Posts by tekHedd 668 publicly visible posts • joined 9 Aug 2011
Exactly! We're approaching something. If you redefine "the singularity" from its previous apocalypse-adjacent meaning[*] to mean "the absolute best result as an LLM can ever hope to achieve," then we're basically approaching it, and always will be, because it's an asymptote. We can put exponentially more processing power into an LLM, but the training data is finite.
"AI" of today asymptotically approaches being a perfect search engine that makes stuff up if the answer was never posted to reddit. And it has no idea whether the answer is real or made up, so it can't even warn you.
The "apocalypse" analogy runs deep. The singularity will always be just around the corner.
* singularity: the point in time when machine intelligence reaches to the point where human intelligence is no longer relevant and change happens so fast that the speed of development is effectively infinite, such that the rate of technological progress is nonlinear in the seminconductor sense, similar to reaching the event horizon in a black hole
I tried a demo of ArcGIS on a project (converting ancient map data for my HOA) and it was, to put it nicely, complete garbage. A bag of hackish python scripts with no error detection or reporting. They still send me a beautiful, thick magazine full of pictures every few months. I guess we know where all that sweet government money goes: glossy brochures.
I'm not surprised it has a few holes in it.
Oh no, it's TEN YEARS old! Might as well set fire to it. Bose lost the plot when old man Bose passed on (as he apparently foresaw). Even this article suggests that it's reasonable to expect to stop using *speakers* after 10 years.
Ridiculous. Streaming radio hasn't gone away. Spotify, last I checked, is still in business. The hardware doesn't wear out. The speakers in my front room are a 30 year old pair of Definitives (Craig's list), powered by a 50 year old Carver and a $11 bluetooth receiver. Last I checked, WiFi was still backward compatible with 2013 standards. Firmware is updatable.
Everything these days is built to a 4 year refresh cycle, like American cars from the '90s. It seems like a good idea from a corporate standpoint, but it really kills the brand. Example: UA Project Rock headphones by JBL: replaceable earpads! But you can't replace them because they stopped making the pads at 4 years when they stopped making the phones. (And it's actually worse than other brands because they use a proprietary attachment mechanism instead of a simple stretch-on pad.) "Repairable" is meaningless branding now, like "green" or "washable."
We complained about planned obsolescence in the 90s, but those companies lacked vision.
It's the "some speech *should* be illegal" argument. We've used it here in the good old USA for years to explain why some objectively bad speech should be illegal. Now an objectively horrid party is in power and now they're using that on what *they* consider speech that should be illegal.
So, I mean, I'm not going to disagree with you (who wants to become unemployable); I am just pointing out that this particular sword has two edges.
The title of this article demonstrates that not only do the authors not know how neural nets work and are trained, but neither do the OpenAI researchers looking into "halluncinations."
Of *course* it rewards making stuff up, making stuff up is the *only* thing generative AI does.
Somebody gets paid to do this research? Paid, probably, like 5x what I make? I would take that job but I'm not good at making stuff up.
I'm still amazed that nuclear is considered dirty power while gas turbines fed by "let's loosen those regulations a bit more" fracking is considered green.
Yes we all learned that nuclear waste was horrid in the 60s, but energy companies have ridden that wave for a lifetime. Things have changed.
Wait, so just one of these GPUs pulls as much power as my house? (Well, not *my* house. I have a lot of crap in here, but a normal person's house?)
Sounds about right.
After the burst, we'll be able to rent this excess capacity, but it will still use a shedload of power, which is not getting cheaper. And the next generation of chips will be more efficient. So, just as developers in the suburbs are building the slums of the future, these AI companies are filling warehouses with e-waste and way more power and cooling capacity than is actually needed.
Smart money is considering how to make money with all that power and cooling capacity, once the obsolete GPUs have been carted away.
Maybe they think that invoking the (undoubtedly real) specter of "China" will protect them, but if they're a US government agency, perhaps they should keep their heads down right now. We're still a very long way from funding those tax cuts for the rich, and any functioning government agency is a target.
