Slashdot Mirror

Counterpoint to your post: The Kentucky Coal Museum is going solar to save money.

That's some weapons-grade irony right there.

The problem with such systems is that we have absolutely no idea how they'll behave at various edge cases.

I fail to see how this is a problem, unless these "edge cases" are things that doctors get correct far more often. My suspicion is that they will not be. As a good example, see the recent study that the older the doctor, the higher the patient mortality rate. There was no causation in the study, only correlation, but regardless of the causation, it's pretty darn likely that AI will soon be doing a better job with diagnoses. AI can be updated with modern understanding and recent studies in a way that a brain which has done medicine for 30 years can't be. We can put far more info into AI systems than we can put into a brain, and actually get meaningful decisions out of it. Theoretically, we'll have less bias with AI, as it won't come with cultural/ageist/sexist/racist notions, and will rely upon data.
 
And as others have pointed out, we can diagnose what went wrong, and then feed it back into the AI systems to learn from. One edge case can instantly be diagnosed and sent to all the AI systems to incorporate in their algorithms. That's going to reduce them even faster, as it will only take a few examples before that particular edge case doesn't happen any more.
 
As to who's responsible, it's who owns and operates the AI. They chose that over a human doctor. Most likely, they chose that because it's cheaper and/or better than a human doctor. If that's the case, they should be OK with taking responsibility for the misdiagnosis, because AI is a net benefit over employing a doctor. If it's not, then they're idiots, and should get sued for a misdiagnosis. Regardless, the owner/operator of the AI will still likely need malpractice insurance.

AC is being sarcastic. For the willfully ignorant/forgetful out there

You're
still
not
safe.

Also, here's a nice piece on them from Ars. A little old but relevant.
https://arstechnica.com/tech-p...

What are you talking about? After Netflix pays Comcast, speeds improve 65%

How retarded can you be?

True, but in this case, Microsoft was only able to develop that patch because the NSA told them about the vulnerability.
"The Shadow Brokers" guys published a screen-shot of one of the NSA hacking tools, apparently called "DoublePulsar". That tool was intended to be installed using the SMB exploit, so the NSA knew that the tool & vulnerability were loose in the wild. NSA then informed Microsoft so they could roll out a patch.

https://arstechnica.com/securi...

Netflix was the cheapskate buying transit from other shitty providers and then acting like they had nothing to with the congestion issues that arose between their ISP and the ISP(s) of their customers.

Bullshit!

It's also quite telling that Comcast refused to install the content caches that Netflix and others offered for free that would have drastically reduced Comcast's peering traffic.

Thought Xiaomi beat them on that feature https://arstechnica.com/gadget...

Errm, nope. Try again. Hint: nearly bezel-less is like nearly pregnant.

Sometimes the trailer is better than the movie.

I'm not sure if "King Arthur: Legend of the Sword" proves or disproves your theory.

The trailer was pretty confusing and then you get King Arthur: Legend of the Sword is bad in ways I didn’t realize movies could be bad

BTW, JUDGE SMASH was unabashedly lifted from one of my favorite ArsTechnica headline/illustration combos.

Remotely exploitable network vulnerabilities shouldn't happen, but there seems no practical hope that they'll stop anytime soon. It would be negligent of legitimate spy agencies to fail to search for them and arguably be able to take advantage of them. Imagine you're trying to find out when an ISIS group is planning a bombing and you discover they're running a messageboard on a Windows machine with an SMB exploit, do you tell Microsoft to patch the exploit?

You never know which of the vulnerabilities you'll be able to use, but if you dedicate sufficient resources to finding them and building exploits for them, then there is a good chance you'll be able to spy on whichever bad guy your agency needs to spy on when the need arises. Getting all the vendors to patch the exploits you find does limit your own agency's ability to spy but you have to assume it doesn't impair your enemies as significantly since the enemy doubtless will have exploits you don't have.

What's the best solution? I suspect the best thing to do is build force-patch worms for every exploit. If you write an exploit, you should also dedicate resources to the task of writing a version of the exploit which pressures the owner of the exploited system to fix the problem. So in this instance, as soon as the attacks started being seen in the wild, the NSA servers should have launched a MASSIVE attack against any and all systems with the vulnerability which would disable the vulnerable systems in the least painful ways along with alerting the owners of the need to update their systems. Instead of getting "your files are encrypted and give hackers bitcoin to recover" messages, the people with exploitable systems should be seeing warnings like "Your system has been temporarily patched by the NSA for your own protection, please secure or update your device to protect it from malicious actors."

