I don't know about Canada, but here in the USA the so-called conservatives and liberals are both populists. From your comment, it sounds like it's true up there, too.
If only. Populists would mean that they're actually in some way responsive to the population; in reality, both parties pay lip service to the citizenry while their true loyalties are to the corporations that bankroll their parties, and effectively get them elected.
This raises an interesting point.... how loudly would the American government be screaming if a US citizen was arrested in Britain for doing something which was perfectly legal in the US but which affected UK citizens and was against their laws???
I don't know... if they were sending out spam, I'd prefer that they be quickly extradited to whatever third-world country still practices breaking-at-the-wheel.
I'm honestly not as concerned about the CPUID as I am about software pulling the MAC address and disk serial numbers. While I can think of some legitimate reasons for userland software to need to know about the processor it's running on (for technical/performance, and not identification, reasons, i.e. identification of the presence of certain features like MMX), I can't think of any good reason why it would need to know uniquely identifying information about other hardware.
Obviously the operating system, in particular the network stack, needs to know the MAC address at some point, but this information shouldn't be passed on to potentially untrusted applications. By running the operating system on the bare metal I am implying that I trust it (most people don't think about it too hard, but you'd better trust whatever's running in Domain 0, because it can do whatever it wants and only report to you what it wants), however I don't necessarily trust all my userland applications to the same degree.
I think it's just common sense that only trusted applications should have access to serial numbers or other pieces of information which can be pieced together to create a per-machine ID. Per-machine is much closer to a per-user ID than an IP address (particularly with the heavy use of NAT), and so it could easily be used to track a user, or prove later that a particular user did something and break anonymity.
Obviously, there should be a mechanism for applications that need it, to get the Ethernet MAC, CPUID, drive status and serials, but those mechanisms should be controlled and limited only to applications that are authorized by the user as having a bona fide reason to get them. To let all software pull up this sort of information automatically, relinquishes a lot of control from the user, to potentially untrusted or untrustworthy pieces of software, and that, I think, is a fundamentally bad idea.
I think a better term would be "deprecated." It's not "obsolete," because a lot of people still use them. However, it's obvious that DirectTV is moving away from them, and would like people to move to newer boxes, and at some point in the next few years, their usefulness will decrease substantially.
Very interesting. I hadn't thought about it that way, but if you allow a requirement that the user has to take multiple photos of the same scene, from slightly different positions (particularly if you require that the target to be visible and in the same position in each of them) then you can start to do some neat stuff. Essentially you could use the target (or other object visible in both photos, as you noted) to find the position of the camera and thus know how far apart the two (or more) images were taken from. Then you could start to construct a stereoscopic model.
I have a feeling that this product is not that complicated, just a hunch, but it would be neat if someone took the concept and ran with it. I could see a lot of applications for some software that let you put a target down near an object, and then take a few photographs of it, and would give you an accurate (or even approximate) 3D model of the surface of the visible portion.
i will be calling Dell ASAP to see if I am affected.
Um, and why would you expect them to give you a straight answer? They'll probably just play dumb and say they've never heard of the problem. (Which will probably be true, at least for the drone you'll be talking to.)
Get out a voltmeter and test it; that would seem to be the easiest solution, and less likely to lie to your face than some Customer Service rep. Probably faster, too.
Until a problem like this becomes terribly public -- and by this I mean more public than just being covered on some technology websites -- I suspect Dell will deny it, except in cases where people absolutely insist that they have a problem, and demand a replacement. In those cases, they'll get a replacement machine just to shut them up.
So I'd just get out the old multimeter, measure the AC voltage from one of the chassis screws to the nearest good ground, and if it's more than a few millivolts, call Dell and tell (not ask) them that you need a replacement unit.
Maybe you missed the part of the article where they said this was a test flight. As in, they were testing the plane. The choice of route was a stunt, granted, but if they hadn't flown a great big "GV," they probably would still have done the test flight, and just flown around in a circle, or some other arbitrarily-defined pattern. It's just that flying in this particular pattern got them some extra press, so why not?
Calm down a little before you flip out, next time.
It doesn't seem like it would be really too hard, if the software is just doing what I think it is.
The hardest part is just picking out the target from the photo. In most interior scenes, the target they're using would probably work pretty well (it's a white square with heavy black edges) although it seems like there are some backgrounds where locating it might be a problem. But there are, if I'm not mistaken, some OSS efforts to do things like automatic facial recognition, and that's a much more complex problem than picking a black-on-white box out. (Particularly if the center of the box is reflectorized, so that it's always 255/255/255 when a flash photo is taken.)
Once you've located the target, knowing it's actual size and how many pixels wide it is in the image, then you can let the user pick any two points elsewhere on the image (which must be in the same plane as the target, and basically perpendicular to the camera's film/sensor) and tell them how far the points are apart in reality. It's just multiplication at that point.
If you look at OSS image-processing software, there are applications around that do much more complex stuff than this: Hugin, and Panorama Tools (the latter are what really do the heavy lifting) come immediately to mind. Compared to joining and sewing a panorama, this kind of measurement seems pretty easy, unless I'm missing something critical.
If I was recommending features for a measurement product, I think the key would be not to limit it to a particular target. Sure, a few printable targets, similar to the one used in TFA's commercical product, would be good for measurement of rooms and houses, but it would also be nice to use smaller things that are typically used for scale in macro photographs. E.g., dollar bills, quarters, width of a pencil, etc. Those would be tougher to automatically recognize, and would probably require some prompting by the user in order to pick out, but would probably appeal to a wider variety of users. Who hasn't seen an eBay photo and wondered what the exact dimensions of something were?
