You left out the category of new bugs created by the patches and "upgrades", especially for new features.

Which somehow made me think of an even worse race condition, though I'm not sure how to describe the level of abstraction here... If the AI search for bugs goes to a certain depth, and assuming there are no bugs in the search algorithms, then what happens when the AI gets a bit smarter and searches a bit more deeply?

Long time ago when I was first studying computer security, but I remember two fundamentals... First it is really hard to secure the valuable targets, because the attackers may be motivated to apply large amounts of resources if the target has sufficient value. At this point I'd try to refine the meaning of "value" in relative terms, but that doesn't simplify things. Meaning itself is context dependent and in one context data may have low value whereas the attacker has imagined a different context where it becomes much more valuable...

The second fundamental is the chain thing. The attacker only needs to find a weak link anywhere in the chain. The defense has to be perfect, and it's really hard to get perfect.

If an AI didn't tell you that, then I think your memory is playing tricks on you, because I was pretty sure that it was much more than a year ago when I first saw the domain squatting reply when I tried to visit slashdot.jp. That was my initial reaction, but...

So I decided to look into the topic with formerly reliable websearch. As regards the google, I don't think you are really avoiding AI by not asking for the explicit escalation. I'm convinced that all google search is now AI-contaminated to some degree.

According to the google and a second opinion from Bing (which is now hiding the Copilot side of its search?) it's my memory that is playing the tricks. They both agreed on 2025 for the DoD. Lots of details that, as usual, feel too realistic to be hallucinatory. Lots of stuff about transitions and management changes. No, I can't recall ever hearing about the Srad thing, though that was apparently the primary URL for some years. Additional data is that Firefox claims I last used the 2013 slashdot.jp password in 2023... So who or what is lying here? I'm mostly counting it as evidence that my memory is getting tricky and that the 2023 event was a successful login...

However this could explain the compromise of some of the older identities on Slashdot itself. The websearch results claim that there was a period when slashdot.jp was being used as a phishing scam and I bet that too many Slashdotters used the same password on the "closely related" websites. The alternative explanation for the Slashdot thing is increasingly senility and even dementia among oldtimers... (Which reminds me of recent bleak thoughts about possible epitaphs for me. Perhaps "Game over, no points" or "Extra time, no score"? The "extra" is for already living longer than I expected to...)

I bet an AI told you that. I'm pert' shure it was shut down well before 2025. However that seems to be another special class of AI hallucinations. I recently asked several of them about when someone died, and I'm pretty sure the "most definitive" answer was off by about ten years, but it had tried to disguise the error by fudging the date of birth, too.

You're appealing to Dawkins, but I think he would disagree with you. It's the teleology again.

However I think that there is confusion because a lot of things we now tag with the label "evolution", such as "social evolution" or "technological" evolution are absolutely Larmarkian. One of my slow books right now is called What is thought? where he goes into a lot of detail about how DNA encodes executable modules as tools for handling various aspects of reality... However I do think he's doing too much handwaving, but he's coming from a computer science perspective whereas Dawkins is building on a deeper understanding of evolution.

There's also <> but I much prefer the simple symbol. But much worse than that is the lack of support for Japanese. So let me confirm that the Japanese angle for not equal also fails thusly:

Yep, nothing there. (Or in the Japanese bit down here...) There was actually a Japanese version of Slashdot for some years. Basically the same website, but with full Japanese capability.

Thanks for clarifying. I'm sorry that I never got a chance to see the entire film, though I've seen many snippets and references over the years. I'll add yours to my watch later...

There are several senses in which free beer should not be confused with freedom. The number kept changing as I thought about different angles, but I finally settled on four as "good enough".

But Slashdot in 2026 has no way to display the not equals sign. I've tried every trick I could think of. Time to ask the AI?

No, AC. No one wants you. There is such a thing as a sock puppet too stupid.

For this story "We" is Microsoft, but really it's a broader thing, and it goes back to the math. So many people hate the criminals and would even be glad to help shut down the criminals, but Microsoft doesn't want any help on that scale. That would involve training people to be wiser and more cautious, but Microsoft prefers suckers who will obey the ads and send money. In Microsoft's case, that's for software that is so buggy and horrendous that criminals LOVE it. But Gawd save Microsoft from well educated customers!

