You are aware, I hope, that during a significant number of those conflicts we lost a lot more of our freedoms than we are currently discussing even the potential of losing right now...
I'm not defending the current administration's policies, but I just think that you should be careful drawing historical comparisons before you know where they're going. President Lincoln -- who history has treated quite favorably -- declared and imposed martial law, suspended habeas corpus, and arrested people that today would probably be termed "political dissidents," including a few members of Congress. (The anti-war Democrats known as the "Copperheads" were the common target.)
When the arrests and courts-martial were declared blatantly unconstitutional by the Supreme Court (under Taney), Lincoln simply ignored the ruling until the conclusion of the war. You can Google this, just type in "John Merryman" or check out Ex parte Merryman (the ruling that was ignored).
That's one of the more well-known and egregious violations, but there are others; the persecution of the Germans in World War I, the Japanese in World War II, and a host of other things, any of which can and were argued to be necessary at the time owing to extenuating circumstances.
So by drawing a historical parallel between 9/11 and any other "war period" in our history, you can quite easily play into the hands of a pro-oppression argument, because there is ample historical evidence for periods of relative oppression (or at least, of substantially reduced civil liberties) during conflicts, followed by a return to normalcy afterwards.
Search engines pay money to the browser makers to get their search engines in there anyway.
Um, care to back that up?
The two biggest browsers right now are IE and Firefox; Microsoft is basically promoting their own product here (MSN Search) so I doubt they're "paying" anything directly, so really your point seems to suggest that Google has paid Firefox to be the default search. That's a pretty strong statement to be making without any evidence.
To a PHB, RHEL is an "enterprise OS," CentOS is a "free software project."
That is to say, RHEL is made by a company and usually purchased with a service or support contract, comes in a bunch of grades/flavors for different applications, and comes with a lot of documentation and even has certified training courses.
CentOS is just that -- an Operating System. They don't support it, they don't service it, and you can't send Sean From IT to them to take a few training courses on how to administer it. You can basically use all the Red Hat documentation, but of course the documentation says Red Hat and not CentOS on it, so it appears more or less "undocumented."
Basically, CentOS is the operating system, while RHEL is one component of a "solution" sold by Red Hat. A lot of people want more than CentOS is offering, which is why Red Hat stays in business selling Free Software. On the other hand, if you're not afraid to support things yourself and don't need handholding, CentOS encompasses most of the actual software bits that you'd use on your computer. (I think CentOS doesn't include the same package manager as RHEL, but I could be wrong.)
I see your point after reading TFA regarding ISP Config, it definitely expects you to install it. But I have to wonder, after checking out ISP Config, if this is a bad thing. It's not as if this is a closed-source or commercial product, so I don't think that the article writer is getting any sort of kickback from recommending it or using it in his easy setup article. It's BSD licensed, actually, so (depending on your personal definition of free, etc.) it's less thorny an issue in terms of use than Linux itself in many cases.
I guess I'm just wondering what the arguments are against using ISP Config, and why it wouldn't be a good thing to use on a production server or why people dislike it. If you're new enough that you're using a Perfect Setup article to build a server, installing a GUI utility (which is all ISPConfig is) might not be a bad idea. The only downside to it that I can see immediately is that you end up running a totally separate Apache webserver and PHP setup for it, in addition to the one you're using to actually serve web pages. This seems like it might double your security exposure, if it's not kept up to date and patched/locked-down correctly.
It's kind of like all those "how to build a blog" articles that tell the user to install PHPMyAdmin in step 3, and then have later steps that are only explained using PHPMyAdmin, even though they could easily be done using commandline SQL commands. I think the assumption is that if you know how to use the MySQL utilities directly, then you probably are above the level of the intended audience of the howto in the first place.
I thought that the big benefit to the BeOS FS was that it had support for an arbitrary number of metadata fields and forks, and HFS+ does that right now (if you wanted it to). The examples I always saw as to why the Be FS was going to be "cool," to a user's perspective, were things like viewing a list of MP3 files from within the OS's file manager: rather than storing Artist, Title, Length, etc., as ID3 tags, viewable only within an application that parsed the file, you could store the fields as file metadata and then use them at the OS level rather than the application level to view / sort / search / whatever. Similarly you could do the same kind of thing for any kind of file, and anyone (with the OS) would be able to read it regardless of which applications they had or wanted to use. It was kind of a neat concept (interoperability problems with other OSes excluded) but as much as I love metadata, it seems like that's not the direction people are going in.
