ekky
@ekky@sopuli.xyz
- Comment on OpenAI and Microsoft are funding $10 million in grants for AI-powered journalism 3 weeks ago:
You thought journalism had reached rock bottom already? Watch this:
- Comment on The popularity of Minecraft has probably led some children to believe you can swim up waterfalls 1 month ago:
Dunno about kids, but I’ve seen my fair share of grown men who appear to think so.
Ladders tend to be more stable if you lean them on the tree trunk, and not the branch you’re about to saw off.
- Comment on How do our brains process reality? I heard our eyes were just low-res cameras and our brains were doing all the heavy lifting in 'rendering' reality. 1 month ago:
I think i read that fighter pilots need to be able to identify a plane in one frame at 300 fps, and that the theoretical limit of the eye is 1000+ fps.
Though, whether the brain can manage to process the data at 1000+ fps is questionable.
- Comment on u can just do stuff 1 month ago:
Hurr hurr, I’m gonna plot f(x,y)=x^2+y^3 where y=x for x limit inf. Checkmate science!
- Comment on Elements of Renewable Energy 2 months ago:
“Batteries” is a rather broad category.
Are we talking hydroelectric batteries? Other potential or kinetic batteries? Chemical batteries (and what subcategory)? Or maybe hydrogen-based power storages?
Since there’s a dam on the list, I’d imagine “batteries” to be electrolytic power stores or hydrogen fuel cells, but the visualization remains lazy and perhaps borderline misinformative.
- Comment on When A.I.’s Output Is a Threat to A.I. Itself | As A.I.-generated data becomes harder to detect, it’s increasingly likely to be ingested by future A.I., leading to worse results. 2 months ago:
You don’t have to sanitize the weights, you have to sanitize the data you use to get the weights. Two very different things, and while I agree that sanitizing a LLM after training is close to impossible, sanitizing the data you give it is much, much easier.
- Comment on When A.I.’s Output Is a Threat to A.I. Itself | As A.I.-generated data becomes harder to detect, it’s increasingly likely to be ingested by future A.I., leading to worse results. 2 months ago:
Oh no, it’s very difficult, especially on the scale of LLMs.
That said, we others (those of us who have any amount of respect towards ourselves, our craft, and our fellow human) have been sourcing our data carefully since way before NNs, such as asking the relevant authority for it (ex. asking the post house for images of handwritten destinations).
Is this slow and cumbersome? Oh yes. But it delays the need for over-restrictive laws, just like with RC crafts before drones. And by extension, it allows those who could not source the material they needed through conventional means, or those small new startups with no idea what they were doing, to skim the gray border and still get a small and hopefully usable dataset.
And now, someone had the grand idea to not only scour and scavenge the whole internet with no abandon, but also boast about it. So now everyone gets punished.
At last: don’t get me wrong, laws are good (duh), but less restrictive or incomplete laws can be nice as long as everyone respects each other. I’m excited to see what the future brings in this regard, but I hate the idea that those who facilitated this change likely are the only ones to go free.
- Comment on When A.I.’s Output Is a Threat to A.I. Itself | As A.I.-generated data becomes harder to detect, it’s increasingly likely to be ingested by future A.I., leading to worse results. 2 months ago:
So now LLM makers actually have to sanitize their datasets? The horror…
- Comment on Microsoft breaks some Linux dual-boots in a recent Windows update 2 months ago:
Huh, is Windows screwing over GRUB and Linux not a bi-monthly experience anymore? Sad that it happened, but glad that it’s become novelty enough to write about.
- Comment on Corn 🌽 2 months ago:
We get to choose the genes when genetically modifying, and it usually takes a few years (plus health metrics and research once complete).
Contrary, when selectively breeding we can breed for traits which we are not guaranteed to actually get, and it takes a few decades (plus health metrics and research once complete).
- Comment on :D 4 months ago:
You can fix that by setting “User Settings->APP SETTINGS->Chat->Automatically convert emoticons in your messages to emoji” to “OFF”
You can still make emojis with “:smile:”, but why would you? ¯\_(ツ)_/¯
- Comment on I think it's extremely invasive that amazon is telling me this 6 months ago:
In what kind of backward and lawless war zone does one have to live to require telling people not to shoot at other people?
- Comment on Stack Overflow bans users en masse for rebelling against OpenAI partnership — users banned for deleting answers to prevent them being used to train ChatGPT 6 months ago:
And then links to a similar sounding but ultimately totally unrelated site.
- Comment on YSK : Dark patterns among large companies are becoming more mainstream 6 months ago:
To expand a little:
While it indeed is annoying, it did mostly go as expected, as in, law makers must always be ready for companies responding to new and more restrictive laws with malicious compliance.
The vast majority of websites don’t actually follow the rules for cookie banners or implement them in as roundabout a way as possible, making them needlessly annoying as it should always be easier and at least as fast to decline than to accept.
