We seem to be at an inflection point with AI -- companies across the board are investing billions of dollars into AI R&D and I expect that we'll start to see some really amazing products and services coming out of this in the coming decades.
This is a real concern. But as I see it, the difference is that the last time around, everyone was sold on the potential applications -- which turned out to be way harder than anyone realized.
This time around, the technology is seeing real applications today, so the valuations are more grounded in reality. So while people are definitely investing based on the future potential, the worst case scenario -- that we're going to hit a wall next week where no further progress can be made in machine learning research -- wouldn't be as devastating as the last AI winter.
At this point there's a lot of work to do, and money to be made, applying the current state of the art even if no further progress can be made.
The next AI winter will come when we maxed out on the current technology, i.e. when model training will become too expensive. If all that is left to do is to run training on top of all of Youtube videos, one might have to wait for 10 years until that becomes feasible.
It probably will be something more like the Kinect: a few products come out that aren't useful at all.
It took decades before people figured out Aluminum was actually useful for things for example. It was a chemical curiosity for the later-half of the 1800s (hmm, this is a cheap metal that is found everywhere. But its weaker than steel, what should we use it for?)
Just because you discover something useful doesn't mean you figure out what to do with it.
Yup. The reason why it was not used for a long time is that it was hard to make. Aluminum is a very reactive metal; when combined with oxygen or other stuff, it sits at the bottom of a deep pit of energy. It takes a lot of effort to get it out of there. The electrolytic process is basically a brute-force approach: throw enough energy at anything, and it will start moving eventually.
Napoleon III had his fancy-dinner utensils made from aluminum, for those occasions when gold did not seem lavish enough. And then cheap manufacturing was invented, and the rest is history.
>Earth Loses 50,000 Tonnes of Mass Every Year. According to some calculations, the Earth is losing 50,000 tonnes of mass every single year, even though an extra 40,000 tonnes of space dust converge onto the Earth's gravity well, it's still losing weight.
10 tons lost a year. Point taken, and very interesting. But it's still small compared to 100 tons of cargo going to Mars 1000 times to build a new civilization on an uninhabitable planet.
The US Marshal Service is basically the judiciary's law enforcement branch. Yes, the DEA is also part of the DoJ, but the DEA is a lot more akin to the FBI than the Marshal Service.
Marshals do prisoner security and transport, run the witness protection program, and are the legal enforcers of the court's orders. For instance, when the Supreme Court ordered the integration of Southern schools, it was US Marshals who actually enforced the order and were deployed to escort students into their schoools.
I don't think our legislators are "crazy", I just think they fall too far on one side of some values spectrum. In this case, Feinstein cares too much about national security and federal LEO tools and not enough about personal and data security.
I consider Mr. Donald a special case. Who would have had audacity to sue an airport over plane's noise... twice!
His Wiki read is really a material for a good movie. From over 150 lawsuits Mr. Trump is involved, the most hilarious seems to be his feud with Bill Maher.
Wiki does a better job at describing it...
Trump sued comedian Bill Maher for $5 million in 2013. Maher had appeared on The Tonight Show with Jay Leno and had offered to pay $5 million to a charity if Trump produced his birth certificate to prove that Trump's mother had not mated with an orangutan. This was said by Maher in response to Trump having previously challenged Obama to produce his birth certificate, and offering $5 million payable to a charity of Obama's choice, if Obama produced his college applications, transcripts, and passport records. Trump produced his birth certificate and filed a lawsuit after Maher was not forthcoming, claiming that Maher's $5 million offer was legally binding. "I don't think he was joking," Trump said. "He said it with venom." Trump withdrew his lawsuit against the comedian after eight weeks.
(a) it took a freedom of information request to make this information public.
(b) the Pentagon did it's own internal report and found that there was no wrongdoing.
(c) that nobody in the government is going to hold these clowns responsible or create any sort of legitimate process for determining whether these flights were legal or not.
"A senior policy analyst for the ACLU, Jay Stanley, said it is good news no legal violations were found"
"the Pentagon established interim guidance in 2006 governing when and whether the unmanned aircraft could be used domestically. The interim policy allowed spy drones to be used for homeland defense purposes in the U.S. and to assist civil authorities."
That's something that really sets me off; the hiding behind the umbrella of "well, that's legal"
Just because something is legal or some twisted definition thereof (eg waterboarding is not torture) does not mean that it's right. And for people of power to hide behind the guise of legality is so often not sufficiently called out and deterred.
