Well, the iPhone, iPod touch and Mac desktops and laptops run the same OS in the same sense that, say, a web server and an Android phone both run "Linux". Which is to say that the core bits are either the same or very similar, but supplemented in each case by different sets of drivers, userland applications, etc.
And yes, the version of the OS that runs on phones and iPods can multitask; any time one of those devices is powered on it's running multiple applications. Apple just doesn't distribute third-party applications which continue running when not active. They claim it's to suit the fact that those are resource- and battery-constrained devices, a claim which you are free to believe or disbelieve at your pleasure.
Under the terms of the GPL, linking is considered a derivative work.
But the GPL isn't the arbiter of what is and is not a derivative work; the copyright laws of your country are. In the US, it's far from clear that dynamic linking creates a derivative work under the law (static linking doesn't come into play because it triggers copyright law in other ways). And if a court were to rule that dynamic linking does not create a derivative, the GPL would pretty much be toast.
So if AT&T is pissed at VOIP (wow, no Telecom has *evaaaaah* been pissed at VOIP), would they... could they.... just possibly..... dial up their *exclusive* partner and subtly indicate, "Hey... would you mind suppressing our competition?".
You do know that there's an approved and free Skype application available, right?
And if you'd take five minutes to actually read the article you'd see that the FCC is pretty clearly aware of that, and that several of their questions are implicitly but still quite clearly saying "we know you've already approved VoIP stuff, so we're wondering why those apps were OK and Google's app isn't".
Some places are simply much more disaster-prone than others.
And some disasters are larger than others. A hurricane, for example, can devastate huge swathes of land because even a "normal"-sized hurricane is hundreds of miles across and can chug along for days or even weeks. But even the biggest, nastiest, meanest tornadoes don't get much above a mile wide and typically blow out very quickly. And those are the rarest of the rare; the combination of short lifespan and small size means tornadoes tend to be extremely localized in the damage they do. And thanks to low population density, the vast majority of twisters do no more than churn up empty. So even though there are hundreds of tornadoes each year, the likelihood of any particular inhabited place being hit is, well, tiny.
Add in the fact that hardening a building isn't terribly difficult (my small-town employer here has a datacenter that can take on anything we're likely to see, and a lot of stuff we're not likely to see) and I just don't see an issue. Oh, and the proposed site is in Manhattan, KS, which already has a nuclear reactor (on the Kansas State University campus) and which got swiped by an EF4 tornado last year. Some houses got chewed up, the university's wind erosion lab was destroyed (irony!), buildings on campus got damaged, and... um, that was about it. The stuff that needs to be secure is secure.
Browser rendering engines? In my application UI? It's more likely than you think, especially if you use Firefox, or any other application built around a XUL runtime. How many CSS-only exploits you heard of for them?
If you can read it, there's a fun discussion on the French Wikipedia over this sort of thing, with the suggestion that instead of "Américains" (literally, "Americans") the French-language Wikipedia should use "États-uniens" (literally, "United-Statesians"). Which brings up some fun replies (my loose translation is supplied here):
Aucun autre pays sur le continent américain n'a repris le terme Amérique dans sa dénomination. Par contre, le nom officiel du Mexique est États-Unis Mexicains . Donc si l'on parle d' États-uniens , désigne-t-on des Américains ou des Mexicains
(No other country on the American continent uses "America" in its name. By way of contrast, the official name of Mexico is "United States of Mexico". So when one speaks of "United-Statesians", does one mean "Americans" or Mexicans?)
My personal favorite:
Le terme États-unien est tout simplement absurde. Le nom complet du pays est États-Unis d'Amérique , le simple bon sens veut que l'on nomme ses habitants, les Américains . Le nom officiel de la France est République française , doit-on nommer ses habitants, les Républicains ?
(The term "United-Statesian" is simply absurd. The full name of the country is "United States of America", so common sense says we should call its people "Americans". The official name of France is "French Republic"; must we call its people "Republicans"?)
One is whether you can answer a couple simple questions: where is this program's executable located, and where is its configuration located? Though in theory there are standards for this, in practice they're not followed or don't really solve the problem. The executable may end up in/bin,/sbin,/usr/bin,/usr/sbin,/usr/local/bin,/opt/bin,/opt/local/bin, etc., etc. The configuration may be in a dotfile in your home directory, it may be in multiple files in one or more directories, it may be site-wide in/etc and customizable (or not) on a per-user basis, etc., etc.
The only platform I've seen tackle this in anything approaching a consistent way is OS X, where you know that there's a directory where the applications live, and you can easily predict where the configuration will be (Library/Application Support/(name of application), either top-level for global config or in the user's directory for per-user config). Simply getting a consistent standard for this on Linux distros would be a big win.