The Hajime botnet may actually already be just the thing I'm describing. I'd prefer to see the NSA take public responsibility, and I'm doubtful the NSA is actually responsible for that one, but it is an example of how it could be done.

If I have a vulnerable system, I'd much prefer to see it hacked by the NSA instead of some ransomware writer. Do I wish it wasn't hackable? Of course, but I accept that anything plugged into a network might be hackable. I do what I can to protect it from everyone, including the NSA. It's not that I'm worried about the NSA (because they have the resources to gain physical access if they really want it) but if I do my best to build secure systems, then it's less likely I'll wake up to a ransomware message some morning.

Germany land area: 348,900.0 sq. miles
United States land area: 9,147,420.0 sq. miles
Over 26 times the land area. In other words, it's easier to get the power
from the source to the people in Germany, than it would be in the USA.
It's the same argument, people complain about, when talking about internet
speed. "Japan & South Korea" have x times the speed and x cheaper price
of internet, than in the USA.
Japan land area: 364,560.0 sq miles
South Korea land area: 97,480.0
It's easier in smaller countries to build out, than the USA.
Ok, so the liberal logic would say, move all the people to large urban population
centers, and make living out in the middle of the nation illegal. Yeah, that would be
the way their liberal minds would work. After all, it's "for your own good".
You want renewable power, fine, but leave coal, gas & nuclear alone.

Well thanks for showing us how your mind works. You spout a random statistic of no particular value, the size of the US versus Germany, but don't consider the population distribution, the same as with your claim about the Internet. Despite numerous and extensive discourses on both flawed argumentations of yours having been available for quite some time. The one they have in common is that you're trying ignore the distribution of population, which is not even across any of the nations, and in fact, you can see how there is plenty of concentrated population in the US with any number of maps. The next part is that you're ignoring the actual state of affairs, as the complaints about power production and internet provision actually don't just depend on the averages, but on particulars, such as North Carolina's complaints about pollution to North Carolina's actions to inhibit municipal internet.

Then, of course, you haven't actually considered that efforts to create a better interconnected grid in the US are ongoing, that alternative power production can work better with distribution.

But instead of looking at those details, let's just let you try to boil things down with an irrelevant argument that isn't particularly important. No wait, let's not.

Of course, that you ended it all with a false conclusion of what you think your strawman concept of a liberal mind would say, just puts another nail in the coffin of your argument. Why do you bother? Is it simply because you don't realize how poor a case you're making, or is that the point? Heck, maybe you're around to make conservatives look bad, that'd be offensive, but at least I wouldn't feel quite as bad.

But I mean really, Mitt Romney gabbled his nonsense over the size of the US Navy and Air Force, what else can I expect?

If the US creates powerful new tools to detect and protect against all things cyber will the US and other nations get some new protection from the Automated Implant Branch (AIB) and Network Devices Branch (NDB) efforts?
With the US government demanding better security products, the NSA and CIA contractors will have to work harder.
Only a few of the better anti virus brands found the equation group https://en.wikipedia.org/wiki/... and Fluxwire Trojan and Archangel efforts.
"Found in the wild: Vault7 hacking tools WikiLeaks says come from CIA" (4/10/2017)
https://arstechnica.com/securi...

Wont someone think of the role of the clandestine services in this rush to improve US domestic cyber security?

Tailored Access Operations (5/15/2014)
https://arstechnica.com/tech-p...

It's actually pretty interesting if true since it is yet another example of a useful service that followed the "first get popular then get permission" model and would not have made it to profitability if it had to go through all the proper legal channels.

Youtube, did it and it worked (bought by Google).
https://en.wikipedia.org/wiki/....

Imeem did it and got popular, but sold itself to Myspace to avoid the legal hole it dug itself into.
https://www.cnet.com/news/repo...

By contrast, Grooveshark was sued into oblivion (bought by no one).
https://arstechnica.com/tech-p...

