Welcome to the new Slashdot. Rob Malda sends his warm regards.

At some point, you have to start asking why you need certain quality of experience in limited environments, and what infrastructure it takes to get there.

The biggest ongoing cost for streaming movies today is CDN storage, in the sense of having enough bitrates and resolutions to be able to accommodate all target devices and connection speeds. As much as people would like to deliver an HD picture to a remote village in the Philippines over a mobile connection on a feature phone, it isn't feasible at the moment for two reasons: they don't need or care about that level of experience, and it isn't technically feasible. The goal of CDN storage is to ensure the edge delivers the content, and the industry has toyed with real-time edge transcoding/transrating to address some of these issues, but fundamentally we are dropping asymptotically to a point on visual quality for a given bitrate and amount of computing power that a codec can deliver at the playback device.

In that sense, I'm shocked that Anne's post didn't mention Netflix's own VMAF, which is a composite measure of different flavors of PSNR, SSIM and some deep learning. But even here, the fundamental is that we are still using block-based codecs for operations simply because of the fundamental nature of most video, i.e. objects moving around on a background. I'm also shocked that Anne didn't discuss alternative coding methods like wavelet-based (e.g. JPEG 2000), but - again - these approaches have their own limitations and don't address interframe encoding in the same way that a block-based codec can. If there was a novel approach to coding psychovisually-equivalent video that would address computing power, bitrate and quality reasonably, I believe it would have been brought forward already.

I think 5G deserves a big mention here that was lacking in Anne's post, because faster connections may solve many of the types of issues that affect perceived visual quality at low bitrates. Get more bandwidth, and you have a better experience. Hopefully 5G will proliferate quickly, but this will be tricky in the developing world where its inherently decentralized nature and the political environments will make its ubiquitous deployment a serious challenge.

In the end, we're all fighting entropy, particularly when it comes to encoding video. Our ability to perceive video is affected by an imperfect system - the human eye and brain. That's why we've made such gains in digital video since the MPEG-1 days. But the fantasies of ubiquitous HD video to everyone in the world on 100kbps connections are just that. When you're struggling to get by and don't have good health care or clean drinking water, the value of streaming high-quality video isn't there from a business perspective, much less a technical perspective. Everyone will get an experience relative to the capabilities of technology and the value it brings to them accordingly. All else is idealistic pipe dreams until otherwise proven.

TL;DR: The selection process is random enough for its purpose, the type of attack proposed would already require access to the data which could be manipulated anyway, and this story is bunk. When someone says that something is "random" what they really mean is that, given a finite number of possible valid values "N", that every attempt to predict that value will result in the correct value only 1/N times over an essentially infinite period of time.

Nominally, random numbers are generated through a true random seed that comes from sources such as radioactive decay, cosmic background radiation, ring oscillator or other effectively chaotic process. This is fed into a pseudorandom number generator which is a giant shift register with specified taps to generate what are nominally random numbers.

Are the implementations screwed up? Sure they are. Can they be influenced deterministically? Of course. Can this be done usefully? Not really given the value of the targets involved and the amount of infiltration required to get there. I emphasize this last point because these professors are indicating that someone could influence the random number generator. Well guess what guys? You would need access to the computer running the spreadsheet anyway, which means you could already do whatever you want to rearrange the results. Why would they waste their time influencing the RNG deterministically?

This story is muckraking bunk by people who again don't really want people to understand security as much as they want to stamp a name for themselves. I'd be much more concerned that this is being handled in a spreadsheet rather than in an air-gapped database infrastructure.

Atmos is a system, not a codec.

Traditionally, you had a pre-mixed channel bed like 5.1 (AC-3, E-AC-3, MLP and AC-4), and 7.1 (supported by E-AC-3, MLP and AC-4). With the introduction of audio objects in 3D space, E-AC-3, MLP and AC-4 are extended - and that is what Atmos basically is.

The problem is how to manage loudness when you have a channel bed and/or objects. E-AC-3, for example, had a substream type originally reserved for future use - in this case, implementing Atmos. Since E-AC-3-based Atmos is backwards compatible with legacy E-AC-3 decoders, Dolby has had to do some tricks to the metadata to insert the objects and keep loudness managed. This can only be accomplished at the renderer, and it requires tight control of the metadata to manage loudness consistently.

