Not entirely unlike Jill Biden telling Joe he’d done a good job, when clearly the opposite had happened.
Not entirely unlike Jill Biden telling Joe he’d done a good job, when clearly the opposite had happened.
I hope Harris doesn’t over prepare, she does need to maintain a certain level of off the cuff affability.
Over preparation was Biden’s fatal error, he was struggling so hard to hit his talking points he couldn’t keep things straight. Obviously things are vastly different with Harris, but she’d be best to not just be a talking point machine.
What conspiracy theory? The Saudis and Qataris have stakes in Twitter.
Qatar Holding, a sovereign wealth fund, is contributing $375m, while Saudi Arabian investor Prince Alwaleed bin Talal, who had initially opposed the buyout, also confirmed he would retain his $1.9bn stake in Twitter, writing that Mr Musk would be an “excellent leader” for the site.
Maybe the Saudis and Qataris will chip in some more cash to help him out.
The show has real Late Late Breakfast Show vibes. An 80’s BBC show where the public took part in more and more over the top stunts. In the end someone died and the show was cancelled 3 days later.
https://www.everything80spodcast.com/the-late-late-breakfast-show-tragedy-of-1986/
They’ve committed to support AM5 (the LGA socket launched 2022) through at least 2027.
Not bullshit. They’re not talking about the heckle they’re talking about the 10 seconds of chanting at 7:18. Watch again. https://youtu.be/2soe8ml_weg?si=uJqnlGGNU4SF8im_
“We envision other types of more complex guardrails should exist in the future, especially for agentic use cases, e.g., the modern Internet is loaded with safeguards that range from web browsers that detect unsafe websites to ML-based spam classifiers for phishing attempts,” the research paper says.
The thing is folks know how the safeguards for the ‘modern internet’ actually work and are generally straightforward code. Where as LLMs are kinda the opposite, some mathematical model that spews out answers. Product managers thinking it can be corralled to behave in a specific, incorruptible way, I suspect will be disappointed.
Yeah, most 3rd person games I like to play with a controller, first person not so much.
I remember the ‘good old days’ of Sun Fire 10k and similar servers. You could replace entire boards of CPU and RAM and the server would keep on trucking.
43% of Google traffic is now ipv6 and steadily growing
https://www.google.com/intl/en/ipv6/statistics.html
CGNAT is only a temporary band aid for reaching services that are yet to present themselves on IPV6. It’s relatively expensive to operate.
IpV6 might be largely pointless on a LAN, and sure NAT is fine there, but ipv6 already running large chunks of the world’s mobile infrastructure. It’s not going anywhere.
One thing the EU got right was reducing interchange fees to 0.5%. The ridiculous situation in the US where airlines have become credit card companies that happen to have planes is madness.
Folks might like kickbacks but they’re paying for them anyway, it’s just hidden in the price and subsidised by folks who don’t use these cards.
The US should follow the EU’s lead here.
There are no M1 devices with less than 8GB of RAM.
The A16 Bionic has as Neural Engine capable of 17 TOPS but 6GB of RAM.
The M1 had a Neural Engine capable of just 11 TOPS but all M1 chips have at least 8GB of RAM.
So the model could run on an A16 Bionic if it had 8GB of RAM as it has 54% more TOPS than the M1, but it only has 6GB of RAM. Apple have clearly decided that a model small enough to fit just wouldn’t give good enough results.
Maybe as research progresses they’ll find a way to make it work with a model with fewer parameters but I’m not going to hold my breath.
Yeah I thought it was a NPU tops issue that’s keeping it off the 17 non pro. However since it runs on a M1 I think it’s more to do with needing 8GB RAM to fit the model.
He called the software integration between the two companies “an unacceptable security violation,” and said Apple has “no clue what’s actually going on.”
I’d be very surprised if corporates wouldn’t just be able to disable it in MDM for their worker’s phones. Not sure it’s Apple who has ‘no clue’ here.
If they keep burning $100k/w on their Vercel bill they might not be around that long anyway!
The thing with serverless is you’re paying for iowait. In a regular server, like an EC2 or Fargate instance, when one thread is waiting for a reply from a disk or network operation the server can do something else. With serverless you only have one thread so you’re paying for this time even though it’s not actually using any CPU.
While you’re paying for that time you can bet that CPU thread is busy servicing some other customer and also charging them.
I like serverless for it’s general reliability, it’s one less thing to worry about, and it is cheap when you start out thanks to generous free tiers, at scale it’s a more complex answer as whether it is good value or not.
It’s called preparation you bellend.