HNHacker News
TopNewBestAskShowJobs

unclebucknasty

4,691 karma · joined February 1, 2013

submissionscomments
unclebucknasty··on New Bedford police officer accused of using Flock cameras to track ex-partner
>maybe exposes hypocrisy

I don't know that hypocrisy is the word, but it does seem that you're going a long way to carve out an exception, where there is no real difference.

That is, it appears that you've simply decided one scenario is okay, even though that scenario itself begs the question. The article on which you're commenting demonstrates one reason that's so—the potential for abuse.

So, seems it all boils down to "out of sight, out of mind" in your framing. As long as you don't see the people invading your privacy, you're okay with believing that no one is. You don't mind that all of the tools of surveillance are in place, because you've defined a scenario wherein it's not possible that they're being used to actually surveil you.

In sum, your thought experiment is essentially, "would this be OK in a world that doesn't exist". But, you're answering "yes" in the world that does.

Maybe that's what you're sensing as "hypocrisy".

unclebucknasty··on New Bedford police officer accused of using Flock cameras to track ex-partner
>I would certainly feel different if followed by an actual human

Why would this bother you? Because of the prospect of physical harm, or the idea that they are watching you/invading your privacy?

unclebucknasty··on Ask HN: Is this normal HN ranking?
>upvotes-to-comments ratio

Makes far more sense than comment count alone.

unclebucknasty··on Ask HN: Is this normal HN ranking?
Seems unlikely. Using comment count as an indicator of poor quality discussion would treat topics with intense interest and engaging discussion like flame wars.

So, I find this explanation hard to believe, especially given that the mods here are active, and that there are other mechanisms for exposing flame wars (including flagging, down votes, etc).

unclebucknasty··on How I use LLMs to learn complex topics
1000% astroturfing—and across social media. Too many posts of the same quality at the same time(s). These things ran in cycles, spinning up and dissipating just as suddenly.

E.g. the entire framing to combat complaints about shortcomings was, "It's not the tech. It's you. You're just not doing it right. Wrong setup, wrong workflow, add this to your .MD, use loops, etc". Every complaint was immediately met with this same treatment by a swarm of vague bro-bots that materialized from the ether. The core message? Always the human's fault.

And, don't get me started on the waves of newly minted expert AI creators, making recommendations without showing a single example of what they'd supposedly built.

I'm sure some bandwagon organic creators tried to cash in on the genre, but encouraging that was also part of the point.

unclebucknasty··on Georgia police officers fired after Flock camera misuse
>The question I hear is "why shouldn't we be pulling over stolen cars?". It's a much harder one to answer than the straw man questions we set up for these things.

The "why shouldn't we be pulling over stolen cars?" question is just a more specific variant of GP's "nothing to hide" question. i.e., the only people asking it are those who aren't stealing cars, thus have "nothing to hide".

The answer is the same.

unclebucknasty··on How I use LLMs to learn complex topics
Sounds about right.

Tangentially, but related: I'm old enough to remember when the spirit of your comment was pervasive on HN.

unclebucknasty··on How I use LLMs to learn complex topics
>LLM prose is annoyingly dense, and the useful information gets lost in a bunch of noise.

This problem doesn't get talked about enough and is second only to the hallucination problem IMO.

AI produces so much noise to wade through in order to find signal, and the more expertise you have in a field the more that costs. That noise directly subtracts signifcantly from productivity gains.

And, I think the problem is directly related to the hallucination problem. It feels very much like an effort to kitchen sink the response in order to provide some value among possible hallucinations.

It also seems to be a byproduct of Gen AI operation. It just fundamentally doesn't understand what it's outputting, so doesn't know how to narrow down to the most salient bits.

unclebucknasty··on How I use LLMs to learn complex topics
I've (elsewhere) written about this diminishing return effect on LLM utility in relation to increasing expertise.

The question: what's the net positive gain of turning people who know nothing in a given field into sub-novices, while weighing actual experts down with work slop and marginal returns?

And I wonder what the true cost is of arming so many novices with that level of dangerous knowledge.

unclebucknasty··on Humans missed 1 in 3 threats approving AI agent commands across 40k game runs
Interesting premise, but there's not much real world meaning here without stats on the percentage of agent-offered commands that are actually dangerous.

If that number is something like 10%, then we have a really big problem. But, if it's .000001%, then it's pretty vanishing. At some point in between we cross a threshold that puts the risk below many other risks that we routinely take (e.g. trusting npm dependency graphs).

Of course, if it's really that low a percentage, then the entire model of "supervising" via human approval really is fundamentally flawed.

unclebucknasty··on Discovery Loop
This reads as hype of the kind: "well, we can't reliably do these rather mundane things with AI, but we're going to run it extra hard and extra long in some novel way, and it will do amazing things".

The issues are with verification and with detecting drift from the goal. These are related, if not roughly the same issue. And, if they can solve this, then they will have essentially fixed AI. Maybe even AGI.

But, if this were the goal, then it seems more reasonable to solve the relatively more mundane verifiable challenges (e.g. generating solid, reliable code). Then, working up from there.