Looking at that photo, I'm not buying that it can measure all those distances from a single photo. I think there is some advertising hyperbole going on here. I get that you could measure all those distances and dimensions, using multiple photos -- one each of every flat surface, moving the target each time so it's the same distance from the camera as the surface being measured -- but I don't think it would work from a single photo.
The only way you could measure everything from a single photo like that, would be if the camera was stereoscopic, or had some other form of depth perception. Otherwise, as you noticed, there's no way for it to know that the window that's closer to the camera is not really bigger than the garage door that's further away.
But since it is not 100% accurate, then its trash.
You do know that's impossible, right? I could use a laser interferometer, and determine the distance between two objects down to a fraction of a nanometer, and it would still not be "100% accurate."
They should make a $1299 "Professional" version, quick. It'll be the exact same product as the $99 version, just in a different color, and with a printed, spiral-bound manual.
It's definitely possible to under-price your product if you're not careful. Actually, having a $1299 version might even help drive sales of the $99 version, because people would perceive the $99 version as a sort of 'deal,' as in "hey, for $99 I'm getting 60% of the features of the $1300 version! That's great! I'll take three."
Yes, actually if you read my comment, I even mention Anycasting. However, the page I was pointing to, lists the geographic locations for the various root servers. Some of them have multiple locations, behind a single IP address, routed via Anycast. The others have only single locations (which may or may not imply multiple physical servers), making them somewhat easier to DDoS, since a world-wide zombie net would be concentrating it's traffic towards one server. On the geographically-distributed "servers" (with Anycast), each zombie would only hit the machine closest to it, so it's harder to take down.
Not sure if I'm entirely understanding your point.
In the example I was talking about above, I was scanning a bunch of old slides, which we can assume the copyright holders have authorized me to scan (hence, why they gave me boxes of slides and asked me to scan them onto the internet-thingamajiggy). So it's not as thought it was an unauthorized derivative work, if in fact it was a derivative work.
My question is basically, would there be two copyrights: e.g., (c) 1939, Grandma (for the original photo), and (c) 2006, me (for the new scan), or just the former.
I think, based on the court case that another respondent cites above, that it would be just the former case; the scan, if it's purely a refixing of the original work in a new medium, without much in the way of creative "work," only the original copyright would apply. So it would be (c) 1939, not 2006 -- a big difference in terms of when the digital copy would hit the public domain (the 2006 one wouldn't end up there, unless I ceded it earlier, until 70 years after my death, while the 1939 one may already be there, or may be there soon -- the law is a bit byzantine on post-1923/pre-1978 works).
So I don't think that the authorization to make the derivative work was really at issue here, or I was assuming for the purposes of argument that I had such authorization, given to me by implication in the request to scan the photos.
They are most likely using this in combination with other more or less 'unique' things to identify a specific machine. It wouldn't surprise me if after this some people would do a more in-depth analysis of their code and find out that it also reads the serial number of the harddrive and gets the MAC address of the Ethernet adapter.
This seems pretty logical. Since they got rid of that hackneyed scheme a while back to give each processor a serial number (wait -- did they get rid of that?), some sort of hash of the BIOS memory, plus the Ethernet MAC, plus the HD serial number, all concatenated together, is probably as close to a unique identifier as you're likely to find on a "per machine" basis.
That said, it doesn't make me feel any better. I wasn't a fan of the processor serial number concept, and not just because it was a serial number in the processor; there were serious privacy concerns with any uniquely identifying, per-machine serialization concept, and that's true whether it's a dedicated number that's being used, or some sort of combination of semi-unique factors.
It's just one more piece of information, sitting in a database somewhere, that could be subpoenaed and used to generally cause trouble. Particularly given how close-mouthed the Skype people are about how their network actually operates (e.g. their alleged encryption, peer to peer communications), I'm not ready to run right out and trust them.
I wonder if it would be possible to run Skype in a sandbox, where the information it's fed could be carefully controlled? On further thought, I wonder what happens when you run it in VMWare or Wine? Do they actually pass information about the hardware up to guest applications? It seems like this behavior would be one that the user should be given an option about, at the very least; I can only think of a few programs who have any reason to be getting the drive serial number, or the Ethernet MAC address, and for the most part they are not userland apps.
It's not like they haven't figured out the whole failover/fault tolerance thing.
That's kind of the point here, actually. Several of the root servers do not have any redundancy. You can see the list at http://www.root-servers.org/. In particular, the A, B, D, E, G, H, and L servers have only a single location a piece.
F, I, J, K, and M, on the other hand, are heavily redundant and have multiple geographic locations, routed via Anycast, so a single client only "sees" the server nearest to them. This makes them difficult to DDoS, because a zombie in S. Korea pinging the J server would be sending packets to the server in Seoul, while one in California would get the one in Mountain View.
What's odd, looking at the list, is that anyone operating something as critical to the internet infrastructure, wouldn't develop some geographic and systems redundancy; unfortunately, I suspect that the government agencies in particular tasked with these responsibilities probably don't keep it at the very top of their priority lists when allocating resources and funding.
I'm sure it is considered in conjunction with other activities. Like if you're otherwise suspected of dealing drugs, use of encryption is "evidence" that you were up to something. Not that it is any more solid, but at least it isn't saying that encrytion, in and of itself, is illegal.