What were the other reasons it was going to be cool, and which of those reasons aren't covered already?
What do call a CEO who makes the decision to chop $400 million off his company's profits?
Although I agree with your sentiment, it's worth pointing out that $400M in revenue -- which would be sales figures -- does not translate into $400M in profit.
Unless of course you're engaging in a little Enron-style math, that is. Software companies may have high margins, but they're not 100%.
I think they mean the 'solutions' aren't ready.
on
Web 2.0 Goes To Work
·
· Score: 1
Actually I think "technologies" here really means "solutions" (a real buzzword-of-buzzwords); basically people are saying/complaining that if they went out right now and grabbed an AJAX package and a Wiki package, melded the two of them together, it would be insecure.
I don't think there's any debate that if you planned it well, you could make a secure web-delivered application using AJAX and which had some wiki-like functions (at least as secure as any other web delivered application), but a ground-up coding effort isn't what most PHBs (or really anyone else) wants. They want something that's basically COTS, and they can basically roll out, customize a little with their logo, and be done with. "Production ready," in other words.
Notwithstanding the naiveté inherent in this ideal, I think what they're saying is that the technology is not quite there because nobody has sat down and designed something that would be secure, but none of the potential users really want to do that. They'd rather just call it "not ready" and "immature" until somehow the work gets done, and they can turn around and deploy it.
Humm... but "Fraudipedia" just doesn't have a ring to it. You have to come up with some better name. Let's see, it's like Wikipedia, but it's kind of not doing what wikipedia does at all. In fact, it's nothing like what Wikipedia tries to do. It's practically the opposite of Wikipedia. Let's see, one common prefix for 'opposite' in English is "un." UnWikipedia. Humm, well you probably shouldn't put Wikipedia's actual name in there, or else Jimbo Wales might come after you with a takedown order and an axe, so maybe we should just drop the "Wiki." Wikipedia is really an encylopedia, and that's not a trademarked word.... so you could call it "Uncyclopedia"!
"Uncyclopedia.org"... Brilliant -- you better go get that domain before somebody else thinks of it.
I knew I had heard about something like that somewhere... thanks.
In my original post when I was thinking about ways you could store energy on a really large scale, it seems like pumping water is probably one of the more practical ones. Batteries, flywheels, hydrogen/fuel-cell storage, all seem like they'd have too much losses associated with them.
Only somewhat related.... I recall an article a while back, I think it might have been in PopMech (don't quote me on that though), about a crazy scheme to build a hydro-powered plant at the Isthmus of Nicaragua? It was kind of interesting because it used the difference in height between the Atlantic and Pacific oceans, but rather than building a canal or something across to carry the water, the idea was to build a big pipe that ran up and over the land, but was completely sealed. So it would basically act like a siphon, pulling the water from the higher side up and over the land (and through the power plant) and down the other side into the lower ocean. A little ridiculous because of the scale -- can you imagine the problems keeping it completely sealed and free of air so that the suction would work? -- but it was neat anyway. I could imagine doing it if you had two lakes near each other and you didn't want to cut a canal through for a conventional hydro plant.
Well, the actual decryption key for the drive's data is probably stored on there somewhere, it's just probably encrypted itself with a pass-phrase that you choose. At least that's how I'd imagine it would have to work. It doesn't make sense for the user to have to remember an entire 1024 or 2048-bit key. Although I guess you could put it on a smartcard or some other kind of physical device, but I haven't heard of that being required.
You're right though, at some point the system has to depends on a key (of some length) not stored on the hard drive, or it would be rather trivial to break...
(2) a national ID could be made more secure and harder to forge than the varied state ID's used now
Now, I have a host of objections to National ID cards, but I really think that the whole "they'll be harder to forge" business is a load of crap.
These things are going to have to get turned out by the hundreds of millions. Whatever great and wonderful security features you think are going to get put in them, they're not going to be worth crap once; the economic incentives to forge them is going to be huge. As people get comfortable using them as IDs for everything--which they will, in short order--they'll become more and more valuable to criminals, plus, as the government finds more and more ways to use the ID's (first, to get on an airplane, next trains, then buses, then amusement parks, music concerts, public transportation, parks and zoos, etc.) and surveillance becomes more omnipresent, the demand for them will increase among people who previously weren't interesting in having a fake ID, "just in case" or because they have something small or embarrassing to hide (an infidelity, or homosexual tendency, going to the wrong church, etc.) that they don't want to become known or linked to them. People desire anonymity, even non-criminals.
