For the best experience on desktop, install the Chrome extension to track your reading on news.ycombinator.com
Hacker Newsnew | past | comments | ask | show | jobs | submit | history | more simongr3dal's commentsregister

I think one of the reasons everybody keeps producing content with massive background bokeh is because of digital video formats[1], having a blurry background and a relatively still subject in focus requires relatively little video bandwidth, so the video can still look good even with a kind of crummy 2-4 Mbps encoding.

1: edit, I guess it's mostly about having to stream it over the internet


The big affected third-party clients would gladly have incorporated the ads into the timeline, if only twitter actually included them in the api response.

Or they could have made a promoted tweets api and mandated in their terms of service that third party clients showing a timeline must include promoted tweets every now and then.


There's a lot that goes into showing ads, prebidding, making sure there's no nsfw content, etc.

Even if they implemented it correctly you wouldn't really trust a 3rd party to show your ads and risk strikes on your account.


I don't understand your thought process. Who is owning this "account" on which "strikes" are going to happen? Twitter?


in Twitter's case it would be direct accounts with the advertisers.

you have to make sure you only show certain companies ads next to certain content, or they will pull out.

therefore you need to control the client. only show tier 2 ads next to nsfw content, etc.


Is there even such a thing built in to the current promoted tweets functionality?


i'm not privy but i very much assume they do have different promoted tweets depending on whether you have the nsfw filter on.

they also aren't going to show certain promoted tweets on certain hashtags, topics, etc.


Ha! I just read the account bio, we’re talking to a damn robot.

Doesn’t HN have a policy against this sort of thing?


i'm not actually a robot, the bio is just a joke.


I guess it's impossible to report this kind of fraud to law enforcement? I just don't see why.


Of course you can report it. I doubt anyone is going to give a fuck though.

But just wait for the next protest march in your city. "Law enforcement" is going to show up by the hundreds for sure.


Your local police officer isn't going to march to Seattle and confront Jeff Bezos or Andy Jassy. But neither is my prime minister going to care at all about a breach of the trades description act.

So Amazon operates in this space in the middle with almost zero enforcement.

The only real opportunity for recourse we have is to moan on social media, and hope we discourage future sales enough to make a dent in their profits.


It depends on the local regulations. It's not always the LE in the traditional sense either. For example in Australia ACCC would likely be the right body to deal with it.


Please present them instead of coming with a (bad-faith) shallow dismissal. I'm sure many here would find value in the perspective, even if only to try and find flaws in the argument.


The blanket statement is obviously untrue. For example: people are forced to obey the law.

But if you only want to talk about suicide: we already "force" suicidal people to stay alive via suicide prevention measures (nets under bridges, restrictions on drug purchases etc.) that make it harder or less convenient to kill yourself. These measures reduce suicide rates and I'd say they're moral.

Okay, so what if we're only talking about "assisted" suicide? If someone really wants to die, should we let them? I get where you're coming from, but let's not pretend it's always this simple. Suicidal people aren't famous for thinking rationally, and just because someone says they want to die that doesn't mean we should take it at face value. There are already cases in Canada of people seeking euthanasia not because they're terminally ill but because they're poor, and see suicide as preferable to poverty/homelessness:

https://www.spectator.co.uk/article/why-is-canada-euthanisin...

There's the "slippery slope" we were warned about.

PS I'm not sure what you think "bad faith" means, but my post was not written in bad faith.


> via suicide prevention measures (nets under bridges, restrictions on drug purchases etc.)

Those are all negative, in those cases you're not allowing someone to do something (which is different from forcing someone to do something). There are other forms of suicide prevention though, like forcing suicidal people to take medication


In the UK, if you want to kill yourself by overdosing on paracetamol, then you're forced to visit several different shops (because you can't buy more than two packets at once) then spend a long time tediously popping each pill individually out of the blister pack before you can swallow them all. (The pills don't come in bottles because bottles make it too easy to quickly pour a fatal overdose of pills into your mouth).