When you get into third-party upmixers, they do all sorts of awful things (*cough*Neural*cough*). Two things they can do due to "artistic" interpretation are to improperly locate the audio in 3D space, and mix in the incorrect level the audio that goes into the speakers. Because of differences in perception in loudness depending on location around your head, and because you aren't mixing the right level of audio at/across a given speaker, the original renderer's interpretation of loudness metadata and location metadata is incorrect. This leads to potentially disturbing variations in loudness and confusion in location of content that is the fundamental basis for Dolby providing an entire Atmos system from authoring to rendering, end-to-end.

The only place upmixers typically exist in devices anyway is in AV receivers and soundbars. Yes, they can exist in the broadcast chain somewhere before encoding and transmission, but broadcasters should know to manage that experience any time object-based audio is in play. As for the rest, Dolby already offers its own upmixer that works with the Atmos renderer. There really is no good reason to go outside of this, and licensees of Dolby technologies are only degrading the end user experience by doing this.

Again, Dolby doesn't care per se whether someone else is using another system, be it DTS or Barco or Fraunhofer. All they care about is that the content owners and distributors don't have complaints because of this. Certain folks who provide premium content, such as HBO, are huge sticklers for audio quality and have been pioneers since the beginning. If they're investing in Atmos, they don't want the downstream experience affected and so Dolby is really doing their bidding ultimately.

So no, there's no conspiracy and Dolby isn't doing this to screw anyone else over. "Blame" the content owners if you want to blame anyone, but Dolby is just trying to provide a consistent experience that has eluded folks for decades now. If you want proof of that, go watch 100 different videos from any large free streaming site and tell me that you won't touch the volume control.

It remains to be seen how much Microsoft has paid for GitHub, but why pay anything if they're unprofitable?

I keep seeing the same behavior that happened during the first dotcom boom - companies valued at stupid multiples of "earnings", including what are technically negative earnings, being valued far in excess of their worth. A company is only worth its future profits discounted at the rate of the next best investment of that money, minus its initial and ongoing investments. The longer it takes to return a profit, the exponentially more difficult it is to recover the initial investment. Only a fundamental change or an external factor like currency inflation can distort that picture into a supposedly rosy one.

Perhaps GitHub can have some of its cost structures reduced by riding on Microsoft's coattails. Perhaps there's some breakthrough that Microsoft can see with them, although I don't think there's a tremendous synergy there. The basic model has been there before (SourceForge), and it could technically be duplicated again by someone else. Many developers/repos will simply bail due to Microsoft's history of changing business terms. Heck, they rolled "Teams" out which is supposed to compete with Slack.

More power to the current owners of GitHub if they get bought out, as it's a great tool. I just think P.T. Barnum really was right, and I'm waiting for the other shoe to drop in this latest boom.

The vendetta against Gaddafi after he had capitulated to giving up his nuclear weapons program is the primary reason NK hasn't given up its weapons program.

It will probably take the withdrawl of US forces from SK and aid to NK, and the formal acknowledgment of the continuation of the current NK regime in order to denuclearize NK. Even then, I wouldn't be certain they don't retain an actual nuke or two secretly now that they have them. This still probably won't address the effectively bigger threat of all of the NK artillery pointed at SK.

Facebook has sold everyone down the river. Republicans, Democrats, you name it, it all goes down to the highest bidder as to who gets the information about the opposition. Meanwhile, they promote stories that get people pissed off and riled up from all sides so that they comment, load pages, check for updates and (surprise!) that enhanced engagement sell more ads and promoted pages/groups/whatever. They don't give a damn about anyone's privacy as long as it can be exchanged for money. This is controlled opposition, and they want us to be the pawns in that game.

I misspoke when I said all, and don't believe that cryptocurrency is immune to total collapse; indeed, we have seen some very high-profile failures for a number of different reasons. What I am saying is that many cryptocurrencies will persist even in more marginal form as ways for exchange of value outside of traditional control of currency and outside of the ability for the investment banks to be able to play a role in them. One specific corollary to this is that I also believe that they will increasingly be used by criminal organizations to launder money which, in and of itself, is a serious problem for law enforcement. Of course, when you have an organization that advised an entire country (Greece) on how to effectively hide its true fiscal condition, one has to bear in mind where the criminality actually exists here.

