fonix232

The LLMs in question aren’t providing data from their training set, but are transforming live data retrieved from the internet. So their date is quite irrelevant, what matters is their ability of contextual data filtering and transformation.

LLMs are great when they work well. Problem is, they hallucinate a lot.

For example I was just trying to research if/how I could stay and work at a nearby airport - I need to leave my Airbnb by 10am but my flight is at 7pm, so I’m thinking of heading right to the airport and just working from there.

Gemini told me that at this airport there’s numerous landside cafés and work pods available.

Perplexity said for sure there will be spots I can work from.

Both were incredibly wrong as they collated information from airside - even though I specifically asked for landside as the airline I’m flying with doesn’t offer early luggage dropoff, so until ~4pm I’m stuck landside.

guess what there is landside? a single cafe with about 10 seats…

Imagine pouring all your effort into a company for years only to be fired because management mismanaged everything…

interestingly it’s NEVER those who made the bad decisions who get replaced or made redundant.

An mSATA or small NVMe in a USB enclosure is the best way to go. No wasting a precious internal slot, and they can be super tiny (2230/2242 enclosures are dime a dozen).

Even if you eliminate personal cars, car infrastructure is still a necessity in most cities - public transport depends on it, workers depend on it (or do you see tradies lugging their shit around on a bus?), deliveries depend on it (let it be your Amazon orders, shops’ deliveries, construction materials and so on).

I agree that we have too many cars on our streets regardless where you live, but uprooting entire existing infrastructure that is relied upon by more than just your target group isn’t going to help the situation.

Maybe in a year or so. My Deck still works fine, not in a rush to replace it. The Frame on the other hand, probably a day one purchase.

Eh, the writing was on the wall the moment DRAM manufacturers refused to expand capacity and instead stopped consumer sales in favour of corporate batch sales.

No such discussion happened. As per above, that group of people were super toxic and couldn’t care less about this specific person…

At most the remaining people would’ve mentioned how his panic attack inconvenienced their night out.

Except as I said NOBODY looked up the term. How many times and in how many forms do I have to state that the MENTION of pulse oximeters was: verbal only, in no way entered into a computer before I started getting the adverts?

The discussion itself wasn’t even that long, colleague asked why I looked so tired, I told him how the previous night we had to call an ambulance and sort the guy out, colleague said “maybe he should get a cheap pulse oximeter, it can be helpful to predict attacks and handle them without a call [for an ambulance]”, and that was it. Topic was dropped.

And no, mutual friends didn’t look it up either as they didn’t care much for the guy (tbf it was a super toxic community I’m glad I’ve left behind).

Nope, as only ~4 people remained at that point, and they wouldn’t even be thinking of researching this topic.

NONE of that data can predict a random occurrence discussion that goes in a specific direction.

A great example is something that happened to me in 2015. One night I was out with friends, and one of them had a really bad panic attack. The next day I was discussing it with a colleague during a smoke break, who recommended he gets a clip-on pulse oximeter. No searches, nothing, literally just a half minute detour in our chat. I repeat, nothing was typed in or looked up or in any way entered into any computer intentionally.

Five minutes later we’re sitting in front of our respective computers and I start getting ads for the very thing. Mind you, we’re still at a point where nothing noted during this discussion was entered into any computer. Explain this.

That’s been happening for well over a decade now, and while “respectable security researchers” call it bullshit… there’s simply too much anecdotal evidence for it to happen organically.

Shitty, old processors? In which way?

Zen 4 is literally just a single generation behind current latest gen architecture. And you’re way off on the pricing too - Zen5 APUs are essentially the AI 300/400 lineup, of which the higher end models still cost well over what Valve would find affordable. Meanwhile the GPU Valve chose to be integrated into the SM is 30-40% more performant than the 890M bundled with the Ryzen AI 370 (the only affordable kinda-high-end Zen5 APU).

So no, it’s neither old nor shitty.

I’d argue that this shall only apply to games that are being made by big, established studios, under big, established labels.

self-published indie games, you should preorder if you can.

Not remembering specifics of a technology, and completely forgetting the base building blocks - the same blocks that you should be using for AI generated code too, BECAUSE YOU NEED TO FUCKING REVIEW IT - is not the same.

I’m an Android engineer by trade. I might not be able to give you the exact interface definition of a BroadcastReceiver, or explain in technical terms the core differences between a TextureView and a SurfaceView (that’s what the documentation is for!), but for sure as hell can tell you if your architecture is good or not, or if the quality of the code you wrote is shite.

If you forgot to write good code and good architecture, in a span of what, at most, a year, then you were never a good engineer to begin with.