And that's why it's a broader story. Not just Microsoft getting rich off off the mass production of dunderheads. It's also how they sell scented soap, preposterous cars, and even dunderhead politicians. A few unscrupulous and extremely greedy people at the top are raking in the wealth and the LAST thing they want is for everyone else to get smarter. Oh, and oof course that's why they LOVE their vision of AI, where most people will be completely mentally crippled by dependence on AI. (But they are sure they are too smart to waste their own "vast intellects" on trivial questions that merely elicit more AI slop. (Also why they regard "benevolent" government as such a vast threat.))

These days I'm actually inclined to see it as running backwards, but after all that is sort of how evolution works. The "pressure" for new species is largely created when overspecialized species go extinct, thereby making available large "energy pools" that are no longer being exploited "efficiently".

I think I mostly agree with your reply, though I think vestigial features can also be useful when conditions regress. I have read a book or two recently that was going more deeply into the modular part of it.

Very productive story and this is my vote for Funniest. (Never get a mod point to bestow.)

However I'm not sure I'm getting the point of the joke on this best of all possible jokes. If it's the "L", then why not Lame Intelligence or Loser Intelligence or possibly even Lost Intelligence? Or maybe I just can't see Ludicrous anymore?

I think you're making a good point, but Subjectively focusing on the karmic impact on yourself is too meta... You're taking too much credit because so many Slashdot identities have a minor reality problem these years?

So I want to tilt it to go for Funny-not-really.

There is more to education than learning how to love money harder and longer. Education is now learning to love money longer and harder.

I think the funniest part of Libertarianism as a fake philosophy is their lack of understanding of their own worship words. At least I've yet to detect any deep understanding, but maybe it's my own confusion showing after too many decades of trying to understand such things. What it seems to come down to for Libertarians is something like "I should be free to do whatever I feel like doing to get more money and the only legitimate purpose of 'government' is to make sure the victims can't complain about any of my negative externalities."

So to completely muddle the water:

[infinity symbol] << money << time << [infinity symbol]

#1^2 Freedom == (Meaningful + Truthful - Coercion^2) Choice{~5} [not equal symbol] (Beer^4+ | Speech | Trade)

Just a couple of mnemonic games to put tags on some of the complexities, not even mathematical philosophy. No units that I can figure out, let alone the measurement errors. "You're pegging my coercion-tometer, but I still say no." (Or vice versa^3 as "I'm pegging your coercion-tometer, but you still say no?")

Mod parent Funny, though I think it could have been funnier if "bombs" was worked into the joke somewhere. Maybe something like "The YOB should have sent smarter bombs"?

My Subject is the joke I was looking for. Low hanging fruit. Of course. I really wish I had more of a humorous imagination rather than just minor UI improvements to tell Claude to implement...

Anyway I'm sure Peter Thiel and Elon Must lose lots of sleep over killing children. Milliseconds and milliseconds. At least every couple of years. Though it's actually the bad press coverage that is annoying them.

Mod parent funny, but it's an old joke.

On the substance of the story, it's fundamentally a whitelist approach, but of course that will create a black market for fake entries in the whitelist. Strangely enough, that may still be acceptable for the primary justification here. Most kids can't afford black market prices, so they will be blocked out as intended.

I've been thinking of the transition point from the other level. Extremely young children have to be protected from self-destruction, but they don't mean anything by it. No conception of any danger. This changes as the kids get older. But with increasingly understanding of the dangers comes great power or great responsibility or cold revenge or something?

Mod parent up, though you didn't do much to even guess how high the number might go. I've gotten quite a bit of data on related topics from Damasio's books, though his stuff is pretty hard to read. I also liked the way Jeff Hawkins approached the topic. (Too bad Mukherjee is unlikely to write a book on the topic.)

The chemical/hormonal sides of decision making are important. And all that vagus nerve confusion and crosstalk... Evolution is not teleological, but lots of genetic cruft just hangs around because there is no sufficiently strong incentive to remove it. And don't get me started on the perversions caused by sexual selection. (But do you have a funnier example than the male peacock's tail feathers?)

No Funny here, even though I thought the disguised-downsizing story looked promising for Funny... I admit that I can't think of much beyond "Couldn't happen to an unnicer company."

As I looked over the comments the main stuff I noticed was some wannabe lawyer trying to defend fascism. Probably one of those jailhouse wannabe lawyers who got most of his legal training writing up his rejected appeals and writs of habeus corpus before the YOB sprang him.

As regards the larger topic area, I don't see how much it matters. It's really a surplus data problem, if if Flock isn't gathering and abusing the data, then it will be someone else. Larger problem is that the best corporate abusers of data tend to be the most profitable. In this particular application "best" includes more discretion. The Flock brand became too visible and toxic. (Though it's also possible "Flock You" jokes are relevant.)