So, you'll have a huge demand, and suddenly you'll have all the forgers in the entire country (and probably a lot of foreign ones) working on making fakes. Given the amount of communication today, I bet it won't take long before the methods of making a fake become widely known.
The science of forgery is just like software cracking, only it's with physical objects instead of bits and algorithms. Just like DRM schemes always get cracked in the long run, all IDs can be forged. All this scheme would do is put all the talented people in that field on one task.
Plus, this doesn't count the "non-forged fakes." IDs that are 'real' physically, but have bad data printed on them. It only takes one bad employee at the factory where they make the card-printing machines or even in local government to pull a Oscar Schindler and make a few thousand fakes. Or for some less-savvy criminals to just break into a DMV and steal the machinery, and combine it with information gleaned from phishing schemes. One set of card-printers and you could steal anyone's identity: it wouldn't depend on knowing how to forge their home state's drivers license.
There are steps you could take to make an ID difficult to forge, but most of them are expensive to implement. We're talking here about a product that would need to be manufactured cheaply, all over the place, and in ridiculous quantities, for a few cents each. The economic realities mean that it would probably be no harder to forge than a drivers license now, and the demand for them would be outrageous compared to what we have now.
I'm not sure exactly myself, but it's not so wildly out-of-the-box an idea that nobody can have thought of it before. I assume there's something wrong with the economics of doing it at the generating station. Maybe it has to do with going down from typical generation voltages to something that can be stored and then back up again? (That would be the problem using batteries...) Other large-scale forms of energy storage, things that could store real MWh's, might be impractical.
Actually, when you think about how hydroelectric power plants work, they do this already: they build water up behind the dam when demand is low, then open the gates further and produce more energy when demand is high. I know it's not the kind of "storage" we're talking about here, but most power plants have some form of output regulation; it seems like the power companies are probably trying to match demand as closely as they can, from their "top down" perspective, but can only get so close.
By putting small storage devices out at the edge, close to the points of consumption and where voltages are low, you might get a lot more effect than taking the same amount of storage and putting it all upstream.
I'm going to go out on a limb here, but I have a distinct feeling that Apple moved more units of PPC Macs than IBM did of RS/6000s. Those things aren't exactly a dime a dozen; you can't just walk down to CompUSA and pick one up to put under the Christmas tree.
I can't find any hard sales figures recently, but back in 2000, at the height of the Bubble, IBM was touting that it had sold "over 1,000,000 RS/6000" servers total to date, while Apple shipped more machines than that (1.12M) in the fourth quarter of 2000 alone.
I've been told on good authority that the biggest users of Power chips have always been embedded systems, and that this is where IBM was putting the majority of its development focus (particularly on lowering power consumption and making them run cooler).
Well, I've heard that the Chinese government has its own TCPM type chip that it either mandates inclusion of, or gives some sort of incentive to companies that install it. Apparently Lenovo computers made for the home market have it in preference to the one used in units destined for the rest of the world.
So it's probably treacherous, but in ways that would be different than ours here in the U.S. It might still be advantageous for a freedom-loving American to get a crippled Chinese PC (and vice versa) because the software to utilize the crippling features of it might not be as widely available or as thoroughly implemented here. I doubt that most software looking for a TCPM is going to be able to use the Chinese security chip, and their software (in the PRC) which expects one probably isn't going to be able to use the one that Dell's installing.
As for your other question, I'm not positive, but I don't believe that the big IBM workstations have them installed (at least the RISC ones). The Opteron based ones might or might not, I'm not sure what the motherboards are like in there. Of course you're talking real money compared to a commodity PC, even if you get an x86-based IBM workstation. It may be 2006 out here but they price those things like it's still 1995. (They were home of the "most expensive PC" last time I checked around.)
Actually I cheered when everyone went to TP wiring, even thought it meant a bundle the size of your arm at one end of the lab, just because it meant that if someone disconnected something somewhere, or pulled the terminator off of the end of the line, it wouldn't take the entire lab down.
I have no idea why, but people used to love to steal those terminators...I think they had to buy them by the gross just to keep the network operating.
