Posts by pip25

Ex-Meta exec: Copyright consent obligation = end of AI biz

Re: Mr Clegg is wrong

Look, if you do perceive that there are issues with copyright as it is currently, then we are not that far apart in opinion anyway. I understand that you believe those cons do not outweigh the positives, and you are entitled to that opinion. I don't want to convince you otherwise, because that wasn't my original point: what I was trying to say was that creators continue to create creative works even in today's climate, where the system that's supposed to empower them is not at all working in an ideal way (again, opinions on the extent of that vary). It would thus be an overreaction to claim that the abolishment of copyright (which, as I mentioned, is incredibly unlikely) would somehow lead to the global end of creative output. Creativity would limp along in some way regardless, just like it does today.

Re: Mr Clegg is wrong

Just google for 2 whole minutes. If you're actually interested, that is.

Re: Technically, he is right.

[...] and is reliable (Wikipedia [...]

I like Wikipedia, but it's been a while since I've seen that word and "reliable" in the same sentence. And I think that's not actually surprising, because if you take a sufficiently complex (moral) problem, you will find that that there are in fact no reliable sources to use, in the strictest sense. Every author brings their own prejudices and opinions into the mix. And even if our AI would be intelligent in a more substantial way than LLMs today, I do not see why we could hope it'd do more with those viewpoints than we could.

Re: Mr Clegg is wrong

That's a bunch of overdramatic hogwash. A case can be made that copyright, in its present form, does more harm than good, and actually fails at what it was originally created to do, which is to support creators. Yet creators still create. Should it disappear overnight, other incentives would take its place, perhaps better ones.

But let's not kid ourselves, copyright won't just disappear, even if our AI overlords demand it; it's way too entrenched in current society. But if the whole AI brouhaha results in some well-needed reforms of copyright - in my book in might just have been worth it.

Builder.ai coded itself into a corner – now it's bankrupt

Re: Wasted potential - there was clearly mismanagement but it wasn't all bad

They had 500 000 000 dollars to spend. Five. Hundred. Million. Dollars.

Even if everything you say is true, burning all that money goes a bit beyond "financial mismanagement" in my book.

Nvidia GPU roadmap confirms it: Moore’s Law is dead and buried

This should not be new to anyone in the industry

The limits of silicon were well-known and talked about decades prior. Whether Moore's Law is truly dead depends on whether someone with sufficient amounts of money manages to reinvent the basic building blocks from scratch. (Nvidia would be in prime position considering it's swimming in cash, but that's assuming those profits won't be spent on luxury yachts or whatever.) Also decades ago, there were talks of basing chips on light instead of electric current. I wonder what became of that research.

Microsoft won't let customers opt out of passkey push

Re: "No password entry or 2FA step is required."

Thanks for the explanation, but this does not seem to mesh with how I saw passkeys being used. My only interaction with them so far has been through the Proton Pass extension, which stores my passkeys encrypted in the cloud somewhere. I can retrieve them on any device I install that extension/app on.

"No password entry or 2FA step is required."

I can understand the first part, since passkeys can basically function as very secure passwords. But why no 2FA? If my passkey is stored in my desktop browser, proving that I have access to my phone as well would offer better security, would it not?

Guide for the perplexed – Google is no longer the best search engine

Re: This is the second Reg article today which references wheresyoured.at

My thoughts exactly - thankfully the text is easily extracted. I've made an effort to validate it by skimming parts of the article, and it seems valid enough. Done by Gemini Flash 1.5.

----

This article argues that the tech industry has entered a state of "Rot Economy," where the relentless pursuit of growth at all costs has led to the degradation of user experience across virtually all digital platforms. The author contends that this isn't a deliberate, Machiavellian plan, but the consequence of countless short-sighted decisions prioritizing short-term financial gains over user needs.

Specific examples are given, including Spotify's redesign prioritizing video content over its core music function, Sonos's app update removing accessibility features, and Meta's frequent, disruptive redesigns of Facebook and Instagram. The author highlights the pervasive nature of this issue, citing manipulative design choices, excessive advertising, microtransactions, and constant notifications as contributing factors to a degraded user experience that negatively impacts users' psychological and social well-being.

The author also draws a distinction between their theory of "Rot Economy" and Cory Doctorow's "Enshittification," arguing that the Rot Economy reflects a broader, more fundamental issue of growth-at-all-costs capitalism impacting the entire digital ecosystem, not just specific platform strategies. This relentless pursuit of growth, driven by figures like Jack Welch and Milton Friedman's neoliberal philosophy, leads to the prioritization of metrics over user satisfaction and ethical considerations.

The author uses a detailed anecdote involving purchasing a budget laptop to illustrate how this affects even low-cost devices, which come pre-loaded with a slow, ad-laden operating system and forced into specific ecosystem usage. This highlights how widespread and impactful the problem is across different socioeconomic groups. This negative user experience is compounded by the rise of manipulative algorithms, AI-generated content, and widespread scams and disinformation.