The high rates of return due to speculative value are already disappearing rapidly. Goldman missed the boat, and I believe their statements are a way to rationalize to their investors their inability to get in on the ground floor of these opportunities. They're probably right that many of these cryptocurrencies are going to "disappear" inasmuch as they are widely traded commodities, but as long as there are people willing to exchange real goods or currency for cryptocurrency, they by definition have a non-zero value. More importantly, I think we will see a significant slowdown in new cryptocurrencies.

Goldman is self-interested in eschewing a method of financial transactions where it does not have the ability to control or extract value out of. It got late to the party and is SOL as far as most cryptocurrencies go.

That said, most cryptocurrencies are substantially overvalued because the underlying value of any currency - crypto or otherwise - has to be backed up by some type of economy. The USD used to be on the gold standard, and only started inflating substantially after it was taken off even though a not-insubstantial portion of that value is in services and intellectual property rather than goods. The inflation of the value of the currency is a natural side-effect of a number of factors, but the ones that are most relevant in this discussion are disparate classes of valuable assets (physical and non-physical), the participants interacting with the currency, and speculation. Also remember that the value of cryptocurrencies is also being exchanges for other currencies, so there are also transaction costs and the actual value of those currencies relative to the cryptocurrency.

In any event, if we use those measures, the inherent value of any currency is the value of the actual goods and services tempered by these factors. That there has been speculation driving up the price is obvious. More importantly, we cannot state the value of all cryptocurrencies is zero strictly because of speculation, because cryptocurrency value is based on the fact that there are people are still willing to exchange goods, services and other valuables including paper currencies in exchange for cryptocurrency!

Goldman is wrong. Blockchain-based cryptocurrencies are here to stay. What isn't wrong is the analysis that states there is overinflated value in the cryptocurrency. We can, of course, also say that of the inflated value of today's normal paper currencies backed by central banks, including speculation with various instruments and the perception of their underlying value. It's the same reason I can purchase currency futures and forwards for common currencies versus requiring special instruments like letters of credit for currencies of little value or with little trade with the currency of question (e.g. try to find a forward for Turkish Lira versus Burundian Francs). The only real difference is how that transaction happens.

And since Goldman is cut out, you better believe that they and JP Morgan and all of the investment banks are doing anything they can to keep themselves relevant in this brave new world of cryptocurrency. Spread FUD, use existing political connections to regulate or shut down cryptocurrency use, whatever. It's just that this time it really may not work.

The three reasons that anyone keeps getting SWAT teams sent to them are the following three factors that must be addressed:

1. Caller ID - it's broken. Unauthenticated caller ID and caller ID spoofing should be treated as a crime since scam artists continue to take over unprotected VoIP gateways. Nothing should be connected to the PSTN without a certificate issued by the PSTN provider, period. This way there's at least some traceability and requires someone to have come on premises or seriously violated the chain of trust far beyond the skiddie level that these little bastards engage in.

2. Police attitudes - militarization of police is rampant with surplus war equipment like MRAPS, Hollywood movie style takedowns and insufficiently-vetted police officers with mental stability issues. Some modicum of rational assessment of a situation without automatically deploying people is necessary. Laser listening devices on windows, drones, or maybe just walking up to the door. It can't be break in, throw flashbangs and yell like a lunatic getting the innocent occupants to play Simon says until they can't comply and someone innocent gets shot any more.

3. Punishment - this one is simple. You SWAT, you get twenty years for each instance consecutive. Someone dies because of a swatting, you're guilty of murder and you get life imprisonment. But wait, you say you have some kind of mental disability? Well no problem, you'll just be committed to a mental facility until your condition is eliminated without drugs. Oh, and are you a provider of a gateway to the PSTN or other services that connect to police and don't work to get this done? You lose your license to operate.

So many people, including myself, are tired of this nonsense. Legislators, law enforcement and telecom companies need to start working together to prevent these things. Otherwise I say they should all be held complicit along with the perpetrators of SWAT incidents in the crimes. It is sheer lunacy that this hasn't been addressed at multiple levels yet.

Wire harnesses are a critical component to vehicle safety. Wires that can degrade during the normal service life of a vehicle can be deadly. Think about a wire harness with insulation that's been eaten that controls the ABS, fuel injection or an airflow sensor, and you hit a bump in the road and it shorts. Now you lose power or braking. Are we willing to have someone's vehicle fail and the people seriously hurt or dead because of a fundamental design flaw?