But you're right, there is a certain elegance to the thin coax. It's the same way with fiber: you have a big rackmount switch, with dozens of 100BT lines running in from the nodes, and one measly half-pencil-thickness fiber line running out as the uplink.
Although there is something fairly impressive about having big cable raceway that's fully loaded with tens or hundreds of TP lines. It's something easy that you can point out to the PHB's of the world when they need a visual for "network infrastructure."
I should have been more clear. The problems with the thumbnails are definitely quite valid. As are the performance and reliability issues. But in the Ars review he knocks the quality of Aperture quite a bit, and says it's RAW stuff looks worse than Photoshop, when PS is (I think) doing some preprocessing in the name of 'looking good' versus accuracy.
It was kind of an indirect response, I was more criticizing the Ars review a little, not anything you said. I just wanted to encourage people to take some of the points in the review with a bit of a grain of salt, although many of its points are valid.
Sorry if that wasn't clear.
Re:Is this the root of EA's problems?
on
EA Spouse Outed
·
· Score: 2, Informative
Sleep deprived cranky game developers can't possibly be very creative, can they?
Nah, artists always work best when they're coddled, fat, and happy. Oh, wait...
Just kidding, although I do think that sometimes a deadline is the kick in the pants that's sometimes needed for people to produce their best work, there's no excuse to just abuse your people continually. You can only maintain that kind of increased tempo for a certain amount of time, before it just becomes fatiguing and output quality is going to drop.
I've actually worked on a few big projects where I remember the "big push" at the end with some fondness. Okay, at the time I probably would have called you stupid for saying that, but in retrospect I knew that it brought the team of people I was working with together and caused us to make a better product than we probably would have done, if we had spread the same number of hours of work out across a traditional work schedule. (Disclaimer, I'm not in gaming, but I can't imagine it's that much different.) However I can only imagine doing that regularly... that's not going to do anyone any good.
Back on topic here -- does anyone know where I can get the backstory on ea_spouse? I didn't follow this too closely when it was going on so I'm wondering if I can fill myself in.
I think before you put TOO much weight on the Ars review, you should take into consideration what jcr said above, because I think it's an important point.
Saying that Aperture's output isn't as pretty as Photoshop's is like complaining that your photos look shittier on slides than on prints, without taking into consideration that with the slide you're looking at your own (and the camera's) handiwork and nothing else, while with the print you're looking at something that's been optimized by someone else (the printer) to look good.
The speed problems are unacceptable though. I just thought the Aperture/Photoshop comparison wasn't a great one; although it's odd to say it, Photoshop has become a "mid grade" application, I think Aperture was going for an even 'more-pro' crowd than the average Photoshop user.
I think in retrospect Apple is realizing maybe that market is smaller than they originally thought.
On the Mac, it's easy - they have Core Audio drivers, and are usable by any well-written Mac application (YMMV with ProTools). But in Garageband, Nuendo, Final Cut, Audacity, etc., they just show up as regular input/output devices. Neat thing with Core Audio is that you can actually route one interface to multiple programs simultaneously (send inputs 1-6 to Garageband, inputs 7-8 are the mix through an outboard piece of gear then brought back to Nuendo, use inputs 9-16 for tape ins in Audacity).
Now that's slick. I haven't really done much with computer-based audio in a while--the last time I had access to a DAW, it was using some version of Protools that wasn't OS X compatible--and I've never used CoreAudio for anything more complex than two channels.
You're right, I can't really think of much reason why you'd want to send various channels to different applications, but it's nice to know the feature exists.
Anybody know of any implementations of this? Seems like there would be a lot of rather obvious uses, but I've never heard of it being used. (Or maybe I have and just didn't recognize it as such.)
My interpretation was that the vector space wasn't explicitly defined, but was assumed to be larger than either k or l.
So basically, n-dimensional vector space where n > l > k.
That was my assumption; although I think you can have n => l > k and it still works (of course when l is n-dimensional and orthogonal, then it's a basis).
As you pointed out, it doesn't make sense for k to be n-dimensional, because then you can't have l>k.
Well, according to the spec you can only restore from ECHELON by going through the FISA_COURTS interface, but everyone's known for a while now it's been backdoored if you have sufficient privileges.
Slashdot Mirror
You are aware, I hope, that during a significant number of those conflicts we lost a lot more of our freedoms than we are currently discussing even the potential of losing right now...
