As much as people seem to clamor for various forms of privacy protection the data shows they only care about it when prompted with questions. People are readily willing to give up privacy for small rewards and don't want to bother with the various protective measures already in place. There is nothing any law can do to really enforce data privacy when consumers don't find that privacy valuable enough to vote with their feet or use existing privacy controls.

There are really two types of `privacy' (often it's more about public but not readily discovered information) violations possible.

1) Security breeches by hackers or data theft by employees.

2) The sharing of personal data with institutions/people the user would object to viewing that information.

There is little regulation (perhaps government supported security information/response/prosecution centers could help) of companies can do about hackers or data theft. Sure, you can fine companies for data breeches and force publication but this creates an unfortunate incentive for companies not to discover security breeches. A well designed law would impose increased penalties for breeches exposed by outside agencies, e.g., law enforcement but even this law would create incorrect incentives for the current executives whose interests are still likely to reduce spending on discovering breeches in the hope that the bad news won't come on their watch.

Besides, I'm highly skeptical that poor security would be remedied by even larger financial incentives.

It's not even clear if such remedies are even desirable. A better law would simply demand appropriate compensation for people harmed by leaked credit cards and the like and leave it up to the companies (and consumers) what level of security is appropriate. Sure, we would be much safer if we replaced credit cards with fancy cryptographic two factor authentication but the costs in convenience and money would far far exceed the costs of making people whole from credit card theft.

This leaves the 2nd issue. The problem here is that the difference between desirable functionality and privacy violations here depends on the user's preferences. Does the user value getting to see free TV episodes more than the cost of having their viewing history shared with advertisers? What about discounts on medical products for similar sharing?

Sure, the law can require all sorts of consent and legal hoops to jump through but as long as people view actually making these calls as too burdensome to warrant real thought/action all you end up with is annoying privacy policies and click through agreements no one reads.

While popular with voters who think they care about privacy as long as they aren't willing to seriously consider it in their consumer choices (evaluating for themselves how seriously a company is committed to protecting their information from inappropriate revelation) such laws are likely to impose more burdensome regulatory costs than benefits to the consumer.

I appreciate the correction but it's not quite what I was trying to say. I realize I was very unclear since reading what I wrote it sounds like I'm just saying there is no hardware/API separation which is not what I meant. I know cell phones don't have software radios and have separate chips for their cellular radios.

My claim was that the firmware on the cellular radio doesn't incorporate any seperate cryptographic signature verification thus allowing any OS level code to modify said firmware or execute commands that might cause unwanted interference (but I take you to be saying the later doesn't occur).

Anyway I could well be wrong here but I'm curious about the answer so figured I would respond and ask someone who knew more.

Well depending on how the WiFi is setup that may or may not matter.

There is some excuse for this differential treatment.

On PCs there is a well defined BIOS/EFI interface that makes it difficult to cause hardware havok. Moreover, if you muck with low level interfaces and do something wrong you just ruin your computer.

Unfortunately, as I understand it phones/tablets don't have a nice seperation between the OS and the code which manages the cellular and wifi radios. Since the device manufacturers wish to enable new radio transmission patterns (or international support) via updates that run at OS level this means that installing an unapproved OS could potentially cause unwanted interference. Practically this probably isn't a big deal but if the hardware vendors don't at least try to lock out alternative OSs they may be subject to legal action (failing to ensure that the intended functionality of the device doesn't cause harmful interference).

Of course the right answer is to either create a level below the OS protected by cryptographic signatures on the updates or to clarify the law regarding consumer software that changes the behavior of the radio to create harmful interference. However, given the lackluster demand for this feature it's probably cheaper just to lock the device down. There is also a concern about enabling tethering since sadly requiring people to pay proportional to bandwidth used turns out to be highly unpopular.

Realistically I doubt the device manufacturers care if you don't use their OS. Unlike jailbreaking which is attractive to many people this will only appeal to the really hard core geeks (and even they may dual boot and buy from the walled garden anyway). Maybe someone can do everyone a favor and define a nice secure layer below the OS so it's easy for the companies to enable dual boot without issues.

