Most people would be fine with 100 Mb: dual condensed HD streams, and relatively low latency. Hell: aside from the ISOs I occasionally download, I'd be fine with 25 Mb.
But.
This doesn't take into account the future. When I upgraded to 2400 baud, I thought I was pretty hot stuff then, too. If we don't keep the bar high, there's no real incentive to increase speed along the fringes -- rural communities, etc. I even know one guy who still uses dialup.
For individuals there really is only one reasonable course here. Ignore the sensational studies and news.
Man, if you're getting your science information from news stories, the problem is you.
Man we live in a world where phantomfive does not even read the sentences he is quoting. Maybe the problem is YOU and maybe it is a symptom of to much screen time.
Screen time is killing our kids, no wait screen time is making our kids smarter than ever, maybe screen time does not really matter, omg screen time is killing our kids...
For individuals there really is only one reasonable course here. Ignore the sensational studies and news. Make sure you kids get some play with things that are not screens. Make sure not all the screen time is 'games' and at least some of it cognitive activities even if it is writing little text games in python/ruby/perl/etc that AI could crap out in seconds, letting them think work and solve a problem matters.
wait 20 years for the science to catch up as it has on cholesterol and saturated fats; tell yourself you did your best and let your kids know to not mess up your grand kids they way you did them.
"patch your shit" totally works too with 1000s of patches per month thanks to 'AI'
sounds exactly like sort of guidance CISA would spend $1 billion in tax dollars coming up with in 2026.
I have negotiated as part of my hiring package some additional time off in my first year for some pre-planed travel, or other event for pretty much every job I have ever taken after high school.
This is really no different. In all seriousness the main motivation for almost anyone taking any job is money. Even if you "love what you do" the vast majority could simply do it (at some scale) as a hobby.
Making this about GTA gets attention, its free advertising for military recruitment. In reality it is essentially four days of extra PTO as hiring / retention sweeter. I would even be surprised if someone thinking of signing on for another tour said, can have my four days in
Point is this seems really smart to me. It is headline grabbing, and it is very likely something that was already on the table anyway.
A lot of states actually have five mile an hour laws, where they explicitly don't write citations for people going five over or less for exactly those reasons. We don't want to be punishing people for minor calibration errors ( could be the cops equipment too ) when the reason for the law is to prevent people from creating dangerous situations. 68 in a 60 zone probably isn't most of the time.
However I bet you that same officer watches people buzz by at 70+ all day long and also does nothing. I don't think rsilvergun is actually far off here. I just think he isn't clear on who is fooling who. All the people with influence do like the status quo. Those speed limits were set a long time ago when equipment really was less effective and less safe. The stopping distance in my modern car is way better than the one I was driving 30 years ago and the speed limit hasn't changed. Cops do like the discretion in pulling people over; they may even tell themselves their subjective judgements about the apparent maintenance status, alertness of the driver, etc justify that discretion that they are not being unfair and making everyone safer. No voter really is asking for more speeding tickets (well not for ones to be issued the them anyway). In fact if you ask any of these people they'd probably tell you justice should be blind/equal/... but all their actions don't really line up with that.
Just like the cop/councilman/whoever who believes in the flock camera, he probably believes he'd never abuse it but I guarantee left unsupervised long enough sooner or later will rationalize 'one little lookup of
I return to flock. Flock is an example. The real problems are:
-Compulsory display of unique identifiers in the age of big data
-Free for all photography of public spaces
If Flock does not ALPN all the things people will just snap a bunch of photos and on phones and Google's AI efforts will end up compiling the same dataset. The costs, customers, use case on the tin might change the effects dont.
As long as everyone is driving around with number plates on vehicles, and letting people snap photos of them everywhere they go in public on connected devices privacy is an illusion. Their real objection to Flock isnt the privacy violation its that it breaks the illusion..
The only spasm Russia is having is a death spasm, which it has been having since about 2014, and that has nothing at all to do with the USA playing world police or not.
It has everything to do with not really opening up the Western world to economic participation by Russia in the early 90s because because Yeltsin could not remake the entire Russian economy fast enough for our preferences. We treated them like that gas where everyone pumps and leaves without ever stopping in the quick-shop portion, they have been starving since.
The despair has placed and kept Putin in power a long with other reactionaries and looters. Ukraine was a desperation play too, because Putin had to see that people were going to start asking 'when is it going to get better' and he had no answers. Russia isn't influential, and has not needed any real checking by anyone. That was apparent in the first days the operation, it was logistical s***-show from the start. They have failed utterly to invade a neighbor they *ostensibly* dwarfed physically and economically; because the truth was Russia they were all bluster with little other than legacy soviet projects and no blow. This should have been apparent to the intel people (I am sure it actually was) because otherwise there would have been no need for 2000s Russia to be playing with an Aircraft carrier they can't even service, and finishing construction of 1970s submarine classes and trying to play them off as contemporary boats.