I'm not defending the current administration's policies, but I just think that you should be careful drawing historical comparisons before you know where they're going. President Lincoln -- who history has treated quite favorably -- declared and imposed martial law, suspended habeas corpus, and arrested people that today would probably be termed "political dissidents," including a few members of Congress. (The anti-war Democrats known as the "Copperheads" were the common target.)
When the arrests and courts-martial were declared blatantly unconstitutional by the Supreme Court (under Taney), Lincoln simply ignored the ruling until the conclusion of the war. You can Google this, just type in "John Merryman" or check out Ex parte Merryman (the ruling that was ignored).
That's one of the more well-known and egregious violations, but there are others; the persecution of the Germans in World War I, the Japanese in World War II, and a host of other things, any of which can and were argued to be necessary at the time owing to extenuating circumstances.
So by drawing a historical parallel between 9/11 and any other "war period" in our history, you can quite easily play into the hands of a pro-oppression argument, because there is ample historical evidence for periods of relative oppression (or at least, of substantially reduced civil liberties) during conflicts, followed by a return to normalcy afterwards.
Search engines pay money to the browser makers to get their search engines in there anyway.
Um, care to back that up?
The two biggest browsers right now are IE and Firefox; Microsoft is basically promoting their own product here (MSN Search) so I doubt they're "paying" anything directly, so really your point seems to suggest that Google has paid Firefox to be the default search. That's a pretty strong statement to be making without any evidence.
To a PHB, RHEL is an "enterprise OS," CentOS is a "free software project."
That is to say, RHEL is made by a company and usually purchased with a service or support contract, comes in a bunch of grades/flavors for different applications, and comes with a lot of documentation and even has certified training courses.
CentOS is just that -- an Operating System. They don't support it, they don't service it, and you can't send Sean From IT to them to take a few training courses on how to administer it. You can basically use all the Red Hat documentation, but of course the documentation says Red Hat and not CentOS on it, so it appears more or less "undocumented."
Basically, CentOS is the operating system, while RHEL is one component of a "solution" sold by Red Hat. A lot of people want more than CentOS is offering, which is why Red Hat stays in business selling Free Software. On the other hand, if you're not afraid to support things yourself and don't need handholding, CentOS encompasses most of the actual software bits that you'd use on your computer. (I think CentOS doesn't include the same package manager as RHEL, but I could be wrong.)
The key is support and branding.
I see your point after reading TFA regarding ISP Config, it definitely expects you to install it. But I have to wonder, after checking out ISP Config, if this is a bad thing. It's not as if this is a closed-source or commercial product, so I don't think that the article writer is getting any sort of kickback from recommending it or using it in his easy setup article. It's BSD licensed, actually, so (depending on your personal definition of free, etc.) it's less thorny an issue in terms of use than Linux itself in many cases.
I guess I'm just wondering what the arguments are against using ISP Config, and why it wouldn't be a good thing to use on a production server or why people dislike it. If you're new enough that you're using a Perfect Setup article to build a server, installing a GUI utility (which is all ISPConfig is) might not be a bad idea. The only downside to it that I can see immediately is that you end up running a totally separate Apache webserver and PHP setup for it, in addition to the one you're using to actually serve web pages. This seems like it might double your security exposure, if it's not kept up to date and patched/locked-down correctly.
It's kind of like all those "how to build a blog" articles that tell the user to install PHPMyAdmin in step 3, and then have later steps that are only explained using PHPMyAdmin, even though they could easily be done using commandline SQL commands. I think the assumption is that if you know how to use the MySQL utilities directly, then you probably are above the level of the intended audience of the howto in the first place.
I thought that the big benefit to the BeOS FS was that it had support for an arbitrary number of metadata fields and forks, and HFS+ does that right now (if you wanted it to). The examples I always saw as to why the Be FS was going to be "cool," to a user's perspective, were things like viewing a list of MP3 files from within the OS's file manager: rather than storing Artist, Title, Length, etc., as ID3 tags, viewable only within an application that parsed the file, you could store the fields as file metadata and then use them at the OS level rather than the application level to view / sort / search / whatever. Similarly you could do the same kind of thing for any kind of file, and anyone (with the OS) would be able to read it regardless of which applications they had or wanted to use. It was kind of a neat concept (interoperability problems with other OSes excluded) but as much as I love metadata, it seems like that's not the direction people are going in.