I think the reason people are unhappy with the opinion is because it might cause people to not use, or otherwise discourage the adoption of, encryption, for fear that it would "look bad" if they ever got in trouble.
E.g., an example that might be closer to most people's hearts would be an IRS audit. Suppose you get audited, and honestly haven't done anything wrong, but get stuck trying to clear your name (because this is how the IRS works: you need to demonstrate to them that you're legit). I could easily see situations where people would be afraid that if their computer was seized and GPG was found, that it would be used as evidence that they were "hiding something" regardless of what was actually going on. So instead, they just begin to fear the tools.
This is a self-fulfilling prophesy: as more people fear encryption tools, less innocent people will keep them around, and they'll become more and more the domain of people with something to hide. Then they gain more of a reputation, furthering the cycle.
The uploaded version would clearly be a derivitave work, but I'm guessing that putting it into another tangible form would mean it's automatically copyrighted right then even if it wasn't originally.
This is actually a fairly interesting question, and IMO an important one. I'm not sure I share your conclusion that the uploaded version is a new work, though. Although it certainly could be, if you changed it (say, retouched, or even just cropped it), a straight scan+upload probably wouldn't be original enough.
It's an interesting question, because I recently scanned hundreds of old family photos and slides. Many of them, provided Congress stops extending copyright indefinitely, will be out of the original photographer's copyright relatively soon (as in, probably within my lifetime -- copyright, like geology, has its own relative time-scales). However, if the act of scanning the photo automatically makes a new work, then it's under copyright for another 120+ years, beginning 2006. Not really a concern to me, since I'd be the copyright holder, but of concern to a hypothetical other party who might want to use them.
I suspect that simply scanning a photo, in its entirety, and uploading it, does not represent enough of a creative act to warrant a renewal of copyright as a derivative work. Essentially, all that is happening, is that the older work is being "format shifted." However, if you were to do any type of alteration to the original photo that wasn't totally automatic, even something like color correction, I could see an argument for protection on the grounds that it was a creative act.
I think that's severely oversimplifying, because rewriting the system to take only one system call would certainly result in more bugs, no?
And I'm not arguing with you in the slightest. However, in TFA, that's pretty much exactly the assumption that the author makes, in order to start off with the diagrams of system calls, and from there get to the conclusion that IIS/Windows Server is "more secure" than Apache/Linux.
I wasn't really making any commentary on that premise, because I'm hardly qualified to, except that if you wanted to undermine the article's conclusions, the premises seem like a good place to start...
It's hard to say, but two immediate problems come to mind. First, and in my opinion most likely, is that they have in place agreements with the major record labels that involve giving the same treatment to all music sold via the iTMS, so that it all has to be FairPlayed. This strikes me as pretty likely, and something that the record labels would insist on; they must realize that online distribution closes a lot of the gap between a small record company, and them, and obviously they want to avoid direct competition as much as they can. So they'd want to suppress anything that a small, independent company could use as an advantage. Hence, demand that Apple apply the same "protection" to all iTMS-sold music.
The other problem, which isn't exclusive of the first, is that the DRM isn't applied once to each song in the store when it's being added to the database, but added at the time of sale (necessary because it's encrypted with a key that's specific to each user), somewhere on Akamai's servers. It might be difficult to the point of being cost-prohibitive to designate one song as being DRM-free, if the system wasn't designed with that capability from the beginning.
I've noticed that even songs that are free downloads (promo songs, etc.) from the iTMS have FairPlay placed on them, even when you can go to the band's or label's web site and download it as an MP3 (so it's obvious that the label doesn't care if it's protected); this makes me suspect that one or both of those problems exist.
It would probably be trivial for Apple to turn off DRM completely, for all the songs in the Store, but difficult both legally and technically, to disable it for just one.
(I'm not trying to sound like too much of an Apple apologist here, to be frank I think the iTMS is an abomination and I wish Apple had stood up to the record companies when they were screaming about the iPod and contributory infringement a few years ago, and remained a purely hardware company and stayed out of the music-retail business; however, at the time creating the iTMS was the best way of eliminating accusations of the iPod as a "piracy machine." It's ironic that Apple's own creation, created to soothe the record companies, is now coming back to haunt them. Well, that's what you get for dealing with the devil.)
Okay, poor choice of words. I just meant that they refuse to license it for use in other playback devices.
And why should they? Steve Jobs is obviously a smart guy; things he's said and written elsewhere make me think that he understands the inherent problems behind DRM.
In short, DRM doesn't work. It works, sort of, only by keeping the mechanisms out of sight, and changing them all the time, as people catch on and figure out what's going on "behind the curtain."
The more people you let see behind the curtain, the harder it is to make work, and keep working, even in the shoddy way that it does currently. Licensing means that specifications and technical documents need to be written, and such documents can be leaked (and are far more likely to be leaked when they're being sent to some licensee in Europe, than kept within a particular technical working group inside Apple US). So if Apple licensed out FairPlay, it would mean that FairPlay would get broken more often, and they would have to dedicate more effort to fixing it, and those fixes would be harder to roll-out, because there would be more users, and multiple online music stores, run by various licensees who might take their responsibilities for updates more or less seriously, etc. etc.
DRM isn't a single technology that you can sell. It's not a word processor. It really is defective by design; that's not just some dumb slogan -- that is reality. Anyone who buys a DRM system, thinking that it's a product they can just use, and then forget about, is a fool. A DRM system is an arms race. It can only work when you're committed to throwing a lot of programmers behind it; programmers who are constantly shoring it up, as people pull the bricks down from the outside. And the work that it takes to sustain is directly proportionate to the number of people who are working to crack it.