You sue the agency. This is how the overbroad application of wetlands regulation beyond the 'navigable waters of the USA' was overturned.

Generally agencies get a great deal of deference in creation and application of their regulation but whether that extends to interpratation of the underlying authorizing act is less clear. In other words you have no chance in court challenging a DEA ruling that crystalized iodine is a meth precursor no matter what the facts provided the law gives them the power to enumerate precursors by regulation. If they are genuinely overstepping the power granted by the law rather than making unwise determinations it's more feasible.

Yah but not one of particular use to terrorists as Iodine tri-whatever is too unstable to make a useful explosive. You start making large batches and it will go off randomly while drying or large parts may fail to detonate.

It's much less of a public safety threat than a gun. The expected harm caused by a man with a pistol far exceeds that of a bomber with this stuff.

How long do you think it will be before some people start using these as a way to exchange child porn or some idiot thinks it would be funny to upload kiddie porn on all the USB sticks and the cops come in and shut the thing down?

Alright, so despite doing a bit of reading on wikipedia I'm still pretty puzzled about exactly how container formats can work. Setting aside fancy features like user menus and things like chaptering it seems to me the primary purpose of a container format is to do two things.

1) Define a format to multiplex many different data streams, e.g., allow packets in the audio stream, video stream, subtitle data to be interleaved so the right data can be available when needed (putting all the video stream first then the audio stream would be a bad idea).

2) Provide synchronization information to let arbitrary video, audio, and subtitle formats coordinate their display.

----

Now 1 seems relatively straightforward. What confuses me is part 2. I mean if we were encoding video as a simple list of pictures and audio in pcm this would seem straightforward. Each packet encoding a video frame gets tagged with the frame number and each audio packet gets tagged with the frame number it should be played with.

However, how does this work given the fact that to display frame 10034 the video codec may need to use information from frame 10000 and similarly with the audio codec. So if I want to jump to frame 10034 the player needs to know to look back at the info for frame 10000. I mean I can think of various ways this might be done but they all would seem to require particular knowledge about how the individual streams work.

Could someone explain how these work or give me a pointer to a good explanation?

Thanks

No, that's not true. Version 2 can simply define a new bit to indicate whether it's version 2 or later.

The real problem with this optimization is it's effect on later versions.

Say one eventually moves to version 12 and each version along the way defines it's own flag. That mean's you've used 12 bits and some very nasty decoding logic to indicate you are a version 12 format when you could have just reserved that one byte and used a case statement.

I mean the annoyance and difficulties created by using a next version flag (and the poor scaling behavior with higher version numbers) is a high price to pay to save a couple bytes in a multimedia codec.

This is a contract with real money and the signatures are exchanging something of real value (their knowledge of the locations). It's a nice clean example of a contract, nothing weird or fishy. Moreover, contract law doesn't work like computer code. You simply include a clause requiring them to accurately disclose their knowledge about such and such a balloons location. If they know or have reason to suspect that it was a fake and they don't share they are violating the contract.

This (unfortunately) reminds me of the ontological argument and similar examples of bad reasoning that manage to avoid being laughed out of the room because they dress themselves in a false shroud of logical rigor.

One version of the ontological argument procedes by defining god to be the being which has the maximal amount of good qualities and continues from there. Now there are other problems with this argument but the giant gaping fallacy is that this simply isn't what people mean by god, a point that wouldn't be lost on anyone if you stripped off all the pretense of extreme rigor and just said, "Hey, something has to be the best thing."

I'm a big fan of using logic to demonstrate that our convential views are incoherent. Indeed, many of the issues mentioned here beg for such a treatment but disguising your hidden assumptions by pointless trapings of rigor (and I'm mathematical logician so I like rigor) gives those of us who actually want to reason about these situations a bad name and enlightens no one.