What were the other reasons it was going to be cool, and which of those reasons aren't covered already?
Ah, ResEdit. I have a warm place in my heart for that program.
The mischief that a person could create with a few dumb ResEdit hacks.... priceless.
Damn, you beat me to it. :)
Every time I see that play put on in a high school, I laugh... though the joke is basically lost if Hamlet doesn't enunciate "country" right.
What do call a CEO who makes the decision to chop $400 million off his company's profits?
Although I agree with your sentiment, it's worth pointing out that $400M in revenue -- which would be sales figures -- does not translate into $400M in profit.
Unless of course you're engaging in a little Enron-style math, that is. Software companies may have high margins, but they're not 100%.
Actually I think "technologies" here really means "solutions" (a real buzzword-of-buzzwords); basically people are saying/complaining that if they went out right now and grabbed an AJAX package and a Wiki package, melded the two of them together, it would be insecure.
I don't think there's any debate that if you planned it well, you could make a secure web-delivered application using AJAX and which had some wiki-like functions (at least as secure as any other web delivered application), but a ground-up coding effort isn't what most PHBs (or really anyone else) wants. They want something that's basically COTS, and they can basically roll out, customize a little with their logo, and be done with. "Production ready," in other words.
Notwithstanding the naiveté inherent in this ideal, I think what they're saying is that the technology is not quite there because nobody has sat down and designed something that would be secure, but none of the potential users really want to do that. They'd rather just call it "not ready" and "immature" until somehow the work gets done, and they can turn around and deploy it.
Humm ... but "Fraudipedia" just doesn't have a ring to it. You have to come up with some better name. Let's see, it's like Wikipedia, but it's kind of not doing what wikipedia does at all. In fact, it's nothing like what Wikipedia tries to do. It's practically the opposite of Wikipedia. Let's see, one common prefix for 'opposite' in English is "un." UnWikipedia. Humm, well you probably shouldn't put Wikipedia's actual name in there, or else Jimbo Wales might come after you with a takedown order and an axe, so maybe we should just drop the "Wiki." Wikipedia is really an encylopedia, and that's not a trademarked word .... so you could call it "Uncyclopedia"!
... Brilliant -- you better go get that domain before somebody else thinks of it.
"Uncyclopedia.org"
I knew I had heard about something like that somewhere ... thanks.
.... I recall an article a while back, I think it might have been in PopMech (don't quote me on that though), about a crazy scheme to build a hydro-powered plant at the Isthmus of Nicaragua? It was kind of interesting because it used the difference in height between the Atlantic and Pacific oceans, but rather than building a canal or something across to carry the water, the idea was to build a big pipe that ran up and over the land, but was completely sealed. So it would basically act like a siphon, pulling the water from the higher side up and over the land (and through the power plant) and down the other side into the lower ocean. A little ridiculous because of the scale -- can you imagine the problems keeping it completely sealed and free of air so that the suction would work? -- but it was neat anyway. I could imagine doing it if you had two lakes near each other and you didn't want to cut a canal through for a conventional hydro plant.
In my original post when I was thinking about ways you could store energy on a really large scale, it seems like pumping water is probably one of the more practical ones. Batteries, flywheels, hydrogen/fuel-cell storage, all seem like they'd have too much losses associated with them.
Only somewhat related
Well, the actual decryption key for the drive's data is probably stored on there somewhere, it's just probably encrypted itself with a pass-phrase that you choose. At least that's how I'd imagine it would have to work. It doesn't make sense for the user to have to remember an entire 1024 or 2048-bit key. Although I guess you could put it on a smartcard or some other kind of physical device, but I haven't heard of that being required.
...
You're right though, at some point the system has to depends on a key (of some length) not stored on the hard drive, or it would be rather trivial to break
(2) a national ID could be made more secure and harder to forge than the varied state ID's used now
Now, I have a host of objections to National ID cards, but I really think that the whole "they'll be harder to forge" business is a load of crap.