Licensing out FairPlay would be a losing proposition for Apple on all fronts. It would force them to lose revenue from the iTMS, which isn't exactly a huge profit center anyway -- as others have pointed out, Apple makes a lot more money on an iPod than they do on the average user's iTMS purchases. Plus, it would mean that they would have to spend a lot more effort constantly fixing FairPlay, and it would create a huge logistical problem -- how do you roll out those fixes to users who may be using some licensee's music store? If Apple doesn't keep FairPlay's facade of security up, the music labels will use it as a bargaining point in negotiations, but they'll be dependent on their licensees, who they don't have total control over, in order to maintain that facade. It's a lose-lose for Apple.
Personally, I don't think Apple will ever license FairPlay. I think they'll pull all DRMed music from the European market, and close the iTMS there, before they'd open the can of worms that licensing would entail. Exactly what would happen at that point is anybody's guess, but there are a whole lot of iPod-owning Europeans who probably want some type of online music store, and Apple is pretty good at PR. They might be able to turn it into some sort of a victory against the governments mandating the interoperability, or against the music labels who won't sell DRM-free music. Or it might backfire horribly and cause a lot of people to run out and buy non-iPod MP3 players in order to use competing online stores (though I doubt it; I don't think that the presence or absence of an online store is a huge selling point of most music players, except those linked to subscription services like Napster).
This is slightly off-topic, but a while back I got interested in OpenVMS, and VAX stuff in general. (I started doing some research because I thought I was going to get stuck doing some turd polishing of old mainframe software, but it never materialized. But by then I was just interested.) Even in hindsight (given that I think we can agree that UNIX-derivatives seem to have gained traction over VMS), it's extremely difficult to find any sort of rational comparisons of VAX/VMS and its architecture and design paradigms to that of UNIX. Whenever someone asks, the response is basically "don't ask, you don't want to start that." Nobody wants to talk about anything that might invite UNIX/VMS comparisons, because it will cause flamewars -- even though such a discussion, at this point, might be interesting and productive. (There are so many people around who aren't familiar with VMS, or anything other than Windows and UNIX, that any perspective besides those would be worthwhile.)
At any rate, it struck me as interesting, because sometimes it's easy to assume that Windows/Linux (or Windows/Mac, or Windows/something) is the first Great OS War. But people have been getting emotionally attached to operating systems, probably as long as they have existed; and ever since, it has helped quash rational discussion, both through flamewars themselves, but also because of self-censorship that occurs, in order to try and prevent arguments.
Actually, his conclusion contains a far more useful test, although it does boil down to common sense:
The difference between these cases is simple: determinism. In the case of the encryption software, the outcome is deterministic. Knowing everything about the mechanism doesn't compromise the security of the outcome. In contrast, for antivirus software the system is heuristic. As such, some things benefit from disclosure, and some things don't. In these two cases, it's obvious. Unfortunately, that's the exception, not the rule. The problem is that many systems contain aspects that are heuristic and aspects that are deterministic.
In essence, the question is to ask whether closing the source really results in any increased security; in the case of DRM systems (his example), it does, because they are broken by default and thus knowledge of the 'algorithm' allows the system to be cracked.
Personally, I would argue that such 'heuristically secured' systems are broken by default, and that there are good reasons why generations of computer scientists have insisted that security through obscurity is meaningless. The "security" provided by such heuristics are of value only to marketing and legal departments, they are not and should not be confused with the security offered by 'deterministically secured' systems (e.g. cryptography is his example). Saying that an application is "secure," when it depends on an attacker not knowing how it works, borders on unethical false advertising.
I don't know how they're doing this in practice, whether they're just sending people out with regular handheld digital cameras or what, but it would probably be possible to rig up a nondescript panel van with side-view cameras, and just drive up the street photographing everything on both sides. (Or, for better results, everything on the right side, and then drive up one side of the street, followed by the other.) If you had a very good GPS receiver in the van, you could geotag each photo, and then crop them as a batch later on, for each house or building on the street.
What would probably be even better would be to use a progressive-scan video camera for image capture, so that you have a continuous feed of images, and then you don't have to worry quite so much about having one house cut between two photos. (Alternately you could probably sew the images together into a continuous linear panorama, but that might give mixed results.)
You might still get shot at in some areas, but it would probably be a lot lower risk than just walking around.
On the other side of the pond, ex-felons can't buy guns & have to do some paperwork to get their voting rights re-established... That's pretty much it. Everything else is social stigma.
It's a lot more than that. They are effectively barred from any 'positions of trust' for the remainder of their lives, unless their record is expunged, which is unusual. This pretty much takes them out of any government or civil-service position (including the military), out of jury duty, and out of most teaching positions in public schools. This is just the stuff that they're barred from by law.
By that "social stigma," they're basically unemployable in all but the most menial jobs, and even in those they're basically un-advanceable. I don't think you can even get a fast-food job with a felony on your record, because they're suspicious that you'll start stealing from the till.
Effectively, they're the modern version of "untouchables"; they become members of a permanent underclass unfit for legitimate occupations. I don't have a whole lot of pity for them, frankly, but it's not hard to see why the recidivism rate is so high. But that creates a feedback cycle; the public expects ex-cons to fall back into crime, thus they're not trusted, thus they can't find legitimate occupations, and thus they do exactly what's expected of them.