The author concludes by calling for greater awareness and accountability, urging readers to recognize the widespread nature and psychological harms inflicted by this system. They name and criticize several prominent tech CEOs (Sam Altman, Tim Cook, Sundar Pichai, Satya Nadella, and Mark Zuckerberg) for their roles in contributing to and profiting from this system. Finally, they call for a collective effort to resist and challenge the status quo, emphasizing the importance of public acknowledgment and discourse as a catalyst for change.

OpenAI's Sora lets ChatGPT subscribers churn out janky text-generated videos

MVP

So OpenAI's release criteria for Sora, a model for generating moving pictures, did not include, uh... the thing being able to handle movement? How curious.

Imagine a land in which Big Tech can't send you down online rabbit holes or use algorithms to overcharge you

So many things are censored or banned in China, some of those bans might be useful through sheer coincidence.

Windows 95 setup was three programs in a trench coat, Microsoft vet reveals

I think the Windows 3.1 installer was still present in Windows 98 as well. Having used 3.1 for quite a few years, I recall finding the shape of the installer's windows "suspicious" back then.

Hugging Face puts the squeeze on Nvidia's software ambitions

Those prices don't seem too competitive

I can rent a H100 SXM on-demand for $2.99 per hour on RunPod, or $1.75 if I allow my workload to be interrupted. (This is probably not the cheapest provider either, just one that comes with easy-to-setup Docker templates.) That $2.5 to $6.74 offer from DigitalOcean is rather uninspiring in comparison.

Techie took five minutes to fix problem Adobe and Microsoft couldn't solve in two weeks

Phew, dodged a bullet on that one

I had to disable Fast Start ages ago for a very simple reason: it never worked, only made my computer hang on shutdown. Looks like I didn't miss anything (useful).

We're in the brute force phase of AI – once it ends, demand for GPUs will too

Wait, he's making sense

Something is not right here. Does he really work for Gartner...?

Feds urge 3D printing industry to end DIY machine guns

Ease of misuse

Just a naïve thought: if everyone really has to have guns, how about giving out guns that CANNOT be made fully automatic with a single piece of plastic...?

If every PC is going to be an AI PC, they better be as good at all the things trad PCs can do

Emulating x86/64 on ARM is slow

Well... duh.

As much as I agree with AI PCs being a load of nonsense currently, anyone who bought an ARM-powered Windows machine for gaming simply did not do proper research. It's not like this is the first, or even the second time people got burned by Microsoft's ARM offerings.

A quick guide to tool-calling in large language models

Re: OK, but why?

It can be useful exactly because the LLM output is error-prone. It's just text. A "tool" on the other hand is an API. Its input commands can be validated, not just syntactically, but also whether they make sense in the current context.

Let's say you have a cocktail-making machine. Its hardware can follow a certain list of basic commands, but not all of them are always valid: there's no point in shaking the liquid container if it's empty, for example. You give the LLM the recipe of the cocktail in plain text format, like something you copied from a website, and it needs to translate it into a list of "tool" commands the machine can use. If it does anything else, starts lecturing you about the dangers of alcohol, or just outputs an invalid command, the system will immediately know that the LLM bugged out and may attempt to recover by restarting the process with a different random seed, for example.

CockroachDB scurries off to proprietary software land

We'll be dropping CockroachDB

With a heavy heart, because I think it's a great piece of software, and we'd probably fit the criteria for a free license - for now. But the yearly license renewal sounds like a disaster waiting to happen; if they change their terms for any reason, we're screwed. That's not something we can risk.

Google's ex-CEO U-turns after saying staff 'going home early' killed winning

Re: An Ex-Googler says...

That's some great insight into Google's inner workings, thanks.

It's all drying up: Microsoft to erase 3D Paint from digital store

We're going to need a Microsoft (Windows) graveyard, too

Aero, then touch operation and the interface formerly known as Metro, Cortana, augmented reality, and now so-called AI. I assume much of the latter will be gone too by the time Windows 12 rolls around.

Could we have a no frills version, perhaps? I'd even pay money for it.

Software innovation just isn't what it used to be, and Moxie Marlinspike blames Agile

Tradeoffs

You have an idea for a service, or found a market gap you want to fill with your own solution. Obviously you don't want to release some unstable crap, but time to market is still of critical importance. Unsurprisingly, in such cases you won't be reaching for some kind of bottom-up, close-to-bare-metal solution. Your software will be much slower than what it could be, but that's the price you pay for getting something done as quickly as possible.

Not all projects are like this. But few of them lack deadlines altogether. Without the abstractions and "black boxes" that speed up development, these projects would not be better or worse, they would simply not exist.

