If Microsoft leadership thinks that they took a wrong turn in the last few years, they are in for a rude awakening (I hope). They took a wrong turn in the late 90s. Probably earlier than that. They have managed to stave off all user feedback for 30 years through litigious bullying and strict vendor lock-in. This isn't about the taskbar, man. This isn't even about Copilot.
The pushback which you are only now starting to perceive is being caused by an entire generation of Microsoft intentionally and actively positioning itself in conflict with its customers.
I understand that once you have a million customers, you can't really treat them right anymore. But Microsoft has not given a single shit about customer feedback, even in aggregate for decades now.
As I read this, all I can think is "too little, too late." I have watched in my workplace Windows go from being a product that we are happy to purchase to yet another piece of technology that we would simply replace were we not yoked to it.
I guess even now they probably still don't care. Microsoft will continue printing money until the sun burns out.
So the chefs are preparing food that has the same macros as ultraprocessed meals (I assume like tv dinners or something?) Why do they keep referring to the freshly-prepared food as "ultraprocessed"?
“Is this processed or unprocessed?” I asked.
Kozlosky smiled. “Ultra-processed,” she said. “Lots of participants can’t tell the difference.”
If the term has any meaning, you could tell very easily. Go look at a freshly fried tortilla chip, and compare it to a tostito. You know which one is which instinctively.
I thought I understood the study but now I'm not sure. I thought the idea was to take the exact same thing you'd get in a tv dinner and make it fresh, so no freeze drying, no preservatives, etc. Then if that food on its own causes the same pattern of health issues, we know it's simply a diet problem. It sounds like they replicated that effect. So they got evidence that ultraprocessing doesn't actually matter all that much?
A ton of vector math applications these days are high dimensional vector spaces. A good example of that for arm would I guess be something like fingerprint or face id.
Also, it doesn't just speed up vector math. Compilers these days with knowledge of these extensions can auto-vectorize your code, so it has the potential to speed up every for-loop you write.
> A good example of that for arm would I guess be something like fingerprint or face id.
So operations that are not performance critical and are needed once or twice every hour? Are you sure you don't want to include a dedicated cluster of RTX 6090 Ti GPUs to speed them up?
I'd argue that those are actually very performance critical because if it takes 5 seconds to unlock your phone, you're going to get a new phone.
The point is taken, though, that seemingly the performance is fine as it is for these applications. My point was only that you don't need to be running state of the art LLMs to be using vector math with more than 4 dimensions.
The entire quoted section in the middle adds nothing. It just keeps repeating the same things over and over, and it doesn't answer the question of how we know the offset at all. Makes me think his "friend" is an LLM.
Just based on this blog post it seems like he wanted the project to be more “professional” in some way that the rest of the developer group didn’t. I wonder if that difference in vision, combined with a (probably justified based on your comment) feeling that he was doing a disproportionate amount of the work lead to an unsustainable situation.
Calling it a post-mortem while others are continuing the project still seems kind of petty, though.
Doesn't look like he was, but then looking at the actual commit list: https://github.com/ublue-os/bazzite/commits/main/?before=e49...
He definitely had more than the 14 commits listed. They might've lost their email due to the conflicts & lost ownership over the commits?
Software history is rife with projects that outlive a person like that leaving, though. Ulrich Drepper comes to mind immediately. They don't own the project.
I can make corn too. I go to the supermarket and hand them these little green pieces of paper, and then I have corn.
Seriously, what does this prove? The AI isn't actually doing anything, it's just online shopping basically. You're just going to end up paying grocery store prices for agricultural quantities of corn.
reply