realityfactchex
realityfactchex on Hacker News
Second, never use the "apps" built into the TV.
Next, don't plug Ethernet into the TV. Plug HDMI cables into the TV.
Use a box of your choice on the other end of the HDMI cables. This is now the box that you trust. (Roku? Shield? Blu-ray? MacBook? PC? Linux? Playstation? XBox? Whatever you want.)
The Smart TV itself is now fine. Choose the devices you plug into it to play content. The TV just shows it.
- you (GROUP 1) are the 0.001%,
and you
- hire the 0.030% (30x more unclean than GROUP 1 on some scale),
to
- build for the 1.000% (33.3% more unclean than GROUP 2 and 1000x more unclean than GROUP 1 on some scale),
then I could see how
- "building for the 1%" would actually feel like building for the general public.
In 2021, Backblaze saw ~14% failure rate at 7 years 9 months. Drive useful life: about 6 years.
That gave us what was a stable-seeming, old "replace drives at 5 years, or you're getting risky" rule of thumb (unless you prefer to replace on failure, of course).
In 2025, Backblaze saw ~5% failure rate at 10 years 3 months. Drive useful life: ~10 years, roughly!
Thanks for all the HDD improvements, industry!
What's wrong with the G502? I like the wired G502 pretty well. Ideally I'd tweak the shape to be a little more like the M500s shape, but I'm happy enough with it overall.
Great question.
As a very satisfied, long time user of 2 extensions in combination, Temporary Containers and Firefox Multi-Account Containers, I wonder the same thing.
I found out the hard way some time back that simply disabling (not even removing it) and re-enabling the Firefox Multi-Account Containers extension would cause all the painstakingly set up containers and their saved credentials/data to get "lost".
If there were to be a way to "export" and "import" sets of named containers with their settings and data, that would be great for not losing which-site-maps-to-which-container-and-is-already-signed-in (whether natively or within an extension).
I buy the highest quality, lowest cost stuff I can find.
There can be a positive story here maybe though:
- More doordash = less grocery
- More farmer's market = less grocery
- More home-grown = less grocery
Though TFA points out,
80% of Americans say they're still trying to spend less
28% are actively trying to cut back on groceries
56% are trading down to lower-priced brands
49% are simply buying fewer items,
44% are leaning harder on coupons and promotions
Effectively, it sort of does that, but really it just listens to the wakeword and opens/switches to the requested app & modality.
FWIW, I get a very different functional result using the Shortcut method vs. asking Siri to delegate natively. To compare, I asked Siri (non-beta here) now to "ask ChatGPT <x>" and I got a top-card with some fairly low quality SEO-ranked weblinks.
1. In Shortcuts app, make shortcut named "AI Voice Mode" (or whatever you want, YMMV)
2. Set it to run the ChatGPT action "Voice Mode" (requires at least the minimum paid tier, I think)
3. To trigger, say "Hey Siri, AI Voice Mode" (or whatever you called the shortcut)
WHAT YOU ARE SAYING
WHAT YOU NEED FROM YOUR AUDIENCE (RESPONSE OR DECISION OR ACTION ETC)
As time goes on, often I say (to myself) "forget that", and write all the detail that is needed anyway, even in email. But only for audiences that may care about the detail (or otherwise are safe to skip the email altogether).
But who uses email at work anymore, anyway, right? I guess some organizations.
That under-states the matter. It is a terrible time, price-wise, to build a NAS.
I'd almost rather have no AI whatsoever and have storage 1/10 the price of pre-AI times.
(If there were a magical choice between having AI and significantly more expensive storage, and having no AI and some program to dump that investment money into getting and somehow leveraging significantly more available storage, that is.)
I'd say it's a "small keyboard" in width and in height.
Mechanical, can be fairly quiet (with red switches), arrow keys, F keys, home/page keys. Well-built IME. Not exorbitantly priced.
It does not require a "service" to consume multiple blogs efficiently, just simply "software" (an app will do).
The below RSS readers pretty much do their job and get out of the user's way. What's amazing is the sheer number of RSS readers that people choose to use; there are so many, people find ones that they personally like.
On desktop: Vienna https://www.vienna-rss.com/
On iOS: R2SReader https://apps.apple.com/us/app/r2sreader/id1608635613
Of course, if one wanted to use a service (or run one oneself), for features like syncing, then that is possible too.
But the basic setup remarkably simple, easy, and useful by itself IMO.
Because of a load-bearing CSS attribute, as I understand it.
So, it's hard to consider it a web-standards supportive platform anymore.
Or, at least, it's a web platform with a technical challenge of not being interactive (so users can post and interact) from web standards supporting devices but lacking whatever HTML standards were introduced since as recently as 2021.
I'd call it a technical challenge. Literally the CSS language framework / build process is just not that flexible.
Semantic design development process became separated from semantic HTML serving somewhere along the way.
Maybe that's fine and quite good for 99% of uses. But I see this one as a glaring technical question mark.
Bringing it back to the titular point in the OP, the "crappy forums" do not seem to (cause some users to) suffer from this problem.
[0] https://meta.discourse.org/t/dropping-ios-15-other-old-brows...
I read it in my early 20s and at that time my main takeaway, as I recall, was to (paraphrasing), "carefully plant the evidence, thus making the conclusion something the audience comes to on their own (they 'think of it' themselves)".
In entrepreneurship, this was slightly dangerous to me, since I immediately started implementing this successfully. Before long, I (apparently) took "credit" for "my idea" (that I successfully got the audience to see, oh so carefully), whereas audience thought that we "came up with it together"!