The months and days before and after CrowdStrike's fatal Friday

Exactly what would you have done in their shoes? Buy Macs or Linux boxes as "backups"? Use multiple AV vendors? And how on Earth was a company like Ryanair supposed to know that their AV vendor was following insane deployment practices...?

It's easy to complain about missing disaster recovery plans after a disaster, but a company will never be prepared for everything - that's part of why they're paying that AV vendor in the first place!

CrowdStrike blames a test software bug for that giant global mess it made

I don't care

Stop feeding kernel mode code with data downloaded from the Internet, you idiots.

We can't rely on a vendor not screwing something up (as evidenced by this debacle), but all validation and certification from third parties is useless if the validated code can be reconfigured dynamically by crap downloaded from somewhere. This needs to stop. Now.

CrowdStrike file update bricks Windows machines around the world

I suspect that so many high-profile organizations are affected, they'll be drowning in lawsuits. So yeah, I think this is it for them.

Re: Pre-emptive threat assessment

Nope.

"Move fast and break things"

"Release multiple times each day"

The result is here for everyone to see.

Among AI infrastructure hopefuls, Qualcomm has become an unlikely ally

"By combining MX6 and speculative decoding, Qualcomm claims these technologies can achieve a fourfold improvement in throughput over a FP16 baseline."

Considering quantization from 16 to 6 bits reduces the model to a bit over third of the original, this doesn't sound all that amazing to me.

An attorney says she saw her library reading habits reflected in mobile ads. That's not supposed to happen

Not surprised at all

About a week ago, I saw Reddit recommending me an article about someone claiming to have solved the "P = NP" problem. I'm no mathematician, and have not come across this topic for long years, if not decades, having learned about it back at university. I'm not that interested in the complexity topic either, though this in particular caught my attention, as I remembered it was a hard problem people were trying to solve for some time now.

I've only checked out the Reddit comments on the article, spent about a minute on that page, then left, soon forgetting about the whole thing... until, a day or two later, "P = NP" suddenly appeared on my page of recommended YouTube videos.

My browser (Vivaldi) has ad and tracker blocking enabled. But the topic had so little to do with what I usually watch on YouTube, I have a hard time believing it to be a coincidence.

Google thinks AI can Google better than you can

"Find the best yoga or Pilates studios in Boston and show me details on their intro offers, and walking time from Beacon Hill."

These are not three search engine queries. The first might be one, though "best" is a horribly vague quality to search for. The second is something I'd want to read on the studio's own site (among many reasons because I want to make sure they're up to date), and the third is a navigation query I may or may not want to use Google's offering for.

If this is the best they could come up with as an example, then I'm seriously worried.

You want us to think of the children? Couldn't agree more

Re: Parental controls

I was ready to use some sort of "old man yells at cloud" meme for my reply, then I realized we're almost the same age, and my amusement turned into sadness.

Yes, like you, I also spent most my school years with an offline PC. The Internet was something I could only access at school, with serious limitations. I survived.

And as it so happens, both of our past experiences are utterly irrelevant to what's happening with kids and the net today. It's been 20+ years. You know just as well as I do what an incredibly large span of time 20 years is in IT. The net changed, its users changed, and in particular, the way kids use it (and how often they use it) has greatly changed as well.

> What does a kid under 16 need to do with a mobile phone and constant access to the internet that I couldn't do in 1996 when I was 16?

Let's see, off the top of my head: attend classes during the pandemic. Get supplementary materials while sick. Submit homework. Stay in touch with their classmates and teachers - you know, the same thing you've used SMS for, except this doesn't cost money per message.

Oh, and all the above examples aren't something the kids come up with: there are schools all over Europe that have transitioned to pretty much online everything. COVID was a watershed moment in this regard.

To give an approachable example for our generation: the Internet today is at a similar spot television was 20+ years ago. An incredible source of knowledge and entertainment - and if misused, can likewise turn you into a passive drolling zombie. But it's a part of our lives, and shielding our children from it is an exercise in futility. Let us teach them how to use it responsibly instead.

Parental controls

The author was doing so well, too, but had to bring up this nonsense at the end. Reminds me of a recent, overall great TED talk about how current society robs kids of their future, which bizarrely ended with the aforementioned call for banning phones under 16.

Parental controls are the same crap the author decries, just with an added bonus of giving parents the illusion of control. Not only do they never work well, they can even actually be harmful on occasion. Ban Facebook or TikTok from your kids' phone? They will look for alternative ways to get there and can easily stumble across scam sites.

Instead of parental controls, what we'd need is actual parenting. Not something IT has the answer to, unfortunately, but even us tech-wizards have our limits. We can't solve everything, so let's not pretend otherwise.

Australia to fund $620M quantum computer claimed to be first at 'utility-scale'

What's up with quantum computing anyway?