I've worked on my own cars for years and seen some really stupid compromises and designs that make regular service difficult or results in failures just outside the warranty period. This, however, takes the cake, and we need to stand up to this by declaring the insulation issue a fundamental safety issue. I'm now thinking about mitigation strategies beyond my standard maintenance that neither I nor anyone else shouldn't have to think about, like underhood blinking lights, sprays, capsaicin tapes, etc..

I would encourage anyone with one of these vehicles to file a NHTSA complaint stating that soy wire harnesses should be banned and recalls instituted to remedy the problem by either (a) replacing the harnesses with standard synthetic non-edible polymers as appropriate to the specific application, or (b) providing coatings that provably prevent rodents from consuming the insulation over the lifespan of the vehicle. We should also inform our congresscritters about this issue.

NHTSA complaint form: https://www-odi.nhtsa.dot.gov/... Congresscritters: https://www.house.gov/represen... and https://www.senate.gov/senator...

The ruling aside, there's no better way to avoid workplace communication monitoring than to use a smartphone with mobile data network connection. Most plans have more than enough data to give you everything you need while you're at work. It's pointless and counterproductive on so many levels to log into anything personal on work machine.

I think this is generally a fantastic idea. Nothing has really changed in subjects such as calculus, linear algebra, chemistry and biology in decades if not centuries for some subjects. Heck, I used my dad's control systems textbook from the 60s to learn. My fear is that publishers will start charging people on the back end of this for more specialized textbooks that are more typical in third and fourth year courses or specialized graduate courses. So, free textbooks for the two-year community college crowd, but $500 textbooks for process control of chemical reactors and digital signal processing.

The real underlying problem here is that student loans are the only type of debt that can't be discharged under bankruptcy, and that has created a moral hazard for post-secondary institutions to accelerate their costs. Tuition has also greatly and disproportionately increased in cost because students can get mortgage-like terms for their student debt, but institutions don't have any responsibility to make sure they graduate or make money. Meanwhile, endowments, perks and expensive buildings keep going up on campuses with little marginal benefit to students. The cherry on top is the IMO bizarre cultural support in this country for post-secondary institutions from alumni and through college sports.

If you really want to solve the textbook crisis, solve the debt crisis in education and allow discharge of student debt in bankruptcy at the same time as you investigate the publishers for any type of RICO or antitrust activity. The system will take a few years to clean out, but the issue will eventually be solved. The best part will be that tuitions will eventually come down to sane levels again, although that will be at the expense of the administrators and faculty who are more concerned about pretty buildings and social justice than they are about academic and human progress.

While I respect what companies like Uber are doing, it seems they could care less about the existing rules and why they're even there. And I'm not talking about the artificial scarcity of the medallion system or taxi company monopolies or the lack of flexibility in for-hire transportation, because that does need to be addressed.

What I'm talking about is a company that repeatedly flouts existing regulatory framework because it wants to "revolutionize" for-hire transportation. Drivers don't have to undergo local training (e.g. London drivers who have to memorize the road system in London prior to licensure). Driver vehicles are not required to undergo commercial-grade inspections for safety. Drivers are specifically disallowed by Uber from purchasing commercial insurance for their vehicles, as Uber claims that they will insure passengers up to $1M per passenger. Either the Uber driver is in violation of state insurance laws because they don't have the minimum required insurance, or Uber is in violation of those same laws by not being a licensed insurer with all of the regulatory and reporting burden of an insurer in that state. Want to guess where that leaves an Uber passenger in a crash?

Even if we ignore all of that, now we come to the self-driving vehicle which, even with GPS, lasers and camera AI, has to match years of a trained natural neural network of the most complex organism known on this planet with tremendous amounts more context to make not only technical but ethical decisions and keep not only the passengers safe, but also the car they're in, other people's cars and property, and most of all other lives that are on the road.

It's not an impossible problem to bound to a certain acceptable level, but not within the timeframe that Uber hopes. When considering its fundamental underpinning is compromised by its ethics and its arrogance that is being challenged by governmental and non-governmental entities, and is subsidized by free-flowing VC money, I can't say that the prediction of the demise of Uber is unlikely.

This is not as straightforward as you would make it out. Lawsuits and other tactics to slow or stop development are a known problem in the state of California. Holding up a dedicated left turn signal for a year to a beach because of an environmental impact is lunacy, and that is how the authorities saw it.

When such actions create artificial distortions and impact others negatively, their right to impose such problems on others stops.