Grr...I mean just consider the free trade example. It sounds as if he is delibrately trying to slip past the reader that our goal is not to maximize the net inflow of 'dollars' to the US not to mention the existance of inefficent equilibriums. I mean I think nearly every protectionist sympathizer I've ever heard is being a total moron but you don't do anyone any favors by failing to mention that increased utility from trade may require transfer payments to compensate for the disparate impacts of trade. Maybe you oppose these on other grounds but if so you need to state the case. Ohh and BTW given the extremely strong evidence that many Chinese are eager to the point of breaking the law to get these 'sweatshop' jobs maybe the reviewer should try harder to believe that other things being equal they leave an individual better off on balance for taking the job. Perhaps by contemplating how much subsistance farming without modern medicine or convienences sucks.

And the god arguments repeat the same problems. Yes, it's interesting that adults treat religious beliefs differently than other beliefs but saying they don't believe in god doesn't accomplish anything. It just redescribes the situation confusingly. People still let their faith influence their attitudes on many policy questions (which they often also treat differently than beliefs about things they can affect). The ESP bit is even dumber. ESP, like most words, doesn't have a stipulative definition but rather is understood by something like prototype resembelance. It's like the word table, you know some things count and others (a bed) don't and evaluate weird new examples (three legged 2 foot radius stool) by their similarity. Besides, no one cares if 'ESP' exists, people care if people can read minds, remote view etc.. whatever you want to call it.

The only half-decent argument listed is the bit about free will. A better statement would be something like this:

Free will doesn't mean unpredictable/random. A person who heroicly rushes back into a burning building to save a trapped dog is exercising free will in that choice if anyone is even if they would make the same choice everytime you (exactly) replayed the situation. Indeed, if you rewound time and gave it another go and they acted differently that feels less like exercising free will. If free will makes sense then choices I make because of my charachter (how I see myself) surely count and not just choices which we might as well have left up to a coin toss. In other words it seems that what makes a choice free is that I get to select the outcome without outside dictation of the answer.

In other words for a choice to be free it must be possible for me to have acted differently, i.e., if I were inclined to select a different option then I could have done so. It doesn't require the absurd criterion that a free choice must be something that *I* don't determine, e.g., by being the sort of person who will race into burning houses. So now if we take the step of identifying ourselves with our brains we see there is no conflict. Had we wanted to make a different choice we would have had a different brain and hence would have acted differently. The key point is that determinism is really saying that *you*, embodied in a brain, in conjunction with the enviornment dictates the outcome while free will is saying that you make the choice. The only trouble arises because we aren't used to thinking of ourselves as brains so we imagine sortof floating outside the physical realm in which case if the physical world dictated the outcome we (not being part of that world) would have no choice in the matter. So once you are cool with yourself as a brain no problem.

Unless a technological approach using sattellites, automated analysis of flikr photos etc.. quickly wins out then presumably the ballons will end up being seen by some people aware of the contest. So suppose you have seen a balloon and wish to use that knowledge to claim part of the prize. What do you do?

Well what you want to do is find a group of other people who have located the other balloons and agree to pool your info in exchange for shares of the prize. Importantly nothing in the rules prevents you from drawing up legal contracts with other participants and if you are clever you might even be able to make delibrately feeding you false information into criminal fraud. So it seems straightforward to ensure there are big incentives to only agree to join such a coalition if you do indeed have valid data. Moreover, your claimed data can be verified relatively quickly for low cost by paying close friends and associates who live nearby to go out and check.

In short it seems to me the optimal strategy is simply to hope you stumble upon a balloon and then creating a legally binding contract with others who claim to have complimentary information.

There is a crisis for journalists as a result of the sudden crash in their industry but that crash isn't the result of some horrible failure of the market for journalism. Just the opposite. The newspaper industry has hit bottom because the internet has made the buisness of reporting so much more efficient. I mean just thinking about the huge number of daily papers across the states carrying the same national and international news on print is enough to make one sick at the waste. Not only does it cost a great deal to publish a print daily but each of these dailies employs editors and layout people to format the same news availible anywhere in their particular style. Many of them even insist on hiring their own reporters even when it's obviously duplicated effort (say reviewing national movies/TV shows).