Oy vey! So, a word to the wise. I now subscribe to more of a "plain honesty but with tact" approach.
Underlying integrity is rigor.
Underlying rigor is education.
It goes deep, for sure, IMO.
I tried two of the Krea 2 models in LM Studio, but loading the downloaded models errored out. (Maybe I'm doing it wrong, since it's an image model.)
Previously: https://news.ycombinator.com/item?id=47800562
> any evidence of this.
"Any evidence" you say? I think something has moved.
As of Autumn/Winter 2025, I can say "Here's a complete enough spec of what I want cloned. Crank on it for a few hours. Give me the clone site as I specified." And frontier agentic tooling does a hands-off YOLO job really well. (Notice: I set it up with good specs/rules to scaffold.) Cost maybe an hour of my time to set up/scaffold, and 3 hours cranking on its own, on a $20 or $60/mo sub.
I think taking the same problem to an "offshore house" (or even Fiverr or whatever) would probably easily cost 10-100x more, and quite possibly with worse (less reusable or less best-practices code quality or less functional or less clean) output overall.
So I think your OVERALL point may stand. I'm just nitpicking the specific aspect quoted above, and maybe to some degree the aspect of
> "LLM generated code is similar, but arguably more expensive and lower quality"
Which to be fair obviously depends on what code one is comparing.
I used to say "nah, it can't replicate apps that well by itself", but then I tried it (on the advice of a fellow commenter here on HN), and was surprised myself.
In any case, you've inspired me to post the original reply I had composed for https://news.ycombinator.com/item?id=48575653 (the immediate parent to your comment), below. This is what I wanted to say, before then deciding to just be grateful for the sharing of the parent's perspective:
""" I'd frame AI as a plausible hypothesis engine, not as a working scientist (yet). I think AI can do some things that look like rational analysis (far better than many/most humans much of the time, perhaps), but I reserve that (most rare and prestigious and important) activity of actual science for humans too, when it counts, for sure.
I get the main article is about the very real "chaos/threat" of no funding, not the "chaos/threat" of AI-articles/"research" nor the "chaos/threat" of "real issues in the state of Science (before funding crises)".
IMO, the state of Science (before funding crises) could be, perhaps, inextricably (though not overtly) linked to the later/current chaos/threat of markedly reduced funding. No? Maybe it's not stated anywhere, but it seems oh so likely, reading between the lines.
If funding cuts, in the medium to long term, lead to a good thing (which would be the best we can now wish for -- and, after all, everything comes and goes in and out with the pendulum of time), it will be a much needed "reset" of science onto a more honest (and net knowledge-learning productive) model.
It (Science) was, arguably, already well by the wayside. Not just sort of expensive (though not very, compared with other budget items). But more importantly: inefficient (to put it nicely). And more importantly still: often (perhaps more often than not!) plain wrong. And that means, sadly: fairly/largely ineffective (degree depending on the domain). Which is the opposite of what is wanted. If you're going to do Science, it should at least be valid, or if it's not, it should be possible for those in their own field to tell that it is not. Else, it's kind of broken.
And if it doesn't serve it's purpose, what can you do but reboot. Reset. Just like a computer.
At least Science can be rebuilt. You just start doing it again (with what you have/can). With more rigor.
Maybe this reads like more of the same. But I don't think "being well funded" correlates well with "doing good science". (Only if the science is measured by the paychecks. Which is economics, not science.) """
At least it's a good thing that we're able to a) observe and b) talk about and c) acknowledge openly(ish) that academic, mainstream, practicing "science" (including as visible to microscopists and all that entails) is currently a "mess".
This allows us to, eventually, address those issues (or die trying!).
Science used to move at a pace of one lifetime after another (pearl clutching 'til the end and confirmation bias and careers built on saving face and economic entrenchments all that).
But I hypothesize that with AI, we can point to "a thing that is not a person with all that is bundled up with that" and say "look, maybe this other train of thought is worth entertaining". Not to say the AI is right. Ideas will stand or fall on their own merit. Just that an AI is not a person outside the field. Normally, an outsider says something, nobody listens. But, if an anonymous AI says something (of course, cleaned up for voice and concision and validated by a human as a first pass), the worst you can say is "ok prove it" or "here is where that is wrong". Instead of: deafening silence.
In other words, I hope AI augments our ability to have those hard conversations that need to be had. Without people losing their jobs due to their prior (understandable) errors, and within the spirit of always using the best available information.
I shared this optimistic indirect use-case for AI with (less technical) friends recently, and they literally were speechless and finally one person said "you're the only one who thinks that".
Am I right, though? There's a there there, isn't there?
I don't want to self-host my photos, too much management.
I don't want to use Apple or Google or MS Clouds for various reasons.
I do want to support a pure-play, independent-ish, profitable, consumer friendly platform. To upload hires shots and have them easy to tag/share/access among those to whom they may be of interest.
I expect to pay, but not through the nose. Reliability matters. I would want the company to be around in 20 years and still reasonably priced/useful. Suggestions?
I liked Maemo 5. Having never used the n810 nor Maemo 4, I suppose didn't know what I was missing.
iPhone did not get copy-paste until mid-2009!
Eventually my n900's microSD jack broke off the board, which I've read was common.
As an original N900 user, I got one of the eBay "refurbed" N900s from China I think a few years ago for fun. It was a piece of junk, literally, like arrived with broken keyboard etc. A clear case of false advertising. I got a full refund.
YMMV. I was really thinking I was buying a proper refurbed N900. Maybe they're out there. Buyer beware.