Short of 'rebooting' the system by summarily executing anyone with a felony conviction, or other similarly drastic methods, I don't think there's any easy way to break that cycle.
Slashdot Mirror
I don't know about Canada, but here in the USA the so-called conservatives and liberals are both populists. From your comment, it sounds like it's true up there, too.
If only. Populists would mean that they're actually in some way responsive to the population; in reality, both parties pay lip service to the citizenry while their true loyalties are to the corporations that bankroll their parties, and effectively get them elected.
This raises an interesting point .... how loudly would the American government be screaming if a US citizen was arrested in Britain for doing something which was perfectly legal in the US but which affected UK citizens and was against their laws???
... if they were sending out spam, I'd prefer that they be quickly extradited to whatever third-world country still practices breaking-at-the-wheel.
I don't know
I'm honestly not as concerned about the CPUID as I am about software pulling the MAC address and disk serial numbers. While I can think of some legitimate reasons for userland software to need to know about the processor it's running on (for technical/performance, and not identification, reasons, i.e. identification of the presence of certain features like MMX), I can't think of any good reason why it would need to know uniquely identifying information about other hardware.
Obviously the operating system, in particular the network stack, needs to know the MAC address at some point, but this information shouldn't be passed on to potentially untrusted applications. By running the operating system on the bare metal I am implying that I trust it (most people don't think about it too hard, but you'd better trust whatever's running in Domain 0, because it can do whatever it wants and only report to you what it wants), however I don't necessarily trust all my userland applications to the same degree.
I think it's just common sense that only trusted applications should have access to serial numbers or other pieces of information which can be pieced together to create a per-machine ID. Per-machine is much closer to a per-user ID than an IP address (particularly with the heavy use of NAT), and so it could easily be used to track a user, or prove later that a particular user did something and break anonymity.
Obviously, there should be a mechanism for applications that need it, to get the Ethernet MAC, CPUID, drive status and serials, but those mechanisms should be controlled and limited only to applications that are authorized by the user as having a bona fide reason to get them. To let all software pull up this sort of information automatically, relinquishes a lot of control from the user, to potentially untrusted or untrustworthy pieces of software, and that, I think, is a fundamentally bad idea.
I think a better term would be "deprecated." It's not "obsolete," because a lot of people still use them. However, it's obvious that DirectTV is moving away from them, and would like people to move to newer boxes, and at some point in the next few years, their usefulness will decrease substantially.
Very interesting. I hadn't thought about it that way, but if you allow a requirement that the user has to take multiple photos of the same scene, from slightly different positions (particularly if you require that the target to be visible and in the same position in each of them) then you can start to do some neat stuff. Essentially you could use the target (or other object visible in both photos, as you noted) to find the position of the camera and thus know how far apart the two (or more) images were taken from. Then you could start to construct a stereoscopic model.
I have a feeling that this product is not that complicated, just a hunch, but it would be neat if someone took the concept and ran with it. I could see a lot of applications for some software that let you put a target down near an object, and then take a few photographs of it, and would give you an accurate (or even approximate) 3D model of the surface of the visible portion.
i will be calling Dell ASAP to see if I am affected.
Um, and why would you expect them to give you a straight answer? They'll probably just play dumb and say they've never heard of the problem. (Which will probably be true, at least for the drone you'll be talking to.)
Get out a voltmeter and test it; that would seem to be the easiest solution, and less likely to lie to your face than some Customer Service rep. Probably faster, too.
Until a problem like this becomes terribly public -- and by this I mean more public than just being covered on some technology websites -- I suspect Dell will deny it, except in cases where people absolutely insist that they have a problem, and demand a replacement. In those cases, they'll get a replacement machine just to shut them up.
So I'd just get out the old multimeter, measure the AC voltage from one of the chassis screws to the nearest good ground, and if it's more than a few millivolts, call Dell and tell (not ask) them that you need a replacement unit.
Maybe you missed the part of the article where they said this was a test flight. As in, they were testing the plane. The choice of route was a stunt, granted, but if they hadn't flown a great big "GV," they probably would still have done the test flight, and just flown around in a circle, or some other arbitrarily-defined pattern. It's just that flying in this particular pattern got them some extra press, so why not?
Calm down a little before you flip out, next time.
It doesn't seem like it would be really too hard, if the software is just doing what I think it is.
The hardest part is just picking out the target from the photo. In most interior scenes, the target they're using would probably work pretty well (it's a white square with heavy black edges) although it seems like there are some backgrounds where locating it might be a problem. But there are, if I'm not mistaken, some OSS efforts to do things like automatic facial recognition, and that's a much more complex problem than picking a black-on-white box out. (Particularly if the center of the box is reflectorized, so that it's always 255/255/255 when a flash photo is taken.)
Once you've located the target, knowing it's actual size and how many pixels wide it is in the image, then you can let the user pick any two points elsewhere on the image (which must be in the same plane as the target, and basically perpendicular to the camera's film/sensor) and tell them how far the points are apart in reality. It's just multiplication at that point.
If you look at OSS image-processing software, there are applications around that do much more complex stuff than this: Hugin, and Panorama Tools (the latter are what really do the heavy lifting) come immediately to mind. Compared to joining and sewing a panorama, this kind of measurement seems pretty easy, unless I'm missing something critical.