They got nothing now, they had little. What is left is some nukes that are strategically impossible to use due to MAD.
Had we just ignored Russia post 1991, let them trade on neutral terms and let people travely they'd probably have ended up with a somewhat functional government and enough going on at home that foreign adventures to chase false memories of old soviet glory would have held enough nationalistic appeal to be seen as domestic political tool worth employing. At least not beyond messing around in some low consequence easily abandon playground like Syria for more than news cycle or two.
A lot of people have very un-even incomes too. Traditionally those folks build up a cash reserve they draw dawn in the lean time and build up at 'harvest'; but for those starting out they almost always have used either a gift or a loan.
If you are a farmer you have to buy the seed for *at least* your first planting and you won't have income until you have grain ready for market. Some more white collar careers feature these booms and busts too. If you are sales person marketing stuff large business you might see huge swings in commission q1 when buyers have new budgets, and again in q4 when they spend due to use it or lose it policies. Meanwhile in q2-3 you have people on vacations, managers holding back spending in case of surprise expenses or shifting upper management priorities. My point is not everyone even relatively successful high earners get consistent paycheck every 2 weeks.
what is different here or appears to be, is that so many people are down the wire, and expenses are rising faster than their forward planning allowed for. That leaves additional short term borrowing to fill in the gap. The problem is the gap plus the additional interest expense leaves them with even less to store up for the next slow period. - Which brings is to the real problem for society at large
Fundamentally savings (that make return of principle certain) are pretty inefficient. That usually means capital that simply can't be employed because to do so implies some risk. So the whole economy slows down because assets must sit idle. This is why stable inflation, reliable legal frameworks, and debated negotiated policy with good forward guidance is so important. These are really the secrets to America's post war boom. Not the only factors but the things that let us take maximum advantage of the environment.
Increasingly we have been breaking these rules with one moral crisis after another, where we have abandon principle and the chickens are coming home to roost. It started with with 07/08 financial crisis response and it has been one thing after another slowly boiling the frogs until COVID hit and we really went completely bat-shit crazy, and then spent the next 3 years letting crazy ride. At this point their IS going to be a train crash. You can see it when the treasury department is trying to game the yen carry trade because they have to prevent bond yields from going nuts, but can't actually manage to pull it off mostly for lack real value to inject.
Its going to get ugly but you also can't sell. Because its going to get ugly no matter what you swap into. One thing I am pretty sure of if you have a lot cash assets you will burned badly. People need to recognize "sell the news", front running it will see you punished not intentionally but because the real estate and corporate markets HAVE to be bailed out if at all possible. The government does not want to see people turned out of their homes in mass, or empty grocery store shelves. Even if they wipe out equity values with forced stock issuance they going to be doing the injection more printed money and FED balance sheet expansion. So it is going wipe out the value of cash and nuke the return on debt at the same time. You might think the rest of the world won't let us get away with it but that is also not true. Sure Japan/China/Others can sell those bonds, but that money then must go somewhere, it could only go to the EU, where again it would have to go somewhere lest it trigger hyper inflation there, which would mean wait for it, they'd probably be back buying US bonds.. The dollar still is the reserve currency, so at least for one round here the US Treasury will 'get' to revalue it radically. I could not hazard a guess what happens after that 'singularity' nor do I know when it arrives. Might be a two weeks or 10 years, my advice to any still reading, is do that read the damn news every day because this will happen slowly then very quickly. Know your account numbers, your brokers phone number, how to make 401k withdrawls etc so you can act when it is time to act.
That is the whole point though! The datacenters don't go offline. They have their own generating plant for backup power. Obviously that isn't going to be as efficent environmentally or cost wise compared to grid scale power generation.
However, it is similarly not efficent to scale the grid to handle corner case capacity situations like a transient switching failure or a plant needing to go offline for emergency maintenance all while maintaining current loads when there are whole bunch of loads that represent 1/3 of the consumption of entire states (see today's other data center article).
All these DCs already are capable of stand alone operation because they have to be should their grid connection get severed. Having it written into their agreements that the grid operators and generator operators also have the right the ask them go standalone, for operational reasons is a good way to rationalize the cost of providing a reliable electric grid to consumers, without adding a whole lot of cost/burden on the data center operators either.
What does that have to do with anything. Environmental footprint is a budget like everything else.
Housing is most peoples largest expense. Your statement is like saying someone should address the cost structure of their housing before complaining they can't afford a thoroughbred race horse. It is a question of priority, most people if you ask them would probably answer that they'd sure like have tripple-crown winner in a stable somewhere, but having a comfortable place to live is higher priority and if they never even own race horse well they are not losing sleep over it.
