eager_eagle
@eager_eagle@lemmy.world
- Comment on French aircraft carrier Charles de Gaulle tracked via Strava activity in OPSEC failure 2 days ago:
you can’t compare it to the islands, because the GPS trace is an area within that green circle. You can only look at the 300m scale in the bottom right, which looks in the ballpark of an aircraft carrier to me
- Comment on Our commitment to Windows quality 2 days ago:
so… some really basic shit that should have been expected in a pre-2010 update + AI
well done, guys
- Comment on French aircraft carrier Charles de Gaulle tracked via Strava activity in OPSEC failure 2 days ago:
You don’t need that assumption. Your assumption can just be “the person and vessel don’t diverge significantly over time”.
Then, if you treat velocity as a vector and compute the average velocity vector over time, you’ll have a pretty close estimation to the vessel’s velocity vector.
After all, if those two average vectors (vessel’s and person’s) were to differ much, they would end up in different locations.
- Comment on I built a local AI movie recommender for Radarr using Ollama 2 days ago:
It’s not, I read the code. It’s not asking the LLM for recommendations, it’s using embeddings to compute scores based on similarities.
- Comment on Astral to Join OpenAI 2 days ago:
Ok, I’m not suggesting replacing humans with AI and I despise companies trying to do this unsustainable practice.
With that out of the way, I’ll restate that LLMs follow some rules more reliably than humans today. It’s also easier to give feedback when you don’t have to worry about coming across as a pedantic prick for pointing out the smaller things.
On your point that LLMs are not improving; well, agents and tooling are definitely improving. 6 months ago I would need to babysit an agent to implement a moderately complex feature that touches a handful of files. Nowadays, not as much. It might get some things wrong, but usually because it lacks context rather than ability. They can write tests, run them, and iterate until it passes, then I can just look at the diff to make sure the tests and solution makes sense. Again, something that would fail to yield decent results just in the last year.
- Comment on I built a local AI movie recommender for Radarr using Ollama 2 days ago:
No, it doesn’t do that. It gets embeddings from an LLM and uses that to rank candidates.
- Comment on French aircraft carrier Charles de Gaulle tracked via Strava activity in OPSEC failure 2 days ago:
that’s how it looks when you walk in circles in a vessel that’s moving
- Comment on I built a local AI movie recommender for Radarr using Ollama 2 days ago:
no one is saying everyone has to ask an LLM for movie recommendations
- Comment on Google Search is now using AI to replace headlines 2 days ago:
wow, this dystopic update is reality not disappointing
yet another reason to not use Google
- Comment on I built a local AI movie recommender for Radarr using Ollama 2 days ago:
chill, this is extracting text embeddings from a local model, not generating feature-length films
- Comment on I built a local AI movie recommender for Radarr using Ollama 2 days ago:
There’s no training, the LLM embeddings are used to compare the plots via a cosine similarity, then a simple weighted score with other data sources is used to rank. There’s no training, evaluation, or ground-truth, it’s just a simple tool to start using.
- Comment on I built a local AI movie recommender for Radarr using Ollama 2 days ago:
that’s pretty cool, this is just the wrong crowd, don’t worry about the downvotes
- Comment on Google gives Android users a way to install unverified apps if they prove they really, really want to 3 days ago:
- Comment on Astral to Join OpenAI 3 days ago:
It’s manageable if you pass along these rules to the LLM. I’ve actually had more success asking the LLM than giving code reviews to some interns, and even someone who coded professionally before.
- Comment on France bids to suspend Shein over childlike sex dolls 3 days ago:
I saw your comments in other threads, and you don’t sound happy at all, but whatever. Have a nice life.
- Comment on Trump Wants to Put You in a Massive, Secret Government Database 4 days ago:
I wish it was just trump. Remember Snowden in 2013? I can only imagine the level of surveillance they have nowadays.
- Comment on France bids to suspend Shein over childlike sex dolls 4 days ago:
It’ll never be normalized because it’s fucking weird, whether that’s legal or not. And you can stop calling anyone with a different opinion a bot, it makes you sound like one.
- Comment on France bids to suspend Shein over childlike sex dolls 4 days ago:
ah, the point is that it’s disgusting, thanks for confirming the emptiness of the argument
- Comment on Firefox 149 adds built-in free VPN with 50GB monthly data 4 days ago:
idk about you, but when I use a VPN I want all the device’s traffic to go through it, not just the browser. So I’d always disable the built-in feature in that case.
- Comment on France bids to suspend Shein over childlike sex dolls 4 days ago:
lol they’re not arguing against because it’s a pointless measure
you’re confusing picking your fights with being supportive
- Comment on Firefox 149 adds built-in free VPN with 50GB monthly data 4 days ago:
ofc, this is a free alternative, not a mullvad competitor
- Comment on Firefox 149 adds built-in free VPN with 50GB monthly data 4 days ago:
with a 50GB quota, I actually believe it’s free
- Comment on France bids to suspend Shein over childlike sex dolls 4 days ago:
“that look like children” is not enforceable. Are you requiring a minimum height? A minimum number of ageing features? A certain breast size? What about cartoon/anime stylized products?
But the core issue is that this literally won’t solve anything and it’s a waste of time and public money.
- Comment on France bids to suspend Shein over childlike sex dolls 4 days ago:
and it finally makes sense, because I don’t think the average billionaire spends money on garbage as much as their average customer
- Comment on France bids to suspend Shein over childlike sex dolls 4 days ago:
I don’t think that banning this kind of product will solve the fact that there are people with a disorder. So what’s the point? Moralism?
- Comment on AI is making us think and write more alike: Large language models are standardizing human expression — and subtly influencing how we think 6 days ago:
so is social media [citation needed]
- Comment on Gryt-Chat: Like Discord but Self Hosted 1 week ago:
“Jonathan”
- Comment on Instagram quietly drops end-to-end encrypted chats 1 week ago:
bro, I swear I won’t read it bro, you can trust me bro - yo
- Comment on Instagram quietly drops end-to-end encrypted chats 1 week ago:
well, they can have true E2EE and still be able to read or exfiltrate the messages, because they control both ends…
- Comment on Instagram quietly drops end-to-end encrypted chats 1 week ago:
shocking