While this all sounds like cookie banners ultimately are a failure because of the misimplementations that companies provided in response, it does function as an eye opener for the common man and stepping stone for the EU for further laws and fines in regard to citizens’ rights to privacy.
- Comment on Caption this. 7 months ago:
Multi-target laser guided auto-lock eye upgrade for sports now available at your local augmentation center. Ads included!
- Comment on Discord Quests have started (ads) 7 months ago:
That does make way more sense. Though, the ad makes it seem like a (clothing) set and not a skin, but I’ve never played the finals so I wouldn’t know.
- Comment on Discord Quests have started (ads) 7 months ago:
So the way that promotion is worded, if I stream “the finals” to a friend for 15 minutes I’m guaranteed to receive that item, shipping paid for and all, like in a quest? Or are they just baiting and entering you in a lottery?
Seems like an easy way to farm real life merchandise.
- Comment on Microsoft in their infinite wisdom has replaced the Hide Desktop icon with Copilot. 8 months ago:
Huh? Isn’t this about Microsoft changing out a button with a well established use, in order to take advantage of muscle memory and the unobservant?
Don’t think it’s much to do with people opposing technological advancement, but rather with opposing another company wanting to making a fool of them.
- Comment on European Court of Human Rights bans weakening of secure end-to-end encryption 8 months ago:
Criminals will always find a way. Make a surveillance state, and they’ll just break the law and use encrypted communication anyway. Might even hide data in other data if necessary.
That said, I’d wager that there are quite a few of those communities hidden in plain and unencrypted sight (discord, fediverse, etc.), but they just keep it small enough to not be found (The ones on discord did get found out eventually, but probably just moved platform). So the question would aris: why do these exist when we apparently have the resources to monitor EVERYONE given the chance?
Best you can do is to report communities and places where it runs rampant to the relevant authorities. That’s much more efficient than the authorities having to make privacy-violating laws and crawl the net themselves.
- Comment on dont question the big orb 9 months ago:
Right, thank you for the heads-up. :)
- Comment on dont question the big orb 9 months ago:
Sight is just the price we pay for the eldritch knowledge* staring at the sun gives us.
- Should not have stared at the sun.
- Comment on FreedomBox is amazing! 9 months ago:
Haven’t tried it, but it does look interesting and worth checking out.
I have been playing around with dietpi lately, and while it feels weird to relinquish control, it does make a lot of things much more easier and faster to setup and maintain. Or at least until you need more specific setups. :P
- Comment on Could you stay in the roundabout indefinitely? 9 months ago:
Depending on the definition, you’re not obscuring traffic if you’re driving 60kmh in the fastlane either, or if only you manage to get over the green light because of your tardiness, as traffic is still flowing, just very, very slow.
Either way, you’ll be slowing down traffic and get in the way of other drivers, unless you are the only person on the road.
- Comment on Could you stay in the roundabout indefinitely? 9 months ago:
Where do you live, OP?
In Denmark it goes in the same category as not driving when the light is green or keeping to the left on the highway.
While you seldomly see someone get fined for it, taking more than two turns in the roundabout is considered an obstruction of traffic, and therefore illegal.
- Comment on The EU common charger : USB-C 10 months ago:
Green, blue, and brown are already compatible (except ground), though I’m not sure about yellow and red.
- Comment on The EU common charger : USB-C 10 months ago:
I’m do not understand why it would not be able to charge with a X to USB-C.
As far as I’m aware, even if one of the two connected devices (such as charger and flashlight) doesn’t have a proper chip for agreeing on voltage, it will just default to 5v. The only thing that should prevent that is a damaged (as in no + or - connection) cable. Even if the data connections are damaged, it should still charge in slow-mode.
Needless to say, I have never experienced that problem even with cheap china cables/chargers, except, of course, for damaged cables.
- Comment on what is the fedipact? 10 months ago:
Not relevant to the question, but the overuse of hearts makes it seem like they play more on feelings than on logic. Kinda feels like a cult.
- Comment on They forgot the LGBTQ... 11 months ago:
The rich Americans are seldomly native Amaricans, so the meme is right about the immigrants holding all the money.
- Comment on What do we think of greenpeace activists? 11 months ago:
Right, I thought it was a group of employees doing this over an extended period of time, might not be a fair comparison after all then.
Not that it’s a fair comparison to begin with, I just held a similar amount of annoyance toward the two.
- Comment on What do we think of greenpeace activists? 11 months ago:
Can’t talk for anyone else, but I put them in the same category as PETA.
They probably do a lot of good, but it feels like the good they do is outweighed by all the bad they do to get there.
Like kidnapping collared dogs from the streets to euthanize them (PETA), or fighting the (at the time) only realistic alternative to oil, giving all the rich oil sheiks a hard on and adding to the already bad global warming problem (Greenpeace).