These things are going to have to get turned out by the hundreds of millions. Whatever great and wonderful security features you think are going to get put in them, they're not going to be worth crap once; the economic incentives to forge them is going to be huge. As people get comfortable using them as IDs for everything--which they will, in short order--they'll become more and more valuable to criminals, plus, as the government finds more and more ways to use the ID's (first, to get on an airplane, next trains, then buses, then amusement parks, music concerts, public transportation, parks and zoos, etc.) and surveillance becomes more omnipresent, the demand for them will increase among people who previously weren't interesting in having a fake ID, "just in case" or because they have something small or embarrassing to hide (an infidelity, or homosexual tendency, going to the wrong church, etc.) that they don't want to become known or linked to them. People desire anonymity, even non-criminals.
So, you'll have a huge demand, and suddenly you'll have all the forgers in the entire country (and probably a lot of foreign ones) working on making fakes. Given the amount of communication today, I bet it won't take long before the methods of making a fake become widely known.
The science of forgery is just like software cracking, only it's with physical objects instead of bits and algorithms. Just like DRM schemes always get cracked in the long run, all IDs can be forged. All this scheme would do is put all the talented people in that field on one task.
Plus, this doesn't count the "non-forged fakes." IDs that are 'real' physically, but have bad data printed on them. It only takes one bad employee at the factory where they make the card-printing machines or even in local government to pull a Oscar Schindler and make a few thousand fakes. Or for some less-savvy criminals to just break into a DMV and steal the machinery, and combine it with information gleaned from phishing schemes. One set of card-printers and you could steal anyone's identity: it wouldn't depend on knowing how to forge their home state's drivers license.
There are steps you could take to make an ID difficult to forge, but most of them are expensive to implement. We're talking here about a product that would need to be manufactured cheaply, all over the place, and in ridiculous quantities, for a few cents each. The economic realities mean that it would probably be no harder to forge than a drivers license now, and the demand for them would be outrageous compared to what we have now.
Maybe the technology doesn't scale well?
I'm not sure exactly myself, but it's not so wildly out-of-the-box an idea that nobody can have thought of it before. I assume there's something wrong with the economics of doing it at the generating station. Maybe it has to do with going down from typical generation voltages to something that can be stored and then back up again? (That would be the problem using batteries...) Other large-scale forms of energy storage, things that could store real MWh's, might be impractical.
Actually, when you think about how hydroelectric power plants work, they do this already: they build water up behind the dam when demand is low, then open the gates further and produce more energy when demand is high. I know it's not the kind of "storage" we're talking about here, but most power plants have some form of output regulation; it seems like the power companies are probably trying to match demand as closely as they can, from their "top down" perspective, but can only get so close.
By putting small storage devices out at the edge, close to the points of consumption and where voltages are low, you might get a lot more effect than taking the same amount of storage and putting it all upstream.
You just need to install the bsodsim package; that'll take care of it.
I'm going to go out on a limb here, but I have a distinct feeling that Apple moved more units of PPC Macs than IBM did of RS/6000s. Those things aren't exactly a dime a dozen; you can't just walk down to CompUSA and pick one up to put under the Christmas tree.
I can't find any hard sales figures recently, but back in 2000, at the height of the Bubble, IBM was touting that it had sold "over 1,000,000 RS/6000" servers total to date, while Apple shipped more machines than that (1.12M) in the fourth quarter of 2000 alone.
I've been told on good authority that the biggest users of Power chips have always been embedded systems, and that this is where IBM was putting the majority of its development focus (particularly on lowering power consumption and making them run cooler).
Well, I've heard that the Chinese government has its own TCPM type chip that it either mandates inclusion of, or gives some sort of incentive to companies that install it. Apparently Lenovo computers made for the home market have it in preference to the one used in units destined for the rest of the world.
So it's probably treacherous, but in ways that would be different than ours here in the U.S. It might still be advantageous for a freedom-loving American to get a crippled Chinese PC (and vice versa) because the software to utilize the crippling features of it might not be as widely available or as thoroughly implemented here. I doubt that most software looking for a TCPM is going to be able to use the Chinese security chip, and their software (in the PRC) which expects one probably isn't going to be able to use the one that Dell's installing.
As for your other question, I'm not positive, but I don't believe that the big IBM workstations have them installed (at least the RISC ones). The Opteron based ones might or might not, I'm not sure what the motherboards are like in there. Of course you're talking real money compared to a commodity PC, even if you get an x86-based IBM workstation. It may be 2006 out here but they price those things like it's still 1995. (They were home of the "most expensive PC" last time I checked around.)
Actually I cheered when everyone went to TP wiring, even thought it meant a bundle the size of your arm at one end of the lab, just because it meant that if someone disconnected something somewhere, or pulled the terminator off of the end of the line, it wouldn't take the entire lab down.