Well, Android/Chromium OS are open source except the good parts. Chrome OS (Chromium OS) is roughly a modified Chrome/Chromium browser on top of the Linux kernel. What mostly makes Chrome OS and Android successful is the Google's services and apps (Google Play, Google Camera, etc) which are usually proprietary software. Also many vendors fork these OSes and make them proprietary by tivoing and closing the sources (except the Linux kernel, but GPL2 doesn't protect against tivo anyway). So it's not so rosy, Android and Chrome OS are almost proprietary. https://arstechnica.com/gadget...

But seriously if the group backing Tizen (Intel, LG, & Samsung to name a few) get their act together they are positioned to swoop in and take the market while Google tries to steer it to their new walled garden. Tizen is OSS top to bottom so there is no locked down portion. Although I'm sure each OEM will try to slap something proprietary on it to make it their own. They need to address the security holes that were recently identified and clean up their codebase but it's theirs to lose.

That is fantastic news that they are targeting 120 fps ! I hope this helps push all the other vendors (phones, monitors, etc.) stop stop targeting a shitty 30 fps experience.

At our Fortune 50 company I'm always educating our UX and Graphic Designers about the reasons why we run our app at 60 fps. Kind of hard to argue when they see a demo first hand. :-) Now if only the rest of the company would get on board ditching the crappy 30 fps that people seem to think is "good enough."

I wonder if Google is trying to target VR at some point placing a safe bet of 120 so they can hit the magic 90+ FPS required? The 120 fps for apps is just a bonus

> Android hung around inside Google for about five years before it launched on a real product.

So basically Fuschia is a tech demo today -- that may, or may not ship.

I wonder if they are going to ignore the whole Android ecosystem or embrace it, because 2 billion devices running Android is pretty hard to ignore.

And how do you think it does that? They hack the browser, only the payload is different. Granted their code doesn't do anything malicious but anyone can take their exploit, pop in a cryptolocker and have their own remotely exploitable 0-day malware. This is the police going black hat along with the NSA.

There are reports of them being compromised via Web pages: https://arstechnica.com/securi...

At pwn2own someone(s) actually managed to break out of Edge and vmware - https://arstechnica.com/securi...

Wouldn't The Thaw be a more appropriate move reference?

I thought the story was click-bait for a British TV series on Sky (and Amazon) called Fortitude. Totally same idea, except in the show the reindeer is a mammoth, and instead of anthrax it's larvae from an ice-age insect that infect people's brains. Coincidence?

Fortitude is the almost scientifically sound TV show you should be watching

SpaceX expects its own latencies to be between 25 and 35ms, similar to the latencies measured for wired Internet services. Current satellite ISPs have latencies of 600ms or more, according to FCC measurements.

If Blizzard makes competitive Starcraft fees problematically high, Riot Games can undercut them

Riot Games is legally prohibited from providing StarCraft. If you're referring to a school dropping one e-sport in favor another, I don't see how skills from one game necessarily transfer any more than skills from baseball transfer to cricket or vice versa.

license for peanuts until the matches can start paying themselves.

In other words, dump licenses until you're a monopoly and then jack up the rates.

Don't competitive gaming leagues already exist? Don't they have rates somewhere?

Professional leagues run into the same problem. See Ars Technica's article about e-sports copyright.

They started sending way more traffic out of their network than they were used to.

That's your misunderstanding. It wasn't Cogent who started sending more traffic, it was the ISPs' own subscribers who were requesting more traffic, which Cogent was able to supply. Charging Cogent for fulfilling the data requests from the ISP makes as much sense as the subscribers billing the ISP for data it sends to them.

The whole mindset of ISPs charging both their subscribers for asking for traffic and their upstream connections for responding is the classic middle-man double-dip, where the aim is insert yourself firmly into a transaction (by preventing alternative arrangements) then squeeze money from any parties remotely concerned. That's not adding value, that's basically extortion. In a free market those subscribers would've shopped around for an ISP who was more willing to assure supply of their requested traffic, but the cozy ISP regional monopolies put a stop to that.

Netflix did the right thing and found another CDN and started negotiating their OWN peering agreements.

