hug

https://x.com/drewpusateri/status/2075708238650089981

and

https://x.com/drewpusateri

The frame is a compositional aid, and it does have quite a strict relation to what the camera is capturing.

The best version of your complaint is that the viewfinder does not have accurate frame lines that account for parallax at various distances, which is true, but there are still ways of using the viewfinder to assist in composition.

If you take two or three shots at two meters, and look at what the camera captured versus where your composition was in the frame, you can immediately intuit a baseline for where the frame lines 'should be' on the viewfinder at two meters. "Occupies the bottom right two thirds of the viewfinder" for example.

Without a reference frame, these intuitions (and your aiming of the camera!) are going to be far less accurate.

It's totally incorrect to state that the area served by either Sydney or Melbourne airport has less economic activity than the area served by "any airport in the US", or even the vast majority of US airports, so whatever laptops-at-airport (and I suppose airpods-in-ears) is a proxy for it sure isn't economic activity.

Yes, I did, and the reason is super straightforward: It's a hardware portrait panel, mounted sideways.

Getting from zero to a fully working OS was a mild journey, but I'd do it again.

This is the primary reason the Minibook X won out in my searches: It's the only small device that has a keyboard layout that puts all of the keys in the right spots.

They're sometimes an odd size, but when I hit the wrong key due to a sizing constraint, I don't even have to think: Backspace, hit the right key with mildly adjusted positioning.

I've tried a few machines with different layouts, and that's never the case - and having to stop and look at the keyboard to find a key interrupts flow in the worst kind of way.

I have this laptop, and it is amongst the best laptops I have ever owned, despite being awful in many ways. It has almost completely replaced my use of my M4 Macbook Pro, simply because I always have it with me. That, and it can run Linux.

I don't share the complaints of the OP about the keyboard or the screen, though. The keyboard is fine, I can hit about 110WPM on it, slower than my regular pace, but enough that there's no dramas. The layout is great: Occasionally there's keys that are too small (looking at you, apostrophe) but everything is at least in the right spot, which is way more important.

The 2K display at 10" is high enough DPI that everything is totally crisp, and you can unlock ~95Hz (bad for video, good for everything else) with a bit of a tweak. You can also smash a byte into the EC at the correct offset and access the full unrestricted BIOS -- mostly to crank the RAM up to 4800MT/s.

I'm running vanilla Arch with Niri and Noctalia, and it's a dream. It's my primary dev machine, used in combination with a remote server with a tonne more grunt. If it broke tomorrow, I'd buy another - and I wouldn't do that with my macbook.

To the OP:

* Accelerometer support, EC-byte-bashing to get BIOS unlock: https://github.com/greymouser/minibook-x-tools

* 95Hz EDID fix: https://github.com/sonnyp/linux-minibook-x/issues/7#issuecom...

It’s not true of their cheaper products, the POs especially.

The EP series are also good bang for buck.

The OP series, along with the TP-7, TX-6 are luxury goods. If you’re trying to save money, they’re not for you. They are nice though.

I don't mean to tar you with a too-wide brush, and I feel like you have a good handle on your personal acceptance for LLM assistance. No complaint there.

I do think, maybe alternative to your view, that LLMs can provide useful feedback to graduate-level employees in most fields.

It is not that the work can be done by LLMs -- we're not there, yet, in software or otherwise -- but that LLMs as useful tutors specifically in regard to denouncing known bad ideas is largely applicable all over.

What I mean by the above is that I have yet to find a truly interesting idea spun from whole cloth by an LLM. They're mediocre at it. They're trained from the aggregate thoughts of those in every industry, and you and I both know that the aggregate of the industry is, generally, mediocre.

Conversely, though, is the hit: They won't be worse than mediocre. An indefatigable tutor who gives no great advice but will counsel you against blowing yourself up (or cutting a limb off with a rope, or falling overboard) is, to me, worth an amount.

The failure modes will get better, the advice will get better. Are we there, now? Unsure. You can tell us all better.

On the ten year horizon, I'd place a bet, though.

Yes, but no room is made for people who see no use for it. There is a forced-consensus that this technology is useful, which I have to combat against at work.

This is the crux of the issue -- The technology is useful. Using it appropriately is probably the thing that people are ignoring, but you're conflating one and the other in your comment.

It is not useful to you in this case, and complain that it is an overall detriment in your industry. Those are fine and reasonable statements and conditions, and I see no reason to disagree with them... But your first statement, people who see no use for it? That is, to me, as off-putting an opinion as the consequence-unaware hypebeasts who are running OpenClaw with access to their trading accounts and can't see why others aren't.

I sympathise with the idea that everyone wants to use the new hammer and so is treating every problem like a nail, but hammers are still pretty good tools. (And you can ignore the ex-NFT-fans hammering on their dicks in the corner.)