If I was recommending features for a measurement product, I think the key would be not to limit it to a particular target. Sure, a few printable targets, similar to the one used in TFA's commercical product, would be good for measurement of rooms and houses, but it would also be nice to use smaller things that are typically used for scale in macro photographs. E.g., dollar bills, quarters, width of a pencil, etc. Those would be tougher to automatically recognize, and would probably require some prompting by the user in order to pick out, but would probably appeal to a wider variety of users. Who hasn't seen an eBay photo and wondered what the exact dimensions of something were?
That link doesn't work (at least, not for me). I think it looks at the referer and won't let you deeplink to the image. You have to go through the blog to see it:w _to_measure_.html
http://jkontherun.blogs.com/jkontherun/2007/02/ho
Looking at that photo, I'm not buying that it can measure all those distances from a single photo. I think there is some advertising hyperbole going on here. I get that you could measure all those distances and dimensions, using multiple photos -- one each of every flat surface, moving the target each time so it's the same distance from the camera as the surface being measured -- but I don't think it would work from a single photo.
The only way you could measure everything from a single photo like that, would be if the camera was stereoscopic, or had some other form of depth perception. Otherwise, as you noticed, there's no way for it to know that the window that's closer to the camera is not really bigger than the garage door that's further away.
But since it is not 100% accurate, then its trash.
You do know that's impossible, right? I could use a laser interferometer, and determine the distance between two objects down to a fraction of a nanometer, and it would still not be "100% accurate."
They should make a $1299 "Professional" version, quick. It'll be the exact same product as the $99 version, just in a different color, and with a printed, spiral-bound manual.
It's definitely possible to under-price your product if you're not careful. Actually, having a $1299 version might even help drive sales of the $99 version, because people would perceive the $99 version as a sort of 'deal,' as in "hey, for $99 I'm getting 60% of the features of the $1300 version! That's great! I'll take three."
Yes, actually if you read my comment, I even mention Anycasting. However, the page I was pointing to, lists the geographic locations for the various root servers. Some of them have multiple locations, behind a single IP address, routed via Anycast. The others have only single locations (which may or may not imply multiple physical servers), making them somewhat easier to DDoS, since a world-wide zombie net would be concentrating it's traffic towards one server. On the geographically-distributed "servers" (with Anycast), each zombie would only hit the machine closest to it, so it's harder to take down.
Not sure if I'm entirely understanding your point.
In the example I was talking about above, I was scanning a bunch of old slides, which we can assume the copyright holders have authorized me to scan (hence, why they gave me boxes of slides and asked me to scan them onto the internet-thingamajiggy). So it's not as thought it was an unauthorized derivative work, if in fact it was a derivative work.
My question is basically, would there be two copyrights: e.g., (c) 1939, Grandma (for the original photo), and (c) 2006, me (for the new scan), or just the former.
I think, based on the court case that another respondent cites above, that it would be just the former case; the scan, if it's purely a refixing of the original work in a new medium, without much in the way of creative "work," only the original copyright would apply. So it would be (c) 1939, not 2006 -- a big difference in terms of when the digital copy would hit the public domain (the 2006 one wouldn't end up there, unless I ceded it earlier, until 70 years after my death, while the 1939 one may already be there, or may be there soon -- the law is a bit byzantine on post-1923/pre-1978 works).
So I don't think that the authorization to make the derivative work was really at issue here, or I was assuming for the purposes of argument that I had such authorization, given to me by implication in the request to scan the photos.
They are most likely using this in combination with other more or less 'unique' things to identify a specific machine. It wouldn't surprise me if after this some people would do a more in-depth analysis of their code and find out that it also reads the serial number of the harddrive and gets the MAC address of the Ethernet adapter.
This seems pretty logical. Since they got rid of that hackneyed scheme a while back to give each processor a serial number (wait -- did they get rid of that?), some sort of hash of the BIOS memory, plus the Ethernet MAC, plus the HD serial number, all concatenated together, is probably as close to a unique identifier as you're likely to find on a "per machine" basis.
That said, it doesn't make me feel any better. I wasn't a fan of the processor serial number concept, and not just because it was a serial number in the processor; there were serious privacy concerns with any uniquely identifying, per-machine serialization concept, and that's true whether it's a dedicated number that's being used, or some sort of combination of semi-unique factors.
It's just one more piece of information, sitting in a database somewhere, that could be subpoenaed and used to generally cause trouble. Particularly given how close-mouthed the Skype people are about how their network actually operates (e.g. their alleged encryption, peer to peer communications), I'm not ready to run right out and trust them.
I wonder if it would be possible to run Skype in a sandbox, where the information it's fed could be carefully controlled? On further thought, I wonder what happens when you run it in VMWare or Wine? Do they actually pass information about the hardware up to guest applications? It seems like this behavior would be one that the user should be given an option about, at the very least; I can only think of a few programs who have any reason to be getting the drive serial number, or the Ethernet MAC address, and for the most part they are not userland apps.
It's not like they haven't figured out the whole failover/fault tolerance thing.
That's kind of the point here, actually. Several of the root servers do not have any redundancy. You can see the list at http://www.root-servers.org/. In particular, the A, B, D, E, G, H, and L servers have only a single location a piece.
F, I, J, K, and M, on the other hand, are heavily redundant and have multiple geographic locations, routed via Anycast, so a single client only "sees" the server nearest to them. This makes them difficult to DDoS, because a zombie in S. Korea pinging the J server would be sending packets to the server in Seoul, while one in California would get the one in Mountain View.
What's odd, looking at the list, is that anyone operating something as critical to the internet infrastructure, wouldn't develop some geographic and systems redundancy; unfortunately, I suspect that the government agencies in particular tasked with these responsibilities probably don't keep it at the very top of their priority lists when allocating resources and funding.