By which you mean Netflix gave up and paid the ISP's extortion demands.

since they face no consequences

They face no consequences unless they post copy (write something). True? Untrue? Don't matter. Tomorrow's another day. Business journalism. Pays bills.

In other news, Apple has $250 Billion sitting around in cash, in spite of spending on a completely insane spaceship headquarters and still not offering a decent tower workstation. Upcoming iPhone 8 rumored to be cool.

Whatever. Probably fit the entire Pentagon inside the courtyard of that spaceship. But if they would just take an 8-figure round-off error on that cash and give it to me, I would be ok with that.

since they face no consequences

They face no consequences unless they post copy (write something). True? Untrue? Don't matter. Tomorrow's another day. Business journalism. Pays bills.

In other news, Apple has $250 Billion sitting around in cash, in spite of spending on a completely insane spaceship headquarters and still not offering a decent tower workstation. Upcoming iPhone 8 rumored to be cool.

Whatever. Probably fit the entire Pentagon inside the courtyard of that spaceship. But if they would just take an 8-figure round-off error on that cash and give it to me, I would be ok with that.

Really dingus. Try reading the article again.

"The EIA measured relative emissions across the US economy as "carbon intensity"—an average of the amount of carbon any sector gives off as it consumes different kinds of fuel. The measurements were applied to five sectors of the US economy: transportation, commercial, residential, electric, and industrial." https://arstechnica.com/scienc...

...is here. Honestly, I can't see the point of putting an i7 in these. If you're doing anything that's pushing all the cores, the heat generated will force it to throttle down, unless the newest chips have become way more efficient than previous gens. Would be interesting to see a shoot-out between the i5 and i7 models.

Not so fast. AFAIK, jurisdiction over the Internet has been removed from the FTC, and it would take an act of Congress to put it back... and that sure as shit don't look likely. Any talk of the FTC, for now, is a head-fake excuse for gutting the FCC and letting Comcast and its ilk get drunk and party at your expense.

Face it, ladies. The Internet is the new telephone system - the FCC should regulate it as a common-carrier. Period. That makes it boring to the carriers, gutting a lot of "opportunities" to squeeze extra money out (like selling your browsing histories), but too fucking bad. The Internet ain't no luxury anymore - shit, your grandma needs it just to get her goddamn meds.

Besides, the FTC is not invulnerable to politics. Maybe they don't have a politically ambitious loud-mouth tool as Chairman who wants nothing more than to see himself on TV, but a GOP-controlled everything can muzzle the FTC, and they will, if the price is right.

Read today's "Slashdot news" yesterday!

https://arstechnica.com/
https://www.theregister.co.uk/
https://www.washingtonpost.com/news/the-switch/wp/2013/08/07/slashdot-founder-rob-malda-on-why-there-wont-be-another-hacker-news/

If by "rules" you mean "got a EEE pulled on them by a megacorp that turned Linux into a proprietary OS they control by cutting off support for ASOP, locking crucial APIs behind a playwall, and making GPL V3 verbotten because they can't pull a TiVo with it?" then yeah I suppose it "won" but if that is "winning" I'd sure as fuck hate to see what you would consider a loss.,/p>

Lets see if Android behaves anything like actual Linux, shall we? So you can just replace Android with any old vanilla ARM Linux build, right? Nope and in fact with each passing quarter less and less phones are able to be "rooted" (which is and of itself an insult to Linux, as having to jailbreak your own hardware is the exact opposite of open) unless you use malware like Kingoroot. You can fix it yourself, right? Nope the drivers are all black boxed and again your changes to the Android source certainly isn't gonna run on that phone you pick up in Walmart. How about the community controlling the direction of the OS, where any coder can supply changes for consideration and possibly get them integrated upstream? Bwa ha ha ha ha...not a chance in hell, Google has exactly zero fuck to give about your code or that of the community, especially since it'll probably be GPL V3 which again is verbotten precisely because they can't TiVo it.

Yeah you "won" alright, you won about as much as that Rube trying to beat the hustler playing third card monty in Times Square. Lets face it you got scammed, had, ripped off, and the truly sad part? You are actually cheering for the guy that fucked you! But don't take MY word for it, lets see what RMS and the FSF has to say...yeah not so good, unless you want to run one of only 2 now out of date phones, otherwise you might as well just be on iOS....yeah you "won" there sparky, you are just full of "win".

