But then why is it still free on Windows? Remove the basic tier entirely, then.

Especially if you’re correct that is a server/client system, then someone will just soon up a Docker container that has all of Windows 11 LTE and the Windows version of this pre-configured, so people can continue to use it in Linux anyway, or switch to an open-source alternative. They’re already supporting and releasing Linux builds, so this doesn’t seem to gain them anything and, instead, may cost them marketshare and goodwill.

I’m confused.

And so the Tragedy of the Commons plays out, yet again.

There’s no cost to being a selfish asshole, so it’s sadly not surprising that many individual actors are destroying the public Internet. Like, how can we align incentives to stop this? Regulations/laws are mostly pointless since the very same tactics used to dodge bot detection also make it incredibly hard to identify the originator.

The only other disincentive with a real cost, that I can think of, would be to poison the data fed to scrapers, so they get bad data? That seems expensive to set up, though.

I think TFA has the best solution idea: make it easy to scrape all the useful data using a low-cost standardized system. Then there’s no incentive to scrape the website using a stupid, expensive crawler in the first place.

Edit: actually, LLMs make poisoning the data fairly reasonable… When there’s a high volume of requests for outdated pages/edit pages/other rarely accessed pages, have the server serve a pre-cached parody version of the root page instead. Pre-build one parody copy of each page with a standardized prompt, like “rewrite this page like it comes from an academic journal of medicine or economics with APA citations for every fact.”

I’ve been sitting this from the rooftops, but nobody seems to be listening. LLMs do not create enough value to justify their cost, and their costs rise exponentially for small, incremental gains. It’s a money pit.

Worse, it’s a massive sunk cost masquerading as investment. Inflated equity validations are propped up by an illusion. I know that timing a crash is impossible, but I literally don’t understand how anyone paying attention doesn’t see what’s coming.

It’s going to be bad. 80% such market declines aren’t atypical, historically.

And this isn’t even touching on the Republican dismantling of the American government apparatus and spending billions in a war to disrupt global supply chains of critical resources.

It’s going to be really bad.

Right, like what’s the ROI of a million computers switching from US-based for-profit Windows to Linux?

Funding open tools/tech/research is exactly the sort of thing that governments are best equipped to do, relative to private sector/individuals. Millions of dollars is a rounding error relative to government licensing costs to proprietary software alone, even ignoring the downstream benefits for the rest of the economy. And if we had lots of money in open software, it would attract a lot more talent to make it even better for the rest of us.

The headlines focusing on AI layoffs are still mostly missing this, too.

The real reason for layoffs in tech are from a quiet recession that’s hidden by circular AI cashflows, correcting for inflated staff count from COVID overenthusiasm, or to free cashflow for throwing more money into the AI pit. (Or all three, of course).

AI doesn’t reduce headcount; it just slightly increases the efficiency of skilled workers to do simple, mundane tasks—and even that requires careful oversight or it’ll inevitably fuck it up.

But skilled workers are quick at mundane tasks… it’s just boring. So it’s not really generating much value, even in the best case.

And the sloppening is showing us all the downsides.

The house of cards is going to come tumbling down soon, and it’s going to be a lot worse than the 1999 dot com crash. I heard that OpenAI needs to pay big bills for datacenter expansion in the fall with money they don’t have—I don’t have the source for it, but, if true, that could be the domino that triggers the crash. Circular cashflows between highly leveraged companies makes the entire chain as fragile as the weakest link.

An 80% market correction wouldn’t be outside of historical norms. Hold onto your hat on the way down!

When it was new, I got a lot of “Certainly! …”

Like, come on, mate. You didn’t even read the first sentence before submitting.

Except that it’ll never work out that way. Open models are almost as good and cost a tiny fraction of the cost of the proprietary models. There are no moats to protect their business model. Anyone can come along and eat their lunch.

AI has no path to profitability since it’s going to be commoditized. There isn’t a big enough difference between individual models to justify the price premium of paying $100/million tokens when open models cost 10¢/million tokens.

And it gets even worse when you consider specialized models; the real future is likely going to be custom training models for specific use cases, trained on the company’s data (and other data too, of course). A much smaller model can be much more successful on tasks it has been trained on. It’ll cost a tiny fraction of the compute of a mega model to train and likely beat mega models on tasks within its training domain. And it can run entirely on the company intranet, so there are no real privacy/security concerns.

Right now, the big players are giving away their compute at cents on the dollar, so there’s not much incentive to run local models. As soon as they start to push pricing to try to become profitable, companies will switch to in-house models.

OpenAI is doomed. I doubt they’ll be relevant in a decade.

The Mozilla numbers are wildly inflated. They were effectively just advertising for Anthropic’s mediocre LLM with that blog post.

Exactly my thoughts, too. QR codes are a great tool, but also an incredibly valuable and opaque vector for scams.

The was one recently where they put scam QR stickers over parking payment signs, so users gave their credit card details to scammers. How are you supposed to catch that, as the end user? It’s not like you know the URL you’re supposed to be going to.

Normalizing scanning QR codes just to access a website is going to be abused by scammers in no time.

It won’t affect Vivaldi. Vivaldi doesn’t automatically ship Chrome code without review.

True that there’s no way to enforce watermarks, but it could be required to mandate that paid AI models include an invisible binary watermark in the last digit of the, say, red colour channel, so odd bits form something like a repeating QR code relative to the even bits; require the fans in keyframes in AI produced video, too.