I'll take a side of AI slop with my beef, if I can have it all sure, but I am perfectly happy to leave the AI until I have enjoyed some nice steak and few burgers thank you very much.
You would not care about things like trafficking, drug dealing, reckless driving etc.
Whatever happen to trust the science? Basic psychology tells you the best way to condition people to not do a thing is to have enforcement be swift and certain. A large amount of crime happens because people expect to get away with it. Every-time someone speeds for example this is almost universally the case. Why do they have that exception? because it is mostly true.
If anything about how we administer society was reasonable we would either, raise the speed limits, because everyone drives faster anyway, or more strictly enforce them. People complain about all sorts of aspect of criminal justice and fairness but nobody does anything about the most obvious contradictions with simplest of answers. The sooner you start understanding that people don't really want to justice, the sooner stuff will start to make sense.
What is worse the police being able to do so dragnets (not good) but the flip side is they actually solve some crimes you do care about, or random companies hovering up data from all kinda of public records and using it for gwd only knows what?
We can argue if surveillance tools like this are valuable enough to policing to justify the intrusions and what the controls should be. 'Get warrant' I expect actually would destroy almost all the value because the main problem they are trying to solve with this are crime spotting, and suspect identification, not 'find this guy' even if they are super effective at that, type 1 errors aside...
The real issues are public recording and data aggregation generally and in the commercial space particularly. You don't want Amazon showing you a little higher prices than your neighbor because they know how clean you keep your car, do you? That is probably the most innocent application.
As far as law enforcement and government go... The reality all this data is going to get swept and aggregated up AND no matter what laws you make will be abused by government, in all sorts of ways we can't even imagine (though not like this is happening already). The laws we make do not matter because there is no public will to hold government accountable at all. The Epstine files are instructive here, a law was very publicly enacted with very wide public support, it required all the files to be delivered to congress and later published. Specifically it called for ONLY VICTIMS to be redacted. Guess what happened the FBI redacted all sorts of things besides just VICTIM identification before the files ever got to the parent DOJ. It took Khanna and Massie about 2 hours to figure that out looking at what they got. Nobody did a f***ing thing! Again laws don't matter here, because they simply won't be followed.
If we want privacy though we need to bite the bullet and recognize the only things are really going to work are, heavy prohibitions against public recording by EVERYONE individuals, corporates, and yes even so called journalists. Similarly strict controls on aggregation and combination of ANY data-source you did not yourself generate first party. For the sake of being able to do research and understand cause and effect we have to allow some of that but probably need to create a national registry of PII, if have any such records on individuals you don't directly employ, care for as a patient, provide a financial service to, the record formats should have to be disclosed, a long with a statement of users in a consistent searchable format, and expected retention period.
At least in theory a lot of the risk isn't some specific drone.
But market risk. The worry would not be so much that we get attacked by/with drones in some way but that we become economically dependent on them and China ends up controlling a big share of the market.
Imagine a scenario 15 years from now, where 80% of home deliveries are done by drone. The trucks have been parked and the drivers laid off for a long time. You can't just turn a key and go backwards. US-Sino relation suddenly deteriorate, a shouting match at the UN Security council ends with the State Department revoking all Chinese national visas and China retaliates by near total ban on exporting goods to the US. Getting parts for anything takes weeks or months while they get laundered thru the EU or elsewhere. In the mean time those drones need parts and maintenance kits, kits nobody domestic has..
It would be crippling.
Is any of the that "likely" does it justify the policy, is the policy legal, all fair questions. I am just pointing out there is a logic to high tariffs on certain goods vs a total ban. If you already have a bunch of Chinese drones in your fleet for example you might want chose to pay the taxes rather than abandon your investment. Uncle Sam might not really have an issue with that, as long as the shop down the street chooses alternative sources.
Depends if I am young, single (so don't need more than one bag of groceries or so), and have enough disposable income - the Porsche, otherwise the Honda.
Exactly there is no real moat here; and more importantly there are other player that have other revenue streams.
Google, Amazon, Microsoft, Oracle, IBM - even if they are not the market leader, even if they are in fact way behind in this space, they don't need to win in the SaaS AI space on technical merit to ultimately gobble up the market they just need to still be in the game with usuably-relevant technology after Anthropic and OpenAI run out of investors willing to set money on fire.
The risks to that strategy mainly are
1) There is some kind of singularity and Anthropic / OpenAI get there first (yeah right)
2) These companies manage to go public and the public remains bamboozled enough to believe the sauce is sufficently secret and specially to just keep throwing money at it for so long technology changes the cost structure enough they actually can run profitable (I also don't see this because if that happens, it will also be cheaper for everyone to just run some really-pretty-good open weight model)
Earth is a beta site.
