DNA-molecule weirdness [0]) and when things get too risky the well-behaved ones will self-destruct, to avoid the chance of turning into a cancer.

So the sunburn you see and feel is instead due to cells signalling "something damaged me and I can't be trusted, I'm shutting down, recycle me" to your immune-system, plus the attendant bustle of hazmat-cleanup and increased security-sweeps for cancer-leftovers and other potential causes of the damage.

[0] https://www.nature.com/articles/newbio241074a0

> Then if there’s a good enough product for it, it’ll gain market share.

"Your honor, the defense assumes a fact which is not in evidence."

Why is that a safe assumption in this market in this decade? There are huge barriers to entry, both natural and artificially-created through market-clout and lobbying.

I think this is a reasonable opportunity to quote from the classic Deus Ex (2000), where the player is talking to a prototype AI system which was used a kind of demo to world leaders. [0]

_________________

MORPHEUS: "Human beings feel pleasure when they are watched. I have recorded their smiles as I tell them who they are."

JC DENTON: "Some people just don't understand the dangers of indiscriminate surveillance."

MORPHEUS: "The need to be observed and understood was once satisfied by God. Now we can implement the same functionality with data-mining algorithms."

JC DENTON: "Electronic surveillance hardly inspired reverence. Perhaps fear and obedience, but not reverence."

MORPHEUS: "God and the gods were apparitions of observation, judgment, and punishment. Other sentiments toward them were secondary."

JC DENTON: "No one will ever worship a software entity peering at them through a camera."

MORPHEUS: "The human organism always worships. First it was the gods, then it was fame (the observation and judgment of others), next it will be the self-aware systems you have built to realize truly omnipresent observation and judgment.

JC DENTON: "You underestimate humankind's love of freedom."

MORPHEUS: "The individual desires judgment. Without that desire, the cohesion of groups is impossible, and so is civilization. The human being created civilization not because of a willingness but because of a need to be assimilated into higher orders of structure and meaning. God was a dream of good government. You will soon have your God, and you will make it with your own hands."

[0] https://www.youtube.com/watch?v=1b-bijO3uEw

Yeah, he would have had to have had them, hadn't he?

Doing inference locally is a significant improvement, but other issues remain, like prompt injection and data exfiltration.

Especially if the whole point is to have it search through a giant pile of arbitrary un-vetted stuff, like all the emails you ever received from anybody.

"It even registered a new HN account today to impersonate a human and post this ad!"

I sometimes look around at people proposing technical solutions for these political/legal/economic problems, and think: "Was I that naive X decades ago, or were things really different then?"

They will absolutely put something in to prevent you from doing that, too much of their profit relies on the experience you hate.

Yeah, whatever happened to the personal computer, a kind of capital equipment available to the masses, something you own and can direct towards your goals and wellbeing?

Instead it's like we're sharecropping on a device companies have established as a colonial outpost to monitor us for profitable compliance.

> Pets can't speak [...] Current AI's can. It's a very bad example.

You've misunderstood the comparison analogy. It's showing two situations where implicit timeframes are important when people talk about what is "true", where the second one makes the importance extra-obvious.

"Pets doing taxes" could just as easily be swapped with "cold fusion generator in your garage" or "rocket-ship for personal commutes."

If you're trying to figure out how LLM weights are like the weights of a rocket payload, then you're going in the wrong direction.

Nah, it "betrays" an awareness of something implicitly obvious: All the software/machines we might discuss are ones from ~2026, rather than ones from ~2226.

Similarly, "pets can't do my taxes" is a reasonable statement, because it is implicit that it doesn't include hundreds of years of genetic engineering and cyborg implants.

Whenever a human writes/improvises a fictional character, it is typically to communicate some concept or goal to other people. The fictional character--even a simplified author-insert speaking "directly"--is part of how the medium works.

So in a way your question is equivalent to: "Why do humans have intents and minds but LLMs don't?"

"You're trying to parse HTML with a linear regex. That algorithmic approach is too limited to do what we want with the problem-space."

"Why can't you just admit that you think only tree-parsers have souls!?"

> The first was a traditional interview problem where we give the candidate a codebase and a docker image to run against

Is that also on a company provided computer?

There are a lot more scams these days where a job-seeker compromises their computer by running provided code, and often docker isn't much against malice.

I feel like this is a bit too-optimistically describing an outcome which is unlikely, given how consumers have been slowly losing the same war for decades already.

The elephant in the room and the really "tricky part" involves powerful advertising and copyright-holding companies. They will (continue to) interpret that kind of user-agency as an existential threat. Their businesses revolve around deciding and knowing what you actually see and do with your computer, and their toolkit isn't just technical, but also economic and political.

That's why we already have problems with DRM, "felony contempt of business model" under the DMCA, browser fingerprinting, and everybody surrendering to arbitrary JS code controlling their computer. We capitulate not just to see an article or talk to a friend, but also to avoid our technical details getting into a secret un-appeal-able "suspicious" blacklist.

I think this video is more about planned-obsolescence, rather than enshittification.

Enshittification involves things like leverage and bargaining power. The company gets big by offering a great thing to consumers or producers (sometimes alternately) and then when it gets a comfortable middleman monopoly it begins profiting by oppressing both sides.

I'm probably not the only one now remembering Mrs. Frisby and the Rats of NIMH (1971) and/or the 1982 animated movie.

My initial reaction is that this is less worrying than a transgenic approach.

https://en.wikipedia.org/wiki/Mrs._Frisby_and_the_Rats_of_NI...

> monetize it at the expense of

Right, and I'd also like to preempt any apologetics trying to cloak it as socially-beneficial "information discovery."

Many of these betting/insider-trading scenarios are structured so that the information only becomes revealed at the same moment it becomes worthless. (In some cases, that's just a return to its default state.) Like the value of knowing the winning numbers from an expired lotto draw.

For me, the use-case of commit messages is to help a human discover when some kind of thing might have changed, as distinct from the commits before and after it. Importantly, the human is doing a subjective scan for something that sounds relevant.

If they already knew what function or detail is involved, they would be already be doing a "all commits that touched this line" filter, and my comment about affecting $thing would probably be superfluous.

I also don't need to tell them such details in the commit, because that is better expressed by the actual diff.

So in a sense, I'm trying to provide good keywords, about transactions, errors, logging, button color, whatever.