There was a lot of hype surrounding it a few years ago, but things seem to have suspiciously quieted down lately. Or are news simply being drowned out by the AI hype?

Devaluing content created by AI is lazy and ignores history

Re: Meanwhile, in another El Reg article...

Thank you for proving my point, though I could have done without it. >_>

Meanwhile, in another El Reg article...

"AI spam is winning the battle against search engine quality"

That is exactly the problem. You can create quality works with AI assistance - and plenty of human input and refinement. But most AI "content" is created with entirely different priorities: minimal effort with maximum payoff. The typical AI output encountered by people today is garbage, and I doubt labeling would help with that.

Boston Dynamics' humanoid Atlas is dead, long live the ... new commercial Atlas

Re: Good old Boston Dynamics

Well, I'll be damned. They've actually learned their lesson... err, well, in part, if the current video is any indication. Past models of the same type were headless, for example the one they've given to the NYPD around 3 years ago, and the reaction was unsurprisingly abysmal.

Good old Boston Dynamics

They make great robots, but are absolutely hopeless when it comes to making them likeable by humans. (And I'm not talking about anything sophisticated, just things like their robot dogs not having a friggin' head.)

And then they're surprised when some of their pilot projects, especially those in contact with the general public, dive head-first into the pavement.

BOFH: The new Boss, Aiman, is suspiciously good – for now

The name had me completely fooled

Yes, I am duly ashamed of myself.

It's 2024 and Intel silicon is still haunted by data-spilling Spectre

3.5 KB / sec

How practical is that? I could be wrong, but you don't actually know which part of kernel memory you need to dump, right? How long would it take to get everything, which would mean you're guaranteed to find what you are looking for?

Microsoft rolls out safety tools for Azure AI. Hint: More models

Hallucinations about hallucinations

I can hardly wait to see them introduce a third layer of "protection", which will no doubt involve an AI checking whether the hallucination-checking AI hallucinated about the hallucinations. (And I guess it's all recursive from here.)

AI hallucinates software packages and devs download them – even if potentially poisoned with malware

Yikes

This sounds legitimately scary considering all those people trying to make the "AI" build entire applications for them with little oversight. (And even if they review the code, who is going to validate every single dependency?)

Sorry, Siri: Apple may be eyeing Google Gemini for future iPhones

Memory bandwidth should not keep a model from running, though it would likely run slower.

A rotten Apple

It seems as if the company had retained most of its negative practices, but has slowly abandoned its former privileged position of being an innovator and a trend-setter. Turning to Google on the AI front would be especially embarrassing when you consider the models that are openly available today. You'd think Apple would at least build something on them.

Voyager 1 starts making sense again after months of babble

Memory dump

That someone at the design team actually thought "okay, if this component misbehaves and we can send stuff normally, dump the whole memory and transmit that instead as debug info" is something I find truly amazing and impressive.

We asked Intel to define 'AI PC'. Its reply: 'Anything with our latest CPUs'

The claim is already untrue even today

There are open-source models today that do not even fit into 64 GB of RAM, let alone 32.

In other words: this is just the usual marketing drivel.

Copilot can't stop emitting violent, sexual images, says Microsoft whistleblower

AI is a mirror

I am getting increasingly convinced that current AI implementations merely make us come face-to-face with all our hypocrisies and perversions... and we're obviously not prepared to witness it. So the model is given an impossible task: be as human as possible without being as biased and fallible as humans generally are. Obviously, that is not going to work.

Toyota, Samsung accelerate toward better EV batteries

Seeing is believing

Toyota has been promising solid state batteries for years, with absolutely nothing materializing. Maybe Samsung will have better luck, but I will take such announcements with a grain of salt until the actual product hits the shelves.

Dell exec reveals Nvidia has a 1,000-watt GPU in the works

Bitcoin mining 2.0?

I can't help but be reminded of how the power consumption really got out of hand on the mining front - only this time, it's Nvidia itself that is apparently fuelling the flames. Energy efficiency might not be what people are looking for in these accelerators, but this still sounds rather extreme.

Mamas, don't let your babies grow up to be coders, Jensen Huang warns

How often is the output of a compiler unpredictable, including changes between runs, using the same input?

Intel urges businesses to undergo AI PC facelift with vPro update

NPU?

I'm feeling really ignorant, but what's the point of that? I thought AI-related stuff is usually much faster when offloaded to the GPU anyway...

China breakthrough promises optical discs that store hundreds of terabytes

Missing important use cases

Audio made the CD ubiquitous. Video helped the DVD spread. And Blu-ray failed to make the same splash exactly because at that point, many people had an Internet connection that was good enough to just stream whatever they were interested in.

So, what would they be selling on these disks with petabit capacity? Because I doubt archival in itself will let it get off the ground.