It's DeepCAD* output, it looks like, which is a JSON payload that is the sketch / extrude / whatever steps, which is itself based on Onshape output.

Looks like you can go JSON -> step files, but not really in such a way that you can modify any of the operations.

* https://github.com/mightyhorst/DeepCAD

That’s also an algorithm. An unsophisticated one, but an algorithm nonetheless.

You can (and should) argue that such a simple algorithm doesn’t “count”, but fundamentally the exact wording of the grandparent post never works, legislatively.

Lawyers will lawyer.

Actually, OpenAI made a similar claim about one of their GPT models a while ago…

Funnily enough that was while Dario Amodei was their research director.

I dug a little into this because I was curious who was more correct here.

From wikipedia, which links to what seems like a relatively reliable source:

The Singapore Department of Statistics broadly defines "Chinese" as a "race" or "ethnic group", in conjunction with "Malay, Indian and Others" under the CMIO model.[10] They consist of "persons of Chinese origin" such as the Hokkiens, Teochews, Hainanese, Cantonese, Hakka, Henghuas, Hokchias/Foochows, Shanghainese and Northern Chinese, etc."[11]

So I would, on the balance of things, think that kccqzy meant what they said, and was pretty correct about it.

It was absolutely not the case two decades ago. There were no other options for an enterprise fleet, 20 years ago, if the question was asked. If you weren't Google (who never asked the question anyway), the answer for managing 25,000 endpoints was to use Windows devices with Active Directory as the management plane. Anyone doing anything else was in for a world of hurt... and that's why every enterprise ended up on Windows, and why everyone targeting enterprise management targeted Windows -- because that's what the endpoints were already running.

What killed AD & GPO was Microsoft, in their bullheaded push toward Azure everything. Instead of listening to what it was that the enterprise customers actually wanted, they designed a system that made sense to them, but to no one else. The original UI was written in Silverlight. It was horrific.

Group Policy and Active Directory are dead, for all intents and purposes.

It's now Intune (via OMA-DM), and Entra. Both of those products are about as bad as you might imagine the "cloud" versions of GP & AD might be.

They are better, in ways -- no longer having to care and feed for domain controllers is nice, and there's no longer an overhead for additive policy processing, so endpoints only get a single set of policy and log on much quicker -- but for the most part, enterprise management of Windows devices is in a worse place than it was ten years ago.

Try to figure out how long it will take an online Intune device to discover a new policy: As far as I can tell the answer is "eventually". There are bandaids for this, because of how infuriating it is, of course, but all time guarantees are basically gone.

Ask me a decade ago what an enterprise should do, and my answer would be straightforward: AD, GPO, Exchange.

The answer now is not simple.

...Embarrassingly, I have typo'd in my original post, and it's too late to edit. Pints are 570ml (not 470ml) everywhere on the East coast - hence why a half pint in Tassie is often called a ten - because it's 10oz, or half a 20oz/568ml pint.

I was decidedly not old enough to drink in 1990 and culture in general in Australia was much less homogeneous back then, so you're probably right for the times.

As someone heavily involved in the hospitality (read: beer) area, this doesn't really line up with reality in Australia: there's only one state (South Australia) that doesn't agree on the major standard sizes: Pints are 470ml, schooners are 425ml, a half pint is 285ml, and a pony is 140ml.

There's colloquialisms for a half: pot, or middy, mostly. Hobart will call a half pint a ten, because it's 10oz, but they also know what you're talking about when you ask for a pot or a half pint.

Then there's South Australia, which will serve you a pint at 425ml, a schooner at 285ml, no one there outside of specialty craft beer bars have any idea what a half pint is, and if you want a proper pint you need to ask for an imperial pint. I have never seen an 'imperial pint' advertised in Hobart - it's just called a pint there.

Source: I have pretty extensive drinking experience in pretty much all of the Australian capital cities, except Perth.

If I had to distill the social status & commensurate behaviour described in your in two words, it might sound suspiciously familiar.

You shouldn't need a medical background to know that having something press on a spot for a couple of hours will leave a depression in your skin.

You have probably fallen asleep on something patterned or folded and have it leave an impression on your skin before: This is no different.

Other places it happens: Watches that are slightly too tight or have ridden up an arm. Glasses arms pressing against your temple or behind the ear. Tight socks after a day wearing them.

It's not a medical problem. It's just general physics.

Not using the combination for one of its ambiguous purposes does not strip it of ambiguity, you've just trained yourself to avoid those circumstances.

That, of course, is one of the pain points that the article addresses: Training yourself to do so is additional cognitive load that never should have been necessary in the first place.