WhatsApp is owned by Facebook. It's encryption is a joke when the right people are asked nicely, hence the "using techniques that 'cannot be disclosed for security reasons.' What they mean is they can't tell you how they did it because it would look REALLY bad if people realized how stupid it is to put your faith in a company that specializes in profiling and biometric data collection; https://www.whatsapp.com/faq/g.... If you're using WhatsApp on Google anything (Android, Chrome, etc.), you're in even worse shape because it's Google for Christ's sake. Remember Dirty COW? Google waited until after the election to fix it while every other Linux-based OS did months ahead of them.

But anyway, Facebook also invests huge amounts of money into cloud computing and AI. That combination one day will make all encryption and anonymity useless because we will all be digitally fingerprinted whether you have an account or not, especially if quantum computing advances, and you can assume your government will get a copy, just like they get copies of your DNA when you fall for the "fun and easy" TV advertised "ancestry" services. This "profile" is going to replace social security numbers. If you want real encryption (at least for now), use Signal (similar to Telegram but better) or a Tox client (similar to OpenVPN but for messaging). More importantly, use your brain. Both are free and open source and support text, talk, video, and file sharing. I would never use anything that important that I couldn't look at the code for. If you could look at WhatsApp's source code, I think security researchers would be horrified. And, Facebook gets caught spying on their mobile app all the time, so I don't see how WhatsApp would be any different. And just because a lot of people use it, doesn't make it the best. Matter of fact, that would make more of a target.

• https://finance.yahoo.com/news...
• https://www.theregister.co.uk/...
• https://www.thememo.com/2017/0...
• https://arstechnica.com/tech-p...
• https://www.pcworld.com/articl...

Some of the above links are kind of old, but note the ISP one. Legally, your internet service provider in the U.S. can sell your browsing information. Because of this, intelligence agencies can just purchase your data for cheap rather than getting a warrant and paying a government employee to waste their time. I'm mentioning ISP because Facebook has been trying for over a year now to bring the Internet to all kinds of places. They would become an "Internet Service Provider." In any case, if the app has an advertisement, you can be tracked.

The real note to take away from this is to realize data can be created and never destroyed and don't put anything on the internet you don't want found. I wish people would realize privacy settings are a joke; they only protect you from the average person. Anytime you see "convenient" or "secure" for a service, just assume it's complete BS because your government doesn't have the time or resources to actually physically search and seize everyone so they have software for it, contrary to "Martial Law" conspiracies; cloud computing makes it easier.

And since this news regarding terrorism, do you know why it was so hard to find Osama? It's because so far as we know, the most technologically advanced thing he ever personally used was a kidney dialysis machine or the Cold War weapons the U.S. gave him. The wor

Oh, he left months ago (according to him)...
https://yro.slashdot.org/comme...

This is my last post here (place is now owned by an incompetent LITTLE DOUCHE in yourself, who has serious issues & can't CODE FOR SHIT - I am proving it RIGHT NOW in fact)!

Erm, wait this one's the last post
https://yro.slashdot.org/comme...

P.S.=> For the hell of it: Doing just 1 more post (just to spite your DIM BRAINED ASS) BEFORE I VOLUNTARIY LEAVE stupid, not you forcing it:

Who am I kidding, he ain't leaving till Ars takes him back
https://arstechnica.com/civis/...

More and more information is coming out that "peer review" is sort of a joke. The basic statistics of many studies isn't even verified. Check this on Ars: https://arstechnica.com/scienc...

While likely true an even more pressing problem is non-scientific clickbait headlines and juiced up summaries and articles about scientific papers/research to simply generate more revenue. No companies seem to care about long term irreparable harm to public consensus. Obligatory xkcd

It's still bad.

More and more information is coming out that "peer review" is sort of a joke. The basic statistics of many studies isn't even verified. Check this on Ars: https://arstechnica.com/scienc...

It seems fairly limited to me. Only Intel CPUs, only Windows 10, special drivers needed.

I was hoping for something with a SATA connector on each end.

Connect one end to the motherboard. Connect the other end to a hard drive. Power on. See a speedup.

