While I agree it is pretty cool, I think they probably did that to avoid a class action lawsuit.
But that does also beg the question, if we do eventually regulate this better, do you own streamed products? IE with Google stadia (and I think Amazon Luna) you buy a game but the game data is streamed to your device. If they shut the service down (like Stadia did) then what happens to your game collection?
It's a hypothetical, but interesting to think about nonetheless
How many Kubernetes administration headaches trace back to the need for automated systems to surgically edit YAML? It’s absurd and YAML may be the worst choice for this use case.
Not sure where in the Nordics you are, but just want to plug ut.no and norgeskart.no for really fantastic Norway navigation.
Both have free apps (norgeskart has a 50kr / year subscription to support and add more features), and both are incredible for navigating trails, marked and otherwise, anywhere Norway.
Yeah...using anything other than an Apple Approved(tm) scsi disk (or, really, anything 'not Apple') was an adventure. One of my least favorite Unixen, frankly. SVR2 based when most of the world on the AT&T side of Unix was SVR3 (and SVR4 was just around the corner), some BSD bits tossed into the blender, coated in a not-very-good MacOS emulation layer. Congrats to you on your perseverance; I recall the TCP/IP stack was...challenging.
A/UX was created so Apple could bid on government contracts that had a "must be Unix and/or POSIX" checkbox (common requirement at the time). IMO, they should have either gone all-in on a "MacOS apps running perfect on a Unix kernel" and picked a better kernel, or done the Atari thing and tossed the legacy OS and gone all in on a "Unix on Apple hardware". Instead it was 'meh' all around.
In the UK, when I reported a crime, I had to pull out the book and read the verse what constitutes an assault for Police to record it, because they didn't know. That said, 3 weeks later case was closed because they couldn't find the perpetrator. That is despite me giving them video evidence and address of that person. Looks like our taxes are well spent.
I live next to a railway line so I'm in the (not particularly unique, and definitely not enviable) position to compare what's on the map to "IRL" trains, and I can tell you it's as good as useless.
- Trains appearing on the map that aren't anywhere to be seen on the tracks.
- Trains on the tracks that don't appear on the map.
- Trains moving away from the station that according to the timetable view shouldn't have left the station yet.
- Trains on the map seemingly stopping and changing direction, only to reverse course once again.
The map shows a single line segment for what is in fact a multi-line stretch of railway. That's okay as a simplification (I guess), but the icons aren't pinned to the line, so appear to be driving off the track, or even on the adjacent street.
As for realtime - even if the data was accurate and timely, a 2Hz refresh rate most definitely isn't realtime.
Sorry if it seems like I'm shitting on it - it's a fun toy, but I wouldn't depend on it for anything important.
The evidence is that unlike Deepseek he does not publish his compute multipliers. Under that argument Deepseek should not publish any of their research either.
I think you’re a troll. We are talking about companies with money that makes countries blush. Let’s not pretend these things exist in a vacuum. Be serious.
In your case, if you have a goal and achieve 10% of it, you have failed the goal. Anything you have gained along the way is irrelevant.
The twenty something guys in my neighborhood set off fireworks from a small boat in the river behind my house. Same guys took an ATV out on the ice when the river froze this winter. They haven’t died yet!
I draw the line at setting the tinderbox that’s california on fire however.
Not sure, I would say it overstates things anyway (it's Twitter, so exaggeration for engagement is par for the course) ... but I would say it's gesturing at something roughly true and it'd be worth discussing the idea.
I’m curious why many tech bloggers choose Substack, when it’s almost trivial to set up your own blog with infinitely more style control, and math/syntax-highlighting etc.
Show dead and you see 23 posts about this exact same thing which gives me pause and makes me wonder, did this guy really post this 23 times?
Why? Or is there some program setup on a loop?
Either way,a legit Linux shell in a browser would be kinda powerful and even if this one isn't legit, Im intrested in adjacent projects...for instance this may be the only way to get a shell in a usable shell on iphone which would be neat and useful
One hardly needs to soar to the exalted heights of "mysticism" or any sort of "ineffable qualities" to come into contact with the limitations of LLMs.
1) Models do not perform the functional equivalent of reasoning at all. When we reason, we don't simply babble out textual derivations of prior examples of "reasoning" to which we have been exposed, arrive at a conclusion, then occasionally state an altogether different conclusion while pointing at the largely irrelevant reasoning to substantiate it.
2) Models have real-world, not-at-all-mystical functional constraints that are directly relevant to the production of everyday economic work in which one attempts to involve them. Their inability to extrapolate or maintain clear mental models leads to staggering, head-scratching mistakes that even a very feeble and developmentally awry human intelligence would not make.
A basic, if well-worn example that was widely discussed in the last year or two:
However, this is emblematic of a much larger idea: the LLM doesn't have any idea what a letter is or what you're asking it to do. This isn't a question of "ineffable qualia"; when it doesn't know what something is at any essential level, it can't competently solve problems related to it. One bumps up against this in everyday programming and all the time.
Also, what is "mystical" about my demand for the kind of scientific progress--no, forget that, any scientific progress--that a functional superhuman intelligence would yield? I am not a "mystic", either; I want functional results, show me the functionality.
$1/day forever is around $9125 using the safe drawdown rate of 4% per year.
So if you’re a bakery and a customer offers $9000+ for a fresh roll every day forever, you should almost certainly take them up on that offer. A smart baker could probably get that number sub-$5000, but you’ll always come out okay around 25x the yearly cost (in this case, $365).
Similarly, if the amortized yearly cost of a customer is $12 (ie, $1 per month), then a $300 forever price is financially indistinguishable from a permanent subscription. (Actually, better: time value of money, they can’t cancel the annuity you buy, etc.)
So there always is a price where that is financially viable.
