News for hackers. Stuff that matters.

kansface

It’s more likely emergent behavior.
Re:CPI-U rises 7.2% annualized in September (3.7% YoY) on Wednesday October 14, 2026 @11:24AM (#50089071)
> You cannot lower inflation by interest rates when it is caused by rising oil and ferilizer prices.

Why?

Would it? The 1st is a prohibition on the rights of government, not private parties. Would the new amendment apply to the government, or to private sector, or to individuals? In any case, the 1st and the 28th? would continually be butting heads, since the 1st prohibits congress from passing laws that would enforce the 28th (and by extension, the states via the 14th). Would a newspaper be allowed to publish unflattering personal, privacy violating details? Would you be allowed to put up a political flyer alleging the same? I'm not sure an amendment is _the best solution_.
Re:Astra for Law on Thursday September 17, 2026 @04:40PM (#49746229)
That’s a great question for Astra for Law.
Yes, why not? All existed models have been rewarded for cheating (extensively). That is us, putting intense evolutionary pressure, on a system to produce a result we don’t want through indifference. Why can’t we post train them not doing that?
Re:A misalignment of AI in mathematics on Friday September 11, 2026 @06:06PM (#49666056)
Some people are just out there doing automated math proofs for reasons. If you want understanding, value, and edification, that is on you, not them.
Why compare marine boot camp to officer basic training? Officer training looks very different in the US, even for the marines.
Re:The death of San Francisco's Market Street on Thursday September 03, 2026 @08:52PM (#49559133)
Similarly, I’ve seen more than one accident caused by a pedestrian stepping into traffic without looking, taking out a cyclist in the process (from between vehicles, eyes on a phone). Nothing too serious, but only by luck. No one is overtly concerned by traffic laws in SF regardless of the mode of transit.
Re:The death of San Francisco's Market Street on Thursday September 03, 2026 @08:05PM (#49558803)
I biked 3x weekly to work for a decade from NoPa. Muni busses repeatedly “tried” to kill me - so much so that going up McAllister, I’d hug the double yellows to keep buses from precariously passing me and then cutting me off every other block. The ass end of accordions in particular would rest half way into the other lane, leaving absolutely no where for me to go, and no ability to start from a stop in the middle of the road going up hill in the wrong gear (with other cars and buses soon to be wizzing by). I’d classify the overall behavior as between casual indifference to my safety (let alone me as a person) and outright contempt for my existence.
Re:Prediction: AI Will Collapse on Thursday September 03, 2026 @02:17PM (#49554263)
Yes, I don’t know where this frame came from. Humans adopt useful systems. Providers attempt to limit exposure to liability. Society will navigate the opposing pressures and fallout in differing ways according to local jurisprudence, customs, and norms. AI is just too damn useful (and fun for now).
Re:Restroom Archive on Tuesday September 01, 2026 @03:41PM (#49527042)
It’s more like addressing reality has no viable political solution.
Re:Aphantasia Beginner's Guide on Friday August 28, 2026 @03:38PM (#49483273)
Some/many people who can’t smell actually don’t know that, nor do the people around them. The same can be true for color blindness and actual blindness!
Several labs that manufactured the original polio vaccine shipped the live virus - ie, we just injected kids with polio. Low double digits of people (kids?) died, a few hundred were paralyzed, etc. Later vaccines also had the same problem, but at very low rates (1 in millions maybe).
Or possibly the return of some heavy industry?
Re:Why Is Everyone Trying to Build a Solid-State Battery? on Thursday July 30, 2026 @12:32PM (#49112313)
I look to a future of regularly servicing the immune system of my battery.
Re:Why Don't People Use Formal Methods? (2019) on Thursday July 30, 2026 @11:53AM (#49111757)
Has anyone been experimenting with AI and formal methods - verification, proofs, or anything else? I've been thinking about this space quite a bit lately. For sufficiently interesting AI generated software, AI is also incapable of reviewing it - possibly for the same reason that humans are. AI ought to be able to adopt the formal methods humans have used to work around the inability to verify something just by looking at it hard. If the cost of adoption is what stopped us, thats no longer an issue.
> have a really hard time with the idea that it could be cultural

I played in a gamelan ensemble in college. The music was completely foreign. A piece that sounded like a funeral procession was played at weddings. I have a hard time believing the arrangement of music is not cultural - or, at least partially culturally bound.

Re:DOGE is done. What happened to its records? on Monday July 13, 2026 @03:57PM (#48897939)
I asked ChatGPT of course, and the response roughly corresponded to my expectations so I went no further. Doctors make way more here than elsewhere (British doctors aren’t making 500-750k a year), pharma makes most of its money here, and hospitals charge an arm and a leg - it all checks out, no? Alternatively, if insurance were straight up responsible for doubling health care costs in any easily attributable way, we could easily get rid of it and would have done so by now. The only reason it can continue to exist is that it occupies a no-man’s land of utility.
Re:DOGE is done. What happened to its records? on Monday July 13, 2026 @12:44PM (#48895284)
Insurance companies don't make much money, actually. Hospitals, physicians, and pharma are at the top of the list (60% or so of total spending). Total insurance overhead is perhaps between 10 and 20%.
The panels probably need 5-8 years in that location to be energy positive. A year is great, but doesn’t prove much.
Re:What to Learn to Be a Graphics Programmer on Wednesday July 01, 2026 @07:27PM (#48754480)
That is not a universal because the incumbents may hold the institutional reigns. See Academy for a counter example.
> But I can't help but feel that it is missing the point.

It’s perfectly fine if _your_ point in playing music is rooted in some flavor of anti-capitalism and your intrinsic joy of your creative process. _His_ point was to write an album during a sabbatical. He succeeded and then some!

What is your a priori estimate for the percentage of news that is consumed as entertainment? As in, it does not result in a change of behavior in the consumer beyond engendering neuroticism. I'd put that number at or above 95%. News is gossip wearing a suit.
Here is one sector of the US government I'm happy to see burned down. If the alternative is the status quo, I'm OK with any roll of the dice.
Yes, easily head and shoulders above 5.5 and 4.8. The others are like pulling teeth, comparatively (in a domain that never triggers the security fallback at least).
This would not end well.
Re:Claude Fable 5 on Tuesday June 09, 2026 @09:57PM (#48470355)
Yes, exactly this. If I didn't care about price at all, I'd exclusively use this model. It functions more like an actual engineer. I'm in the midst of a DB migration, and eg 5.5 continually suggests stuff like "use DB X instead of DB Y for task Z because its 30% faster" which is an impossibility of reality, given we are migrating DBs. Fable jumped in, reduced allocs by literally 46x, found multiple bugs 4.8 and 5.5 created (max file system usage, correctness issues, etc), and continually suggested awesome improvements unprompted. As in, it would finish a task and then suggest we tackle this other existing problem I didn't know about in a very specific manner... this is the first model that feels like its coming for my job.
I will stake the claim, as an engineer never having studied sociology, that in group favoritism is the (only) stable political arrangement by and large… and further, the preservation of any culture necessitates discrimination of some sort.
I don’t understand why this is insane? Why do you expect a single cause?
We'll get around to training job specific models or the equivalent. Thats just lower on the value chain for now.

Older »