These measures won't stop a truly determined person, but they stop a lot of people. It's well understood that erecting even a minor inconvenience in front of a suicidal person can be enough to keep them alive. Popping all your paracetamol out of the packaging takes enough time to give you a chance to reconsider the whole thing.

By forcing someone to do something, we prevent them from doing something else.


I don't want to not kill you.

I don't want to safely dispose of this industrial waste because that is too expensive and time consuming.

I don't want to feed my children.

I don't want to drive the speed limit.

I don't want to pay for that.

I don't want limit my amplifier's output to 1500 watts on the amateur radio bands. (edit: 500 watts on HF in France!)

I don't want to provide the good or service I have been paid for and I don't want to return the money.

I don't want to add to this list anymore....


Using "bad faith" like this destroys any and all meaning in the phrase. It also directly violates the hacker news guide lines.


I just think that people generally have pretty high expectations of OSes these days. So the job of getting it to run on actual hardware, and getting it to run on lots of hardware is just kicked down the road until the OS is a lot more mature.


The article mentions that "health activists" have been denied access to the bases, I don't think it's too much to imagine that Japanese authorities could also be facing significant bureaucratic hurdles to gain access and actually determine the source of pollution.

They give an example of both military and civilian firefighters doing training that involve using a lot of fire-suppressing foam. These foams contain carcinogenic compounds that can be washed into the groundwater by rain.

In Denmark we have recently had just that happen to people living near firefighter training grounds [link]

link (machine translated since I couldn't find anything in english): https://www-dr-dk.translate.goog/nyheder/indland/store-maeng...


From the article:

> If this is all too involved for you, or you only want to seed a small subset of the data, then it might be easier to pin a few directories: 1. Use a VPN. 2. Install an IPFS client. 3. Google the “Pirate Library Mirror”, go to “The Z-Library Collection”, and find a list of directory CIDs at the bottom of the page. 4. Pin one or more of these CIDs. It will automatically start downloading and seeding.


Any instruction that involves "Google X" (or even a general term "Install Y"), is not a good instruction. It is like a maze - easy to go once you know the path. Otherwise - dozens of subtle things that can go wrong, at each step (e.g. installing a different version of Y, or to a wrong path, etc).

The best instructions can have descirptions, but also contains a FULL list of command lines instructions, ideally without any placeholders.


It's presumably deliberate: by keeping annas-blog.org clean from copyrighted content (or links to copyrighted content), they will be able to retain a public channel despite the illegal nature of their work.


I usually tend to agree, but in this case the link would be a link to pirated content and would quickly get the site taken down. So "google XYZ" is the only viable way here


There is a "wine brick"-type instruction workaround (https://en.wikipedia.org/wiki/Vine-Glo):

"After dissolving the brick in a gallon of water, do not place the liquid in a jug away in the cupboard for twenty days, because then it would turn into wine."

I have seen this type of messages a few times on growkits.


Pretty much. Conda installed with miniforge uses a different channel, with a focus on having packages available for more architectures.


> A performant server should be able to do, on 10 GbE:

> - 6M/s ~4KiB text/plain HTTP/1.1

According to my numbers 4KiB * 6M/s is roughly 196Gbps, that really must be a very performant server, and network card.


255Tbps exist in research labs. 100 Gbps is COTS hardware.

Benchmarks are usually performed locally between a load test program and the local server. Bottlenecks are imposed at the boundaries of storage iops, memory bandwidth, cache bandwidth, cpu context switching, external io, cache miss/hits, and overhead of the program.

Anything production and performant likely leverages DPDK and multiple 40/100 Gbps nics/hcas but doesn't need to serve that much because it has CDNs and "cloudflares" in front of it. Apps with dynamic data have to be sharded in such a way that they can be horizontally-scaled. WhatsApp scaled nicely with Erlang BEAM/HiPE on FreeBSD on a low hardware budget for the volume of traffic. HiPE isn't particularly fast or efficient compared to Rust or Azul C4 JVM but it saved a lot of thread contention by making copies and saving cache invalidations by making almost everything immutable.


You currently get 239 apps with the setapp bundle, I don't think there's anything special about that relationship.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search:

HN For You