hootz

5 ms·

61.
by hootz 4mo ago
For me, it has a positive connotation! In my experience, Chinese Model means cheaper, but still quite effective model you can use for millions of tokens without burning your entire wallet in seconds. That's why I get more excited over
62.
by hootz 4mo ago
I do it too, but I can see why this can be a problem for users. There should be an "official" scan for potentially malicious changes. I use a third party AUR scanner to help me with this.
63.
by hootz 4mo ago
Oh yeah, that's the name of it. But I guess something similar happens with checklists, you do it so many times without anything bad ever appearing that you start to subconsciously assume nothing will ever happen. Why check the rotor of
64.
by hootz 4mo ago
There are some AUR hooks that can help. I use https://github.com/Sohimaster/traur which also has scans for orphan package takeover patterns.
65.
by hootz 4mo ago
For every single update, for all your AUR packages, all the time. You know that thing where if you make a security review feature obnoxious, after some time people will just accept everything without even looking? Yeah...
66.
by hootz 4mo ago
Arch is not used in enterprise solutions because of the AUR? Can't you just not use it?
67.
by hootz 4mo ago
Making a modern analogy, reading this feels kinda similar to reading about the Backrooms, but with a bigger, existential dread. Amazing.
68.
by hootz 4mo ago
What is "EA" in this context? I see a lot of people using this initialism.
69.
by hootz 4mo ago
Like Chaplin in Modern Times, we will tighten screws until we lose our minds.
70.
by hootz 4mo ago
It's how American companies have always worked, of course it is monopolistic practice, but those things are rarely illegal because the US absolutely loves their corps. Look at Google, Microsoft and the likes, this is the norm.
71.
by hootz 4mo ago
But it seems like that was not genuine concern, but instead a tactic to pivot to closed models and an API service with an excuse to do so, breaking the public's expectation that they would be a non-profit making open models, like their
72.
by hootz 4mo ago
They are not artificially cheap, they are still cheap even when hosted by independent inference providers. Are all providers subsidizing their open-weight models?
73.
by hootz 4mo ago
I still remember it. "Open"AI going API-only because GPT-3 is really really dangerous, so forget the Open in our name and all of that, you can't download our models anymore and must request access to them because they pose a
74.
by hootz 4mo ago
That doesn't match my experience at all. I can't see myself saying in 6 months that the current model I am using is useless, that makes no sense. In fact, I did go back to DeepSeek V4 Flash for most of my problems as it is way che
75.
by hootz 4mo ago
People can work around those if they are open-weight.
76.
by hootz 4mo ago
I guess either that or closing the bug bounty program, but I still believe closing it is worse than automated triage, even though both suck.
77.
by hootz 4mo ago
AI reviewing AI submitted bug bounties. We have reached the dead bug bounty program theory.
78.
by hootz 4mo ago
You are only priced out if you only care for SOTA right now and can't wait for the inevitable cheap model coming in 6 months. DeepSeek, Xiaomi and Moonshot are already really cheap and match frontier performance from 6 months ago.
79.
by hootz 4mo ago
>Opus is too small, increase the impact of the name. Okay, how about Mythos? >Increase it even more. Right, then Cosmos. >Even more! Even more? Let's try Aeon. >MORE, EVEN BIGGER ALRIGHT, TRY OMEGAPANTHEON 7.8 THEN
80.
by hootz 4mo ago
My bet is that Mythos is still over-hyped and the cybersecurity fear and guardrails are mostly marketing to force company partnerships through Glasswing and get public attention.
81.
by hootz 4mo ago
>The reports I get from people I know are that my company is not on the extreme edge of vibecoding, so leaving it for a potentially worse environment is not a good trade. I'm in a similar situation right now. I might get a job offer
82.
by hootz 4mo ago
This is really good! Thank you for this skill, it seems like this method of learning is working really well for me, it is much more engaging, and I have just learned new things about my team's project by using it. I'll definitely
83.
by hootz 4mo ago
The difficult part is to also be careful as to not become one of those YouTube "stoics" who just stay in "I don't care" state for everything. We can't control other people, but we can try to influence them towa
84.
by hootz 4mo ago
Have you tried the open weight models, but not locally? Like, using it from a provider. That way, you get access to better models while still not using private closed models, anyone with enough compute can host them, not just the big AI cor
85.
by hootz 4mo ago
Yeah, feels kinda weird to think about using a mouse pointer utility with licensing DRM.
86.
by hootz 4mo ago
But that is Mac only.
87.
by hootz 4mo ago
The Task Manager that many times also stopped responding or was permanently stuck behind a frozen full screen game? On Linux, KDE has the plasma System Monitor, and if you can't use it for some reason you can switch your entire session
88.
by hootz 4mo ago
First, you need a skiddie neighbor who knows about your speaker and has an AI agent that can read this article, 2nd...
89.
by hootz 4mo ago
Oh yeah, for some reason the companies with the highest risk products seem to be the ones that care less about security. Don't even get me started with "smart" bulbs and cameras that each individually connect to your local ne
90.
by hootz 4mo ago
>Email from SingCERT stating vendor "do not consider this to be a vulnerability, as it does not present a cybersecurity risk." So wirelessly writing custom firmware to someone else's device that is connected via USB to the

More ›