I know, that's really cynical. These are cynical times here in the good ol'.
I mean, if you accept that widespread adoption of IPv6 and the end of IPv4 is a good thing, this plan is good and I see no problem with it.
We may safely ignore IPv6's detractors, because its simple elegance makes it a no-brainer for... it's simple easy-to-understand ... its. I'm sorry where was I going with this?
Anyway, if you want to drive adoption, just make the use of IPv4 anywhere in the system a hard incompatibility with systemd. This will of course first require moving the rest of the network stack into systemd, which I assume is already on the roadmap. This will both win hearts and minds from the hardcore Linux developer community, while gently moving the backbone of all computing architecture to IPv6.
So, I can't see how people can stand using other tools after trying the JetBrains environments. And I hate their huge focus on pushing out new AI tools, although it does have the advantage that it's distracting their developers from more stupid ease-of-use-removal changes like "auto hiding window controls" that look fancy and make for nice PR shots but make the tools actually harder to use.
But, I digress.
They've earned some credibility, and I feel we should give them the benefit of the doubt, maybe just a little: since their primary business model is to make programming as easy as possible without the fuss, they have no choice (considering that thousands of Vibe code fanbois have deep pockets), no choice that is but to have an AI tool at least as good as possible to make with the current tech. Whether or not it's a dead-end. And, thank goodness, they have made it optional for the end users, unlike certain other companies I could mention. I mean, if you know anybody in sales, marketing, or management, you know about the pressure to put AI in things. Having it as part of the product is *not* optional. If they want to stay in business.
And I do want them to stay in business. The alternative is unthinkable
Obvious troll is ob...wait OP is serious.
The only real problem with Linux is the endless, corporate-backed push to make it into Windows, so they can make it too byzantine to work on without corporate backing, cryptographically lock down the "official" brand name releases, and "fund development" of those by pushing ads and training AI on our... well anyway that's the problem.
"The most trustworthy mechanism to get the good stuff is GitHub."
And it's certainly not problematic in the least that GitHub is now a Microsoft joint as well, and pushes its own AI and other features on you willy nilly. I'm sure there is no conflict of interest there.
And it tries to do everything as one giant monolithic app. Hmm, where have I seen that design philosophy before?
It's classic overengineering: oh someday somebody might want to use a Sharpie to take some notes while editing magic, byzantine config files, so it needs to have integrated support for ordering Sharpies from AWS, and not as a pluggable add-on but built in to the kernel with every install. What if they want colored sharpies? We need to add support for that. What about generic brands? In fact, let's just add general support for online shopping of all items, not just writing tools, but let's make sure there are sharpie-specific options in the first release that are later obsoleted and cause it to quietly fail without logging an explanation why.
But I digress. Does SystemD have a built-in email client yet?
Have you submitted a PR? The community is very supportive.
Yes, I know this is "learn to code" for distros. But honestly, "oh it doesn't persist my settings" sounds like a very fixable problem. At the very least, submitting it as a bug and then helping to reproduce it will almost certainly result in happiness.
I've been using Devuan since very early--the only real problem with the whole project is that they picked a name that autocorrects to Debian.
I don't need a discussion of ancient history to know that a couple years ago we had functioning economies and established borders (regardless of how annoying, arbitrary, or inconvenient they may be for some people's plans), and now we have an army crossing that established border uninvited.
Let's see, you'll need a laptop, a cheap SDR, and specifications that have been publicly available for decades. And maybe a little knowledge of directional antennas, so you don't get caught. Engage the brakes, which dumps the pressure and stops the train immediately...
Could be handy if you want to block traffic... but where's the profit? Nobody robs trains anymore. (retro crime!)
AFAIR, trains are designed so that if there is a separation or other break in the line, it will release the pressure, causing all the brakes to come on.