Sure, but it's hard to make the case that using aerial surveillance to assist in lawful arrest operations is "not right". Weather the surveillance is manned or unmanned isn't of ethical concern. To which branch of government it belongs may be of ethical concern, but I think such concern is better described as "legalistic".
So at a micro level, this may seem like a good idea. I believe we need considering such activities in a larger context.
For example, there have been multiple laws passed/proposed in various countries because "think of the children" or "terrorism". The Patriot Act was written before 9/11 and passed in a hurry. To the point, where the original architect of the act now regrets his actions [1]. Or the fact that the act is now overwhelming used for prosecuting drug activities instead of its original intended purpose [2].
> "Weather the surveillance is manned or unmanned isn't of ethical concern"
I would disagree with this statement. It's the difference between a police stakeout and them planting a webcam [3].
I repeatedly stress at work to people involved with big data projects that want to advance the notion of a data lake. "Just because we can does not mean that we should". I'm not convinced that the legal route is sufficient to check such activity and is why I'm a proponent of systems that prevent bad actors from intercepting (zero knowledge?) instead of "we promise not to look" solutions.
I'll end with:
"First they came for the Socialists, and I did not speak out—
Because I was not a Socialist.
Then they came for the Trade Unionists, and I did not speak out—
Because I was not a Trade Unionist.
Then they came for the Jews, and I did not speak out—
Because I was not a Jew.
Then they came for me—and there was no one left to speak for me."
The article talks about less than 20 missions over a 9 year period. This is hardly mass surveillance and it's disingenuous to suggest it is simply because the platform was unmanned. I'll add that it's detrimental to the quality of discussion we enjoy here on HN to conclude such hyperbole with a holocaust reference.
I think all of us on HN are familiar with the qualitative effects of "big data" both in general and as it pertains to surveillance. I hope we're all sophisticated enough to realize that the issues at hand revolve around scale, probable cause, and data retention/mining, rather than the canard of whether the platform involved was manned or unmanned.
My original comment did not mention anything about manned vs. unmanned. I was merely responding to your claim that it does not make a difference which I do not agree with.
As to your statement that it's <20 over 9 years, at the risk of sounding a little tin-foilish, but that's what is publicly being admitted to.
There's could be, which is of course anything, and is known to be, which AFAIK is only JSOC aka the folks in charge of seal team 6 and delta force, lately of the drone attacks in the middle east, etc.
The Sec Defense can make available for "expert advice" and "training" and stuff like that, and that's what JSOC operates under. AFAIK they're the only army command running under those rules on a continuous basis.
Note that the Army does bust active duty MPs for doing things off base.
Another rarely understood part of the Posse Comitatus act is a state governor can do anything he wants inside his own state with NG / ANG troops. Its only when they're federalized or operating outside their state that they can't enforce laws. NG troops can "pretty much" do anything their governor orders them to do within their state borders. The active, reserve, and guard components are at least sometimes deeply intermeshed. Would not surprise me for a guard pilot to casually be called an "army dude" by a journalist with no military experience (aka all of them)
4 was by far the best 'classic' Civ experience. That being said, I think the one unit per tile change in Civ5 was needed and added a lot of strategy to actually fighting. My hope in Civ6 is that tiles are smaller and that cities and other features can take up multiple tiles.
At no point in the war was the Luftwaffe capable of (a) clearing the English Channel for such an invasion and (b) keeping the channel clear for reinforcements. It might have been possible to get an initial landing party on the shores of England, but reinforcing them would have been extraordinarily difficult.
The Luftwaffe would have been harassed not only by surface ships with rather formidable anti-aircraft systems, but also by the RAF which they continuously failed to knock out of the war.
Let's also not forget that Germany possessed no vessels truly capable of taking an invasion force across the Channel to begin with.
I guess I don't really see the value to businesses. People have an awful tendency to (a) overrate themselves dramatically and (b) have wildly unrealistic expectations for what they're worth and what's out there. So you wind up with a bunch of "10/10" Ruby engineers who think they're worth 150k, when the reality is that they're probably 7/10 guys that are worth 2/3s of that.
Yup. The problem with "solving" ad blocking is that it isn't lucrative. Ad blockers are trivial to create, but very difficult to monetize. These guys are as much scumbags as shady advertisers who bombard you with terrible ads.