I'm sure it is considered in conjunction with other activities. Like if you're otherwise suspected of dealing drugs, use of encryption is "evidence" that you were up to something. Not that it is any more solid, but at least it isn't saying that encrytion, in and of itself, is illegal.
I think the reason people are unhappy with the opinion is because it might cause people to not use, or otherwise discourage the adoption of, encryption, for fear that it would "look bad" if they ever got in trouble.
E.g., an example that might be closer to most people's hearts would be an IRS audit. Suppose you get audited, and honestly haven't done anything wrong, but get stuck trying to clear your name (because this is how the IRS works: you need to demonstrate to them that you're legit). I could easily see situations where people would be afraid that if their computer was seized and GPG was found, that it would be used as evidence that they were "hiding something" regardless of what was actually going on. So instead, they just begin to fear the tools.
This is a self-fulfilling prophesy: as more people fear encryption tools, less innocent people will keep them around, and they'll become more and more the domain of people with something to hide. Then they gain more of a reputation, furthering the cycle.
Well, I'm hoping that some of this "scatter shot" technique being used by the media companies in their so called war on piracy starts backfiring.
Where's Dick Cheney when you need him?
The uploaded version would clearly be a derivitave work, but I'm guessing that putting it into another tangible form would mean it's automatically copyrighted right then even if it wasn't originally.
This is actually a fairly interesting question, and IMO an important one. I'm not sure I share your conclusion that the uploaded version is a new work, though. Although it certainly could be, if you changed it (say, retouched, or even just cropped it), a straight scan+upload probably wouldn't be original enough.
It's an interesting question, because I recently scanned hundreds of old family photos and slides. Many of them, provided Congress stops extending copyright indefinitely, will be out of the original photographer's copyright relatively soon (as in, probably within my lifetime -- copyright, like geology, has its own relative time-scales). However, if the act of scanning the photo automatically makes a new work, then it's under copyright for another 120+ years, beginning 2006. Not really a concern to me, since I'd be the copyright holder, but of concern to a hypothetical other party who might want to use them.
I suspect that simply scanning a photo, in its entirety, and uploading it, does not represent enough of a creative act to warrant a renewal of copyright as a derivative work. Essentially, all that is happening, is that the older work is being "format shifted." However, if you were to do any type of alteration to the original photo that wasn't totally automatic, even something like color correction, I could see an argument for protection on the grounds that it was a creative act.
I think that's severely oversimplifying, because rewriting the system to take only one system call would certainly result in more bugs, no?
And I'm not arguing with you in the slightest. However, in TFA, that's pretty much exactly the assumption that the author makes, in order to start off with the diagrams of system calls, and from there get to the conclusion that IIS/Windows Server is "more secure" than Apache/Linux.
I wasn't really making any commentary on that premise, because I'm hardly qualified to, except that if you wanted to undermine the article's conclusions, the premises seem like a good place to start...
It's hard to say, but two immediate problems come to mind. First, and in my opinion most likely, is that they have in place agreements with the major record labels that involve giving the same treatment to all music sold via the iTMS, so that it all has to be FairPlayed. This strikes me as pretty likely, and something that the record labels would insist on; they must realize that online distribution closes a lot of the gap between a small record company, and them, and obviously they want to avoid direct competition as much as they can. So they'd want to suppress anything that a small, independent company could use as an advantage. Hence, demand that Apple apply the same "protection" to all iTMS-sold music.
The other problem, which isn't exclusive of the first, is that the DRM isn't applied once to each song in the store when it's being added to the database, but added at the time of sale (necessary because it's encrypted with a key that's specific to each user), somewhere on Akamai's servers. It might be difficult to the point of being cost-prohibitive to designate one song as being DRM-free, if the system wasn't designed with that capability from the beginning.
I've noticed that even songs that are free downloads (promo songs, etc.) from the iTMS have FairPlay placed on them, even when you can go to the band's or label's web site and download it as an MP3 (so it's obvious that the label doesn't care if it's protected); this makes me suspect that one or both of those problems exist.
It would probably be trivial for Apple to turn off DRM completely, for all the songs in the Store, but difficult both legally and technically, to disable it for just one.
(I'm not trying to sound like too much of an Apple apologist here, to be frank I think the iTMS is an abomination and I wish Apple had stood up to the record companies when they were screaming about the iPod and contributory infringement a few years ago, and remained a purely hardware company and stayed out of the music-retail business; however, at the time creating the iTMS was the best way of eliminating accusations of the iPod as a "piracy machine." It's ironic that Apple's own creation, created to soothe the record companies, is now coming back to haunt them. Well, that's what you get for dealing with the devil.)
Okay, poor choice of words. I just meant that they refuse to license it for use in other playback devices.
And why should they? Steve Jobs is obviously a smart guy; things he's said and written elsewhere make me think that he understands the inherent problems behind DRM.
In short, DRM doesn't work. It works, sort of, only by keeping the mechanisms out of sight, and changing them all the time, as people catch on and figure out what's going on "behind the curtain."
The more people you let see behind the curtain, the harder it is to make work, and keep working, even in the shoddy way that it does currently. Licensing means that specifications and technical documents need to be written, and such documents can be leaked (and are far more likely to be leaked when they're being sent to some licensee in Europe, than kept within a particular technical working group inside Apple US). So if Apple licensed out FairPlay, it would mean that FairPlay would get broken more often, and they would have to dedicate more effort to fixing it, and those fixes would be harder to roll-out, because there would be more users, and multiple online music stores, run by various licensees who might take their responsibilities for updates more or less seriously, etc. etc.