I have no idea why, but people used to love to steal those terminators...I think they had to buy them by the gross just to keep the network operating.
But you're right, there is a certain elegance to the thin coax. It's the same way with fiber: you have a big rackmount switch, with dozens of 100BT lines running in from the nodes, and one measly half-pencil-thickness fiber line running out as the uplink.
Although there is something fairly impressive about having big cable raceway that's fully loaded with tens or hundreds of TP lines. It's something easy that you can point out to the PHB's of the world when they need a visual for "network infrastructure."
I should have been more clear. The problems with the thumbnails are definitely quite valid. As are the performance and reliability issues. But in the Ars review he knocks the quality of Aperture quite a bit, and says it's RAW stuff looks worse than Photoshop, when PS is (I think) doing some preprocessing in the name of 'looking good' versus accuracy.
It was kind of an indirect response, I was more criticizing the Ars review a little, not anything you said. I just wanted to encourage people to take some of the points in the review with a bit of a grain of salt, although many of its points are valid.
Sorry if that wasn't clear.
Sleep deprived cranky game developers can't possibly be very creative, can they?
...
... that's not going to do anyone any good.
Nah, artists always work best when they're coddled, fat, and happy. Oh, wait
Just kidding, although I do think that sometimes a deadline is the kick in the pants that's sometimes needed for people to produce their best work, there's no excuse to just abuse your people continually. You can only maintain that kind of increased tempo for a certain amount of time, before it just becomes fatiguing and output quality is going to drop.
I've actually worked on a few big projects where I remember the "big push" at the end with some fondness. Okay, at the time I probably would have called you stupid for saying that, but in retrospect I knew that it brought the team of people I was working with together and caused us to make a better product than we probably would have done, if we had spread the same number of hours of work out across a traditional work schedule. (Disclaimer, I'm not in gaming, but I can't imagine it's that much different.) However I can only imagine doing that regularly
Back on topic here -- does anyone know where I can get the backstory on ea_spouse? I didn't follow this too closely when it was going on so I'm wondering if I can fill myself in.
I think before you put TOO much weight on the Ars review, you should take into consideration what jcr said above, because I think it's an important point.
Saying that Aperture's output isn't as pretty as Photoshop's is like complaining that your photos look shittier on slides than on prints, without taking into consideration that with the slide you're looking at your own (and the camera's) handiwork and nothing else, while with the print you're looking at something that's been optimized by someone else (the printer) to look good.
The speed problems are unacceptable though. I just thought the Aperture/Photoshop comparison wasn't a great one; although it's odd to say it, Photoshop has become a "mid grade" application, I think Aperture was going for an even 'more-pro' crowd than the average Photoshop user.
I think in retrospect Apple is realizing maybe that market is smaller than they originally thought.
On the Mac, it's easy - they have Core Audio drivers, and are usable by any well-written Mac application (YMMV with ProTools). But in Garageband, Nuendo, Final Cut, Audacity, etc., they just show up as regular input/output devices. Neat thing with Core Audio is that you can actually route one interface to multiple programs simultaneously (send inputs 1-6 to Garageband, inputs 7-8 are the mix through an outboard piece of gear then brought back to Nuendo, use inputs 9-16 for tape ins in Audacity).
Now that's slick. I haven't really done much with computer-based audio in a while--the last time I had access to a DAW, it was using some version of Protools that wasn't OS X compatible--and I've never used CoreAudio for anything more complex than two channels.
You're right, I can't really think of much reason why you'd want to send various channels to different applications, but it's nice to know the feature exists.
Thanks for the info.
Anybody know of any implementations of this? Seems like there would be a lot of rather obvious uses, but I've never heard of it being used. (Or maybe I have and just didn't recognize it as such.)
My interpretation was that the vector space wasn't explicitly defined, but was assumed to be larger than either k or l.
So basically, n-dimensional vector space where n > l > k.
That was my assumption; although I think you can have n => l > k and it still works (of course when l is n-dimensional and orthogonal, then it's a basis).
As you pointed out, it doesn't make sense for k to be n-dimensional, because then you can't have l>k.
Well, according to the spec you can only restore from ECHELON by going through the FISA_COURTS interface, but everyone's known for a while now it's been backdoored if you have sufficient privileges.