*THAT* would have sold millions. This? Not so much.

It's also limited to 200 series Intel chipsets (i.e. Kaby Lake) or newer and only on the i3, i5, and i7 (not the lower end variants).

https://arstechnica.com/gadget...

I prefer the Ars technica article: https://arstechnica.com/gadget...
Gives a much more critical look at the product. The other reviews seem to be fawning over the new tech too much without doing proper real world comparisons

SteamOS is very likely to change that. Not a huge space in that market and it will likely get much tighter as manufacturers jump into that market. It is likely even game publishers and gaming studios will jump into the Steam OS market https://arstechnica.com/gaming..., as it grows. In the back of managements mind will be escaping M$ licensing fees and controls and creating a easier to access gaming market. Valve would likely do far better if the opened up SteamOS to broader investment and sharing of control. To further develop the SteamOS valve needs to start distributing popular FOSS titles (small returns but great advertising and drawing more people to their platform).

Try explaining that to the legacy mainstream media dinosaurs that are still busy taking Google to court for spidering, indexing, and linking to their content, despite the debacle of Spain a few years back, and see how far it gets you. Common sense is in short supply in some corners of the Internet, and fairly large corners at that.

ArsTechnica: once-more-with-feeling-climate-models-dont-exaggerate-warming

Do you prefer a particular brand?

No, it wasn't. But the company they just filed a copyright infringement lawsuit against is: https://arstechnica.com/tech-p...

The funny thing is that the kickstarter juicer is clearly the superior product -- it's less expensive and lets you put your own mix of produce in the bag that it squeezes.

I don't see why regulatory agencies shouldn't be able to test products.... IF they are doing it properly though.

Because if Subway is right on this one, and it sounds like they are, they have all the rights to sue CBC for it, and this isn't only to the benefit of Subway, but also to the benefit of the public.
https://arstechnica.com/scienc...

Basically, if the ArsTechnica article is right, CBC used a bad method to jump into a conclusion and premeditated an article about it for some reason. That reason could be pure incompetence or perhaps something worse, but it certainly damaged the fast food chain reputation for no good reason.

Rebuilding that sort of reputation can be extremely costly, and the fast food chain could lose far more than 210 million for it. Unfounded rumors usually already cost far more than that for other fast food chains, a regulatory agency going out of it's way to publish something like that can be far more damaging.

We'll see how it goes.

Subway will win the lawsuit. https://arstechnica.com/scienc...

Op, a bit of research (always helpful) would reveal that Subway has an excellent case against the CBC. https://arstechnica.com/science/2017/03/food-scientists-weigh-in-on-50-subway-chicken-test-its-100-weird/

Actually they do have a hydrogen car for consumers; see https://arstechnica.com/cars/2.... Not exactly for the mass market yet though.

Just a nitpick though; I agree with your point.

Gelsinger set a goal to "literally get to the point where radios are integrated into every product we build."

arsTechnica: Intel researchers put WiFi inside... 9/14/2014
I bet you don't get to see everything that happens on those.

I was surprised to learn that moz://a was involved with this research.

The article says:

Commissioned by Mozilla - the organisation behind the Firefox operating system - the study was designed to find out what it is that limits people in the developing world from grabbing the opportunities offered by the web.

And what did this research discover?

That unemployed Africans use their smart phones for gambling and for defrauding financial lenders:

Nairobi resident Evans has no regular employment and the money he does earn is used for betting.

His smartphone became a tool to research and improve his betting techniques.

He bet on football matches and used the phone to research the teams and their statistics.

In Kenya, mobile phone Sims are often tied into mobile money providers, such as M-Shwari, and this offered Evans another way to borrow money other than from his friends and neighbours.

Unlike a personal loan, which he tends to repay to prevent fights, he is much more blasé about online loans.

"He was defiant of strict loan repayment rules, which may have landed him on a financial blacklist," said Ms de Reynal.

"He felt that as long as he threw away the Sim, he could buy another with no consequences," she added.

And I suppose it also found that some Africans are gullible enough to fall for obvious prank apps.

Still, I'm not sure what's stupider, this research or moz://a's disastrous $35 Firefox OS phone.

I can't believe how badly moz://a screws up any time they try to get involved with third-world regions.