And, that's exactly what gives this the hype smell. No use for solving problems that don't get the oohs and aahs. Just straight to NAE Grand Challenge problems.

unclebucknasty··on Kill The Cookie Banner
>This is a jaw-drop moment for me every single time I observe someone else using the web and quickly clicking "accept"

There's a mismatch between the velocity at which people visit sites and the time it takes to navigate the cookie particulars of each site.

And, we can dismiss this as people being uninformed or lazy but the reality it is that's actually not so unreasonable. Cookies are in some ways near the bottom of the list where privacy is concerned, given everything else from breaches to search dossiers to device finger-printing to mobile device location-tracking to the ubiquity of cameras in the real world, and on and on.

The idea that we're clawing back privacy in any meaningful sense by blocking a few cookies here and there is kind of quaint.

unclebucknasty··on Claude Opus 5
I'm sure the marketeers would love for the public's assessment of complex versus easy to conveniently shift per their release cycles; or for the public to simply forget their prior marketing.
unclebucknasty··on Claude Opus 5
Recent releases have said something to the effect (paraphrasing here):

"Use <less expensive or older model> for everyday tasks and <other non-critical stuff>. Use <more expensive or recent model> for complex coding tasks, refactoring large code bases, etc.".

Then, the next model/release emerges and the previous "best for complex" gets demoted to "everyday".

Obviously, it's all relative. But, it does beg the question: was the previous model really good for complex coding tasks or no? I mean, how is it now suddenly only good for the "easy" stuff?

unclebucknasty··on There is a shadow hanging over this Fable thing
>it's the real lesson that should be learned here

I believe the real lesson is that we need to fix government. Too many things that people assumed to be codified were actually only ever enforced by social contract.

Until now, we've largely operated within a band of norms that served us fairly well, if imperfectly.

However, we're now seeing what's possible when the social contract is shattered. We need to codify in ways that insulate government from wide variances in the reasonable operation of our form of government. And, we need to root out regulatory capture while we're at it.

Government involvement should be the people's voice. We need to restore that in earnest versus eliminate government involvement; else we're merely a corporatocracy.

unclebucknasty··on WH proposes rules giving political appointees final approval on research grants
>That's at best a misunderstanding...bad-faith...

No. Analyze the thread more carefully, particularly the original comment to which I replied. Should help any good faith reader to see that it's the opposite.

>That is 100% correct. Congress controls spending

You'll see that I actually introduced that fact originally to clearly delineate the roles, whereas GP was blurring / reassigning them to make his point. I added that Congress's other major role here is in oversight, which corrects the GP's assertion that political appointees are needed for accountability to the people. i.e. I'm saying that mechanism exists, Consitutionally. That destroys his primary argument—that this is about accountability.

You seemed to have overlooked that fact (in addition to my other points), in much the same style as GP. Perhaps his rhetoric has worked on you a bit here.

>Congress delegated the details of that role in this case to the president, and the president wants political appointees making these decisions, not scientists and subject matter experts.

That is not what's happening here, and reads like a complete misunderstanding or calculated twisting. The "in this case" bit is actively misleading. The OMB already executes spend management. There is no special "case" here. The regime is using the OMB to politicize the process by claiming it was partisan—i.e. using the same well-worn tactic in its ongoing attack on science and other matters.

>I'm not sure that's actually true in general

Of course it's true. Statistically.

>that trust is not absolute

Never the assertion. Immaterial.

>and things have happened (like initial COVID response

In fact, the left experienced a temporary bounce in scientific confidence during the initial COVID response, before settling down to pre-pandemic baselines. Meanwhile, the right experienced a roughly 20 point drop in confidence that has persisted.

unclebucknasty··on WH proposes rules giving political appointees final approval on research grants
You've restated your flawed assertions, you continue to reassign the roles, and you're conflating Congress with political appointees.

>The problem scientific institutionalists face is that they've squandered a lot of public trust over the decades

The left generally trusts science and the scientific community, while the right has fallen prey to the right-wing war on science and truth. This war was explicitly designed to enable exactly what is happening here—the transfer of more power to the right, rationalized by a seeded distrust of institutions.

Hence, it's not surprising that the people who want political appointees in charge of science are on the right.

unclebucknasty··on WH proposes rules giving political appointees final approval on research grants
The statement, "everything is just people" begs the question. That question is about appropriate roles.

No one is debating that Congress has the power of the purse. That is one of their primary roles. They appropriate, but obviously cannot and should not make every detailed decision, particularly where expertise is required and political neutrality is preferred. Accountability is another primary Congressional role. That comes through oversight, not day-to-day decision making on behalf of those being overseen.

Even if it were desirable to have politicians making decisions in place of scientists, granting that decision-making power to political appointees instead of Congress actually undermines the public's representation and further shifts the balance of power to the Executive.

unclebucknasty··on WH proposes rules giving political appointees final approval on research grants
Many of us did vote for sane ideas, like allowing scientists to make decisions about science. For instance, we knew RFK Jr would be a disaster and here we are, dealing with a resurgence of preventable diseases.