Once competition drives most local papers to focus on local intersts and everyone to publish online it will free up a quite substantial amount of money for real reporting. Though actually a lot of what journalists call real reporting is duplicated effort for the sake of status. I mean does it really help the public understand what's going on better to have 40+ journalists at the white house press briefings and who knows how much AV equitment? If they just sent over a single camera crew and agreed on a way to pick questions there would be no harm to the quality of reporting. Much of this is just done because historically that behavior signaled prestige and seriousness in the news industry.

I don't think the newspapers are doing anything wrong. But when technology lets you accomplish the same job with disruptively less total effort (delivering news to the nation) many people are going to lose their jobs and most of the companies in that industry will go out of buisness. I feel sorry for the people with careers in the industry but I think there is every reason to believe that after things settle down there will be just as much investigative reporting and important journalism. There will just be less redundancy and a more efficient use of reporting resources.

While obviously unnecessery from an addressing point of view it does offer some interesting advantages in terms of security. In particular rather than bothering with expensive hardware memory protection one can simply rely on the extreme sparseness of memory usage to provide process isolation...at least if you do other things right (randomize layouts etc..)

Of course most people say they don't want targeted ads. I mean it's like asking "Would you like your taxes to be raised." Of course people say no because other things being equal people would prefer zero taxes and almost no ads. Of course if you asked whether you would prefer the government increase the gas tax or income tax, or whether a tax hike would be a worthwhile price to pay for universal health care you would get very different answers.

In this sense ads are quite similar to taxes. We would prefer to get our government services for free and our media without any ads but since that's not possible question that matters is what tradeoffs you would like to make. I suspect if you asked people whether they would prefer to get tageted advertisements, a substantially larger number of untargeted ads or no longer get free webmail and the like I think you'll find many people take the first option.

Also it's well documented that people seem to only really care about privacy/psuedo-anonymity/targeting when someone asks these kind of ominously phrased questions. I mean all the 'experts' go around saying vague ominous sounding things about losing privacy/etc so you ask people a question about privacy/etc.. and they think, "Hmm, I don't really understand what all the fuss is but it sounds kinda ominous and scary...you know I probably should be more responsible and worry about these issues." So your never finding out how much people actually mind losing their privacy and the like, they may not even notice. All you're really measuring is how much vague scary statements make people think they should worry about something.

Ohh, and the fact that so many experts seem to take these concerns seriously is no more useful. If you dismiss a danger like that rather than looking thoughtful and concerned you appear less serious. I mean it's the same reason doctors and other medical experts in the mainstream media never get up and say, "Pshhaw, marijuanna use by teens isn't anything to worry about," despite having uncontrovertable data that it is way less risky that many other accepted activities. Once society views something as posing a risk and takes it to be a serious matter then any role models who refuse to treat it as a matter of serious concern are considered irresponsible and blamed for any harm that might have been avoided by being overly cautious.

Some quick observations:

• First judges have generally applied balancing tests between free speech and publicity rights so this means that if this result stands EA would probably still have to shell out to Madden for Madden Football but if you appear as simply a name, likeness and stats on a sports team you probably won't be able to win a lawsuit based on any kind of publicity right. In other words being depicted in the game and being plastered all over the box to move more copies are different things.
• Surely this is a minimal requirement to have anything like freedom of expression for video games that public figures can be depicted as inhabiting historical/counterfactual situations without requiring any kind of licensce, permission or payment. Otherwise video games would be unable to comment on any issues of public concern or historical interest unless they made all the participants happy with their depiction. If you make an LAPD game you shouldn't need to get OJ's permission to help pursue him in the "high speed" chase.
• More broadly I feel that any kind of publicity right for public figures going beyond false endorsement style laws is incompatible with the first ammendment. If I want to write a novel depicting Bill Clinton's secret life as a butt kicking comando only good taste and not the law should stop me. Video games deserve no less protection and taken together these two observations essentially eviscerate publicity rights.

To put the point another way: Just because you participated in historical events doesn't give you the right to influence the (fictional) depiction of these events. If you don't like your likeness appearing in the media then don't let yourself become a public figure.