So, if FRED controls the brakes, it probably supplies pressure to the rear portion of the train (less than half). If there's no pusher engine, and it applies brakes for the entire rear of the train, and the head is still pulling, and there's a curve... this would be Bad[TM].
The companies selling phones and OSs want to pretend that this "AI agent" is working for you, the customer. So, they play down the fact that it is in fact a giant Shodan-like program residing in their private datacenter. They want you to give their thing permission to buy things online, run programs on your desktop, make calls on your behalf...everything! They definitely don't want you thinking about how its training data and prompts might be tainted by nefarious (or good) intent, about what might happen if some hacker removes the ethical constraints from it, how a government or corporation might use it to shape your worldview, and a thousand darker uses that sound hyperbolic but if they occurred to me just now writing this sentence, you know much nastier people have been thinking about them for months.
You think the people who don't understand the technology are afraid? Try talking to people who do.
My money is on: they've hooked their stupid Copilot into every corner of Windows and all of its core support products, with the ability to control them all, and released the integration all at once, breaking everything in ways you can't work around, and which Copilot will screw up every more badly as you try to work (and cannot disable).
"list loads of plausible but fictitious AI products that we won't be able to live without" sounds suspiciously like a decent starting point for an AI prompt to generate a starting point.
One thing chatbots are quite good at is creating plausible but fictitious answers. :D
It'll be vetoed by people with no sense of humor. Although I would assume we will even get pushback on that...
"He didn’t have a sense of humour and, like most people without a sense of humour, prided himself on the sense of humour he hadn’t, in fact, got."
Why would a DOGE employee use a VPN?
1) Perhaps you are getting paid well and don't want to wreck your career by quitting DOGE, but would really prefer not to be directly implicated in what you're pretty sure is going to be an absolute dumpster fire.
2) You're torrenting videos from your work computer and running a VPN because of that
...and you get a great discount on this Russian VPN software through your work contacts...
I mean, Incompetence and Nefarious Intent are only possible explanations.
They're not mutually exclusive.
A third strong possibility: He's not really the one in control. All this means that is whoever is pulling the strings is either stupid or evil, or both.
Most likely possibility: the person(s) pulling the strings is nefarious, but all that experience playing people off against each other on reality TV has taught him to throw up a very impressive smoke screen that keeps enemies off balance. He's also petty and self-centered, and throwing his personal vendettas into the mix, creates even more confusion.
We're being charitable when we assume incompetence. Stop being charitable and face the truth. We're being played.
Let's Encrypt obsoleted their "legacy" automatic renewals recently. I guess if updating certificates is your full time job this is not a PITA.
"Just automate it using a magic script that you have no idea how it works and it might go away at any time because you need to move to the latest hip mechanism for automatic renewal for the children" isn't an answer that fills me with warm cozy happy feelings.
So we're supposed to believe they're just now going to start shipping all queries to the cloud, instead of the much more likely "eventually we will get in trouble for spying on everything you say around the house, so we're now officially going to stop pretending we don't."
The increase in traffic will also mask which other keywords trigger data delivery.
(I wish this felt like paranoia, but it's just how business is done these days.)
"AI continues to deliver plausible, but wrong, answers to questions."
Working as designed. #wontfix
ALL output from Generative AI is hallucination. "AI accuracy" is (human) confirmation bias. (..confirmation bias often makes this mistake, where probabilities are involved, see also gambling.)
^ Anyone who says otherwise is scamming you.
Yup, the bureaucracy is bloated, inefficient, definitely corrupt in places, slow...broken even. And, it's probably the best government you're going to get, in a somewhat outdated, "version 1.0" representative democracy. If we had a working Congress we could have upgraded to version 2 or something by now, but that's clearly not an option. But it had some safeguards, which look to some like "activist judges" and "bureaucratic waste." By the end of the year, we'll be really missing those things.
Thinking you're going to fix it by "burning it all down" and putting it in the hands of a super rich playboy who goodness only knows how much contempt he has for the masses is... it's the plan? That was the Genius's plan. That was his whole plan.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