I flip between macOS and Linux and, occasionally, Windows. On one of my laptops, insert is also a Fn switch away, so I have to either remember that this machine needs Ctrl-Fn-F11 specifically when I'm copying from terminal.

On another keyboard I have the same problem, but insert is mapped to a different key entirely, so it is ctrl-fn-equals, and fn is on the opposite side of the keyboard from ctrl.

Contort my fingers in which way on which keyboard? Mental load and annoyance I don't need.

People will argue with you because your initial quoted sentence is chock full of fallacies.

* Caddy's complexity (especially when it comes to TLS) is not arbitrary, it's to meet the needs of auto-renewal and ... y'know, hosting sites on TLS.

* Caddy's SDLC is not, as far as I understand it, especially rapid.

* Implying that "military grade" is some level of encryption beyond the minimum level of encryption you would ever want to use is silly.

* The Manjaro website is not "the equivalent of a poster", and in fact hosts operating system downloads. Operating system integrity is kinda important.

You may have reasonable arguments for sites that are display only, do not out-link, and do not provide downloads, but this is not one of those circumstances.

About the only thing about the weather I can tell from my window is whether it is currently raining or not.

The temperature inside is not at all indicative of the temperature outside, the sun being out doesn't mean it is warm, and I don't really have any useful indicators of wind, unless the windows are rattling, but that doesn't let me know if there's a stiff breeze.

I could walk over and open up my balcony door and experience it all personally, but checking my phone or watch is faster and more accurate, and also gives me the forecast at the same time.

Based on both assessment, I estimate the overall compromise period spanned from June through December 2, 2025, when all attacker access was definitively terminated.

FTA.

Spoofing the user agent and decoding capabilities and [...] is a useful way to unblock things that are crippled on various browsers, indeed.

The problem here is requiring hardware-attested DRM: Widevine L1 on Edge on Windows, and Apple FairPlay on Safari on MacOS. The only way to get hardware attested DRM is via browser specific (i.e.: native code) support that interfaces with the OS & GPU drivers. You can't get there through an extension.

It's this section that puts the lie to the entire project: https://github.com/Pickle-Pixel/netflix-force-4k?tab=readme-...

It appears to only be useful on Edge on Windows.

Did I, though?

I understand it spoofs all of the checks it can, but the only Chromium browser that supports Widevine L1 (a requirement for 4K) is Edge, so even if all of the check spoofing works, it still won't do 4K.

There's even a table in the README that describes this exact scenario.

I may be an idiot, but: What does this actually, y'know, achieve? It seems the answer to me is probably nothing?

It doesn't work on Firefox. It appears not to work on Chrome. The suggestion is to use Edge, which on Windows already gets 4K support in Netflix anyway.

Of course -- but CentOS' upstream was RHEL, no?

Because in its brevity it loses all ability to defend itself from any kind of reasonable rebuttal. It's not an actual attempt to continue the conversation, it's just a semantic stop-sign. It's almost always used in this fashion, not just in the context of LLM discussions, but in this specific case it's particularly frustrating because "yes, you're holding it wrong" is a good answer.

To go further into detail about the whole thing: "You're holding it wrong" is perfectly valid criticism in many, many different ways and fields. It's a strong criticism in some, and weak in others, but almost always the advice is still useful.

Anyone complaining about getting hurt by holding a knife by the blade, for example, is the strongest example of the advice being perfect. The tool is working as designed, cutting the thing with pressure on the blade, which happens to be their hand.

Left-handers using right-handed scissors provides a reasonable example: I know a bunch of left-handers who can cut properly with right-handed scissors and not with left-handed scissors. Me included, if I don't consciously adjust my behaviour. Why? Because they have been trained to hold scissors wrong (by positioning the hand to create opposite push/pull forces to natural), so that they can use the poor tool given to them. When you give them left-handed scissors and they try to use the same reversed push/pull, the scissors won't cut well because their blades are being separated. There is no good solution to this, and I sympathise with people stuck on either side of this gap. Still, learn to hold scissors differently.

And, of course, the weakest, and the case where the snark is deserved: if you're holding your iPhone 4 with the pad of your palm bridging the antenna, holding it differently still resolves your immediate problem. The phone should have been designed such that it didn't have this problem, but it does, and that sucks, and Apple is at fault here. (Although I personally think it was blown out of proportion, which is neither here nor there.)

In the case of LLMs, the language of the prompt is the primary interface -- if you want to learn to use the tool better, you need to learn to prompt it better. You need to learn how to hold it better. Someone who knows how to prompt it well, reading the kind of prompts the author used, is well within their rights to point out that the author is prompting it wrong, and anyone attempting to subvert that entire line of argument with a trite little four-sentence bit of snark in whatever the total opposite of intellectual curiosity is deserves the downvotes they get.