It wouldn’t work for text, obv, and this would be trivially easy to strip in post-processing, but it’s technically and legally possible, with low cost. Of course it wouldn’t affect local models, but not many people are running good image/video genAI locally anyway.

VB is still great. I use it to fully automate a lot of Excel-based tasks, like looping through thousands of Excel files to pull values or update spreadsheets with new features.

Sure, there are more powerful tools, but VBA is built in and works really well for some tasks like that.

I’m vibe coding a fairly complicated bash script to fully automate upgrading a web server at work. For context, I have over 2 decades experience in programming/data analytics/tech, but I’m a Linux and server admin newbie.

It’s comically bad at it. Like, I had to tell it not to post passwords to the production database to console and plaintext log files. Then, about a dozen prompts later, it does it again. The restore script rm-ed things (as sudo) before checking that it had a valid backup file to replace it with. It keeps deleting the comments in the code snippets I send it to update/fix, even when explicitly told to keep the comments. I asked it to prepend time to live commands (i.e. not “dry run” echos), and then it deleted them all again when I asked it to refactor something unrelated.

It’s been great learning for me, and I’m definitely getting this job done faster and to a higher quality than I could on my own, but holy hell these scripts would have been a disaster if someone just ran them “as is”. I’ve needed to fix dozens of errors that could have really screwed things up.

I wonder how often people go through their vibe coded outputs with the careful attention and care it needs. I’m guessing infrequently. LLMs are just word prediction machines; they don’t understand anything.

Jesus Christ, what the hell did I just read? I didn’t clue in to RT until I saw the comments, but that leaves me even more confused—it didn’t even try to pretend to be a news article. It’s blatant propaganda. I thought RT was supposed to be, well, competent at producing propaganda.

The average cost per square foot of a new condo was $1,189 in the first quarter compared with an average resale price of $859, a 38 per cent difference.

Seems like a weird way to frame that number. Sure, 1189/859 is 1.38, but that’s the percentage resale prices would need to rise to match new unit build costs. The way it’s framed sounds like they mean 859/1189 (“prices of resales are lower than the costs of new builds by x%”), which is .72, a 28% difference.

Math nerd alert.

Seriously, though, I would expect better from a financial news publication. AI slop, I’m guessing. (Edit: not the whole article, mind, I just mean that number framing and/or calculation looks, to me, like a typical AI error. The rest of the article is well written, imho.)

Fully automated targeted scams are new. That wasn’t possible before. The cheap scalability is the scary part.

When I received the first email from my “headhunter”, I was drawn in by how professional and customised it seemed. The writing was of a good standard and the sender was clearly familiar with my profile. It felt personal. Even five years ago, says Rosser, you could often spot a scam just by looking at the grammar. “But they’re so clever now.”

“The growing accessibility of AI means that criminals have way more leverage than they ever did before,” Webb says. “They can produce these scams much faster. They can make them more relevant, and there’s a much higher level of sophistication.”

This was the most interesting part, to me.

In the past, scammers deliberately made their pitches obvious, so only “suckers” would fall for them. With AI, it’s now scalable to make the whole thing targeted enough to be believable.

And that’s truly scary.

The funny thing is that a lot of Conservatives think that the profit motive + free market efficiencies will reduce costs, and that public systems are full of lazy incompetents who are wasting taxpayer dollars.

And… They’re not entirely wrong, either. And I say this as an educated NDP voter. They’re mostly wrong, because of many factors that mostly boil down to 1%ers building laws and systems to further concentrate wealth. And that wealth they’re hoarding is from profit motive leading to lower costs and the free market increasing efficiency… And also from profit motive leading to higher prices and that most of the reduced costs aren’t efficiencies, they’re cutting corners to deliver inferior services.

But their world view:

• Makes sense
• Is simple
• It’s internally consistent and coherent
• It’s supported by lots of research (mostly in theoretical economics, mind you…)
• Blames “others”/outsiders for their problems
• Etc. etc. etc.

There’s a reason uneducated, young, white, male voters are flocking to the Cons. The Left’s answers are, by comparison, complicated and messy, with lots of grey area, and involve significant change to systems that have perpetuated white make privilege. So, even though uneducated white males would be far better off in a progressive world, it’s much harder to convince them.

I skimmed most of this thread and didn’t see anyone mention that Steam actually supports third party stores. They let developers sell game keys on other storefronts for free (with limits, granted—the number of keys they can generate depends on sales on Steam, I think.)

Fanatical and Humble only exist because Steam handles all of the games delivery infrastructure for them. That’s, like, the opposite of monopolistic behaviour. Name another tech monopoly giving their services away for free so other directly competing businesses can profit.

I keep forgetting how useless Epic is.

Every once in a while, I want to scan my Epic library to see what’s there… and it doesn’t even seem to have a library feature? I need to use a separate app just to see all my games on their storefront.

Then, occasionally, I’ll want to check out what people are saying about their free game offers… and they don’t have reviews?

They don’t support Linux.

Their Android app keeps redirecting to their website for basic functionality.

Do they even have a method for devs to show patch notes or game updates? I haven’t seen any.

I mean, great that they’re giving developers a bigger cut, I guess, but 88% of nothing is worse than 70% of actual sales. Why would I, as a customer, ever try to shop there? It’s a terrible UX missing many features I have grown to expect.

So, yeah. The author of this article gets it.

Edit: and why bother listing your game there, either? Another storefront to manage is a decent amount of overhead work, I’d expect. You’d need pretty good sales for the effort to pay off.