In fact, "unelected bureaucrats" have been the key to whatever degree of success this democracy has enjoyed. Politicizing everything replaces non-partisan expertise with political loyalty and favoritism. It's a direct path to the destruction of critical institutions, undermining the public trust, and authoritarianism.

unclebucknasty··on Fidelity Won't Let Fund Holders Donate to Southern Poverty Law Center
>I guess Fidelity doesn't want to help fund hate groups

Your "guess" is not the stated reason. FTA:

>“Consistent with our grant-making standards and practices, the organization is not an eligible grant recipient during the ongoing investigation.”

In fact, WRT Fidelity's actual disposition on funding hate groups, the SPLC reported in 2023 that their donor advised fund had been consistently used to that effect, including anti-LGBTQ, anti-government, anti-Muslim, and hard right groups.[0]

[0]https://www.splcenter.org/resources/hatewatch/extremist-cryp...

unclebucknasty··on EFF is leaving X
>They still get more engagement on X than on Bluesky.

Is this the right metric? Or would having 98% of their impressions lopped off by the platform factor in? What if they were 100% suppressed? Would it still be "political" for them to leave? If not, then what's the threshhold?

And, if the platform is suppressing them, then isn't it the platform that's playing politics? How are they absolved, and why should EFF stick around to give them its imprimatur of legitimacy / neutrality?

unclebucknasty··on Why the US Navy won't blast the Iranians and 'open' Strait of Hormuz
Arguably is an understatement.

Perhaps you're considering only the European theater, but even that would have been significantly more challenging for Russia without the U.S. tying up (and degrading) Axis resources and manpower throughout Europe and elsewhere (e.g. the Pacific). Japan could have very well opened an eastern front for Russia.

And, it was the U.S. that forced a two front war that prevented Germany's fuller focus on Russia's western front (millions fewer troops). Not to mention U.S. logistical and material support to the Soviet Union, which may well have prevented their industrial collapse.

Even with all of this support, the fatality rates for Russia were astronomical. To this day, it boggles my mind that one nation lost ~26 million people in a single war.

Hard to imagine how they would have succeeded without the U.S.

unclebucknasty··on Show HN: I AI-coded a tower defense game and documented the whole process
All of this, and highlighting this part:

>You would be faster by the fact that you know that it's correct and you don't have to review it. Helps greatly with mental load.

I keep thinking maybe it's me who's just not getting the vibe coding hype. Or maybe my writing vs reading code efficiency is skewed towards writing more than most people's. Because the idea of validating and fixing code vs just writing it doesn't feel efficient or quality-oriented.

Then, there's the idea that it will suddenly break code that previously worked.

Overall, I keep hearing people advocating for providing the AI more details, new approaches/processes/etc. to try to get the right output. It makes me wonder if things might be coming full circle. I mean, there has to be some point where it's better to just write the code and be done with it.

unclebucknasty··on Show HN: I AI-coded a tower defense game and documented the whole process
>a strong and thorough idea of what you want, broken up into hundreds of smaller problems, with specific architectural steers on the really critical pieces.

Serious question: at what point is it easier to just write the code?

unclebucknasty··on 'The Worst Internet-Research Ethics Violation I Have Ever Seen'
https://archive.is/Oq6Tw
unclebucknasty··on Ask HN: Has anyone else noticed a recent sentiment shift on HN?
>There was a time when Musk and his fans were riding a wave of enthusiasm, but with Tesla and Starship problems they've dialed it back.

That's the thing though: those stans were (and still are on some other platforms) very vocal, irrespective of his copious ups and downs (e.g. prior Tesla problems, Twitter debacle, provocative/bad behavior, etc).

In fact, it seemed the worse things were the more vocal they became (like a PR army), and that included here on HN until very recently, IMO.

unclebucknasty··on Ask HN: Has anyone else noticed a recent sentiment shift on HN?
>It's not a left-right thing. Even apolitical stuff gets downvoted. Comments about health, obesity, diet tend to get a lot of downvotes...

Well, everything has become political though, right? I mean, you mentioned health, but that's actually a hyper-political topic.

unclebucknasty··on Ask HN: Has anyone else noticed a recent sentiment shift on HN?
>HN goes through its phases

I've definitely observed this over the years, but there has also been a fair bit of consistency in the form of certain through-lines and more balance.

In any case, what I'm describing is a pretty sudden shift—in particular the recent "re-normalization".

unclebucknasty··on Demolishing the Fry's Electronics in Burbank
>for one reason or another.

Exactly that. It was never that it straight-up didn't work. It was just that there was some issue.

I first encountered it with a TV that literally had a dead pixel. From there, the next 3-4 purchases featured something wrong. Monitor's built-in settings menu didn't display, cordless phone speaker issue, etc.

Dead pixels.

No way it was random. Funniest part was they'd get snippy with their return policy, like you were the problem.

unclebucknasty··on Demolishing the Fry's Electronics in Burbank
And each one of those items would somehow have a dead pixel.
← PreviousPage 2 of 34Next →