hedgehog @ hedgehog @ttrpg.network

Posts

2

Comments

851

Joined

2 yr. ago

You say that like it’s not compatible with the ones people already have.

Each credit reporting agency offers this option, at no charge …

It is highly recommended to lock your credit. Frankly, it should be locked by default. In September of 2017, Equifax announced a data breach that exposed the personal information of 147 million people.

Note that, before this incident, it wasn’t consistently free. I remember it being free to lock, but costing $20 or so to unlock. A law passed in 2018 required credit bureaus to offer freezes and unfreezes (and to fulfill them within certain time frames) for free.

Also note that you might need to look for a “freeze” instead of a lock. Experian charges $25/month for their “CreditLock” service, for example, but they offer a free security freeze.

Heads up that this article is from March 2021.

Are you seriously asking why a commonly supported 27 year old format doesn’t need a fallback, but a 2 year old format does?

From the article:

The “$100 for 1,000 likes” that influencer Gyan Abishek mentioned in one of his videos seems to be greatly exaggerated, based on the various payment portals that I have seen.

Someone else they quoted said $3-$10 per 1000 likes, which is still quite a bit.

I haven’t dug into the test data or methodology myself but I read a discussion thread recently (on Reddit - /r/jpegxl/comments/l9ta2u/how_does_lossless_jpegxl_compared_to_png) - across a 200+ image test suite, the lossless compression of PNG generates files that are 162% the size of those losslessly compressed with JPEG XL.

However I also know that some tools have bad performance compressing PNG, and no certainty that those weren’t used

I mean, I’m just pronouncing the letters aloud: A-P-N-“Gee.”

Compared to something like JPEG XL? PNG] is hands down worse in virtually all metrics.

Until we circle back to “Jpeg XL isn’t backwards compatible with existing JPEG renderers. If it was, it’d be a winner.”

APNG, as an example, is backwards compatible with PNG.

If JPEG-XL rendered a tiny fallback JPEG (think quality 0 or even more compression) in browsers that don’t support JPEG-XL, then sites could use it without having to include a fallback option themselves.

Probably the least relevant benefit of APNG over GIF: Unlike GIF, I can even pronounce APNG with a soft G and not feel gross about it. (Like I’m betraying the peanut butter brand and my entire moral framework at the same time, y’know?)

I should’ve tried that. I ordered an Apple Watch silicone band dupe (direct from the vendor’s site) for like $5. Same band would have been $50 from Apple, except the color I wanted wasn’t even available anymore.

Band came in the wrong size, so I wrote the vendor. They apologized, might have refunded me, and sent me a pack of like 10 different colored bands. Unfortunately those were all in the same size as the first one

Wow, what a terrible set of moves by whoever at AMD made that call. Lack of CUDA support is the only thing keeping me from buying AMD GPUs, and I’m pretty sure I’m not alone.

This product looks awful.

First, ever since they added ads to Google TV back in 2021 (even on the Nvidia Shield TV), it’s been a subpar experience. Well, it was for me, at least - maybe it’s improved, but I switched to Apple TV as a result and haven’t looked back.

Second, why would anyone get this over an Nvidia Shield TV or an Apple TV, other than ignorance or an incredibly strict budget? The Apple TV 4K is $130/$150 new and the Shield TV is $150 new. The Shield TV, which came out in 2017, is faster than this. The Apple TV 4K is 16x faster. And if you get either refurbished, get an older Apple TV,

For anyone on a strict budget, the $30-$50 Chromecasts make way more sense than this device. Yes, they’re ending production of those, but there are still competitors near that price point.

The only thing I can think of is that they’re banking on brand recognition or are hoping the segment of people without smart home hubs who are unaware of alternatives (like the $35 SmartThings Hub Dongle) and who aren’t in the Apple ecosystem is big enough.

Yes - you can set multiple daily limits (they reset at midnight and that can’t be changed), and each one can apply to one or more apps, categories, or websites. You can also select almost all the apps in a category and omit a couple, but then future apps in that category won’t be limited automatically. And you can choose specific apps to never be limited.

So you could set a 3 hour limit for Social apps, Games, a couple individually chosen other apps, and some specific websites, as well as a 5 minute limit toward the Facebook app and facebook.com, if you wanted.

If you mean the screen time tracking, then I don’t know think you can do that, but it gives you both your overall time as well as breakdowns by category (at least the top few categories), so you can do the math on your own.

You could also change the PIN to something random that you’ve written down somewhere inconvenient, or stored in a password manager.

And Apple will never just let users decide that. They consider it anti-user to force us to make choices.

Apple lets you set app, category (“Social” is a category), and website-specific limits, though, so you can absolutely make that choice.

It's relevant because, when people talk about "AI" that's not actually intelligent (ie. all AI), they're being incoherent. What exactly are they talking about? Computers in general? It's just noise, spam, etc.

If your objection is that AI “isn’t actually intelligent” then you’re just being pedantic and your objection has no substance. Replace “AI” with “systems that leverage machine learning solutions and that we don’t fully understand how they work” if you need to.

Did you watch the video? Do you have any familiarity with how AI technologies are being used today? At least one of those answers must be a no for you to have thought that the video’s message was incoherent.

Let me give you an example. As part of the ongoing conflict in Gaza, Israel has been using AI systems nicknamed “the Gospel” and “Lavender” to identify Hamas militants, associates, and the buildings that they operate from. Then, this information is rubber-stamped by a human analyst and then unguided missiles are sent to the identified location, often destroying entire buildings (filled with other people, generally the family of the target) to kill the identified target.

There are countless incidents of AI being used without sufficient oversight, often resulting in harm to someone - the general public, minorities, or even the business who put the AI in place.

The paperclip video is a cautionary tale against giving an AI system too much power or not enough oversight. That warning is relevant today, regardless of the precise architecture of the underlying system.

AI can be used in automated solutions and it doesn’t intrinsically have to be supervised. It being or not being intelligent is irrelevant - it can still cause harm.

I use --format-sort +res:1080, which, if my understanding of the documentation is correct, will make it prefer 1080p, the smallest video larger than 1080p if 1080p isn’t available, or the largest video if nothing 1080p or larger is available.

res is the smallest dimension of the video (so for a 1080x1920 portrait video, it would be 1080).

Default sort is descending order. The + makes it sort in ascending order instead.

And for any of the "AGI won't happen, there's no danger"...what if on the slightest chance you're wrong? Is the maddening rush to get the next product out without any research on what we're doing worth a mistake? Scifi is fiction, but there's lessons there too, and we're ignoring them all because "that can't happen" is stronger than "let's be sure".

What sorts of scenarios involving the emergence of AGI do you think regulating the availability of LLM weights and training data (or of more closely regulating AI training, research, and development within the “closed source” shops like OpenAI) would help us avoid?

And how does that threat compare to impending damage from climate change if we don’t reduce energy consumption + reliance on fossil fuels?

Besides, even with no AGI, humans alone can do huge damage with "bad" AI tools, that we're not looking into either.

When I search for “misuse of AI” I get a ton of results from people talking about exactly that.

My guess is they thought they were 99% done but that the 1% (“just gotta deal with these edge case hallucinations”) ended up requiring a lot more work (maybe even an entirely new sub-system or a wholly different approach) than anticipated.

I know I suggested the issue might be hallucinations above, but what I’m genuinely curious about is how they plan to have acceptable performance without losing half or more of your usable RAM to the model.