DRM isn't a single technology that you can sell. It's not a word processor. It really is defective by design; that's not just some dumb slogan -- that is reality. Anyone who buys a DRM system, thinking that it's a product they can just use, and then forget about, is a fool. A DRM system is an arms race. It can only work when you're committed to throwing a lot of programmers behind it; programmers who are constantly shoring it up, as people pull the bricks down from the outside. And the work that it takes to sustain is directly proportionate to the number of people who are working to crack it.
Licensing out FairPlay would be a losing proposition for Apple on all fronts. It would force them to lose revenue from the iTMS, which isn't exactly a huge profit center anyway -- as others have pointed out, Apple makes a lot more money on an iPod than they do on the average user's iTMS purchases. Plus, it would mean that they would have to spend a lot more effort constantly fixing FairPlay, and it would create a huge logistical problem -- how do you roll out those fixes to users who may be using some licensee's music store? If Apple doesn't keep FairPlay's facade of security up, the music labels will use it as a bargaining point in negotiations, but they'll be dependent on their licensees, who they don't have total control over, in order to maintain that facade. It's a lose-lose for Apple.
Personally, I don't think Apple will ever license FairPlay. I think they'll pull all DRMed music from the European market, and close the iTMS there, before they'd open the can of worms that licensing would entail. Exactly what would happen at that point is anybody's guess, but there are a whole lot of iPod-owning Europeans who probably want some type of online music store, and Apple is pretty good at PR. They might be able to turn it into some sort of a victory against the governments mandating the interoperability, or against the music labels who won't sell DRM-free music. Or it might backfire horribly and cause a lot of people to run out and buy non-iPod MP3 players in order to use competing online stores (though I doubt it; I don't think that the presence or absence of an online store is a huge selling point of most music players, except those linked to subscription services like Napster).
This is slightly off-topic, but a while back I got interested in OpenVMS, and VAX stuff in general. (I started doing some research because I thought I was going to get stuck doing some turd polishing of old mainframe software, but it never materialized. But by then I was just interested.) Even in hindsight (given that I think we can agree that UNIX-derivatives seem to have gained traction over VMS), it's extremely difficult to find any sort of rational comparisons of VAX/VMS and its architecture and design paradigms to that of UNIX. Whenever someone asks, the response is basically "don't ask, you don't want to start that." Nobody wants to talk about anything that might invite UNIX/VMS comparisons, because it will cause flamewars -- even though such a discussion, at this point, might be interesting and productive. (There are so many people around who aren't familiar with VMS, or anything other than Windows and UNIX, that any perspective besides those would be worthwhile.)
At any rate, it struck me as interesting, because sometimes it's easy to assume that Windows/Linux (or Windows/Mac, or Windows/something) is the first Great OS War. But people have been getting emotionally attached to operating systems, probably as long as they have existed; and ever since, it has helped quash rational discussion, both through flamewars themselves, but also because of self-censorship that occurs, in order to try and prevent arguments.
Personally, I would argue that such 'heuristically secured' systems are broken by default, and that there are good reasons why generations of computer scientists have insisted that security through obscurity is meaningless. The "security" provided by such heuristics are of value only to marketing and legal departments, they are not and should not be confused with the security offered by 'deterministically secured' systems (e.g. cryptography is his example). Saying that an application is "secure," when it depends on an attacker not knowing how it works, borders on unethical false advertising.
I don't know how they're doing this in practice, whether they're just sending people out with regular handheld digital cameras or what, but it would probably be possible to rig up a nondescript panel van with side-view cameras, and just drive up the street photographing everything on both sides. (Or, for better results, everything on the right side, and then drive up one side of the street, followed by the other.) If you had a very good GPS receiver in the van, you could geotag each photo, and then crop them as a batch later on, for each house or building on the street.
What would probably be even better would be to use a progressive-scan video camera for image capture, so that you have a continuous feed of images, and then you don't have to worry quite so much about having one house cut between two photos. (Alternately you could probably sew the images together into a continuous linear panorama, but that might give mixed results.)
You might still get shot at in some areas, but it would probably be a lot lower risk than just walking around.
On the other side of the pond, ex-felons can't buy guns & have to do some paperwork to get their voting rights re-established... That's pretty much it. Everything else is social stigma.
It's a lot more than that. They are effectively barred from any 'positions of trust' for the remainder of their lives, unless their record is expunged, which is unusual. This pretty much takes them out of any government or civil-service position (including the military), out of jury duty, and out of most teaching positions in public schools. This is just the stuff that they're barred from by law.
By that "social stigma," they're basically unemployable in all but the most menial jobs, and even in those they're basically un-advanceable. I don't think you can even get a fast-food job with a felony on your record, because they're suspicious that you'll start stealing from the till.
Effectively, they're the modern version of "untouchables"; they become members of a permanent underclass unfit for legitimate occupations. I don't have a whole lot of pity for them, frankly, but it's not hard to see why the recidivism rate is so high. But that creates a feedback cycle; the public expects ex-cons to fall back into crime, thus they're not trusted, thus they can't find legitimate occupations, and thus they do exactly what's expected of them.
Short of 'rebooting' the system by summarily executing anyone with a felony conviction, or other similarly drastic methods, I don't think there's any easy way to break that cycle.