Organization of files created by end users is tougher; various platforms have tried different approaches -- "My Documents", specific locations for different types of media, etc. -- and so far nobody's really managed to beat this, with the result that every user either ends up with their own ad-hoc organizational system or (more commonly) ends up with a single directory holding "UNTITLED-1", "UNTITLED-2", and so on. And this isn't really related to users being dumb or not knowing how to use a computer; it's due to the fact that (1) we still impose a physical metaphor (a desktop, with files organized into folders inside) even though it's increasingly irrelevant and (2) we still haven't noticed that managing individual pieces and collections of data is a tedious and repetitive task, and that automating such tasks is a big part of why we use computers in the first place.
There are only two ways socialized health care can save money
This is what we call a false dilemma.
While my personal opinion is going to stay personal for the moment, you're neglecting a variety of factors which affect the cost of health care. Most obvious is the ability of large entities to negotiate better rates; I have personal experience of working in health insurance, and saw this first-hand. When you represent a large group of patients, negotiating with a large group of doctors and/or hospitals, you can easily work out average costs which are far lower than any individual would manage but which still let everyone involved remain profitable. The federal government, incidentally, tends to get better rates than even the biggest private insurers, because it represents a staggeringly large pool of patients through Medicare.
Other factors -- reducing the number of middlemen, simplified interoperability, etc. -- would doubtless come into play as well, and none of these are "socialized magic". They're well-known and long-recognized market forces.
(incidentally, the largest contribution to the cost of health care is the cost of malpractice insurance premiums, which do not in any way correlate to the size of malpractice damage awards -- costs related to malpractice suits in the US have grown at roughly the rate of inflation in recent years, but the premiums for malpractice insurance have absolutely skyrocketed over the same period, in some cases rising by eight to ten times the amount of the corresponding rise in costs. This has nothing to do with what happens in courtrooms, and much to do with what happens in stock exchanges)
Personally, I think the higher profit-per-unit is a better bet in the long run, since it's a bit easier to survive a downturn in the market (when you're not making much on each unit sold, you need to sell a lot more to stay afloat -- that's why, for example, Dell is hurting pretty bad right now).
I mistyped -- meant "profit", not "profit margin" (too many arguments too early in the morning).
What I had in mind was a comparison like:
Operating system sales: 75 cents on each of a hundred million units
Premium hardware sales: 75 dollars on each of a million units
Which is, as far as I can tell, a reasonable pair of numbers to pull out of my ass (though nobody in the general public knows for sure how much Apple makes on hardware). This underscores the fact that you can make some pretty good money on hardware if you manage to market it correctly, and the fact that you don't have to be on everybody's home computer to be raking in the dough.
They could probably kill windows overnight if they invested in mainstream hardware drivers, and got quickly to the critical mass where hardware manufacturers have to develop drivers for them.
But is that their business model? Or is it a good business model compared to the one they currently have? Even setting aside the costs of supporting any old piece of hardware someone wants to run on, the difference between the OS business and the premium PC business is pretty significant. The former has a couple orders of magnitude more unit sales, but the latter has a couple orders of magnitude higher profit margins.
You're making the fatal mistake of assuming that, say, a court proceeding works according to the sorts of logical rules you're used to. This is by far the easiest way to end up on the wrong end of a court's decision.
If you work for a record company and you set out to create the impression that you're listening to and distributing bootleg copies of an album (ahead of its release date or not), I highly doubt that your boss will think it's a great prank if you pull the "honest, it was just a file of random data!" defense. More likely, you'll be thrown out on your ass. And if you do end up in legal proceedings, well, be careful about creating a reasonable facsimile of copyright infringement, lest the court impose a reasonable facsimile of judgment upon you. Judges have a funny sense of humor about that stuff sometimes.
User listening data is not really that useful as a tool for filing lawsuits. It is useful as a tool for tracking and potentially identifying leaks. For example: suppose User X listened to a new album ten days before it was actually released, and is friends (on the site) with User Y who listened to it twelve days before the release date, and User Y is friends with User Z whose profile matches up with an intern at the studio. Odds are that User Z -- the intern -- is the source of the leak.
To try to better fit your analogy, wouldn't this case be more like, say, holding the phone company responsible for the alcohol-to-kids store because they listed their telephone number and address? We all agree that this selling-to-anyone store is breaking the law... we just don't think telling people where the store is and what the store sells is illegal.
Well, actually you're making an important mistake by using an incorrect analogy.
If the phone book changed its title to "The Underage Alcohol Hookup Bay", changed its listings to a bunch of messages saying "ALCOHOL SOLD 2 N E 1 NO ID REQUIRED UNDER AGE A-OK!!!!!", etc., and then put a little label on the front saying "you kids be good and behave yourselves, we don't want any law-breaking in here", would they have any reasonable case for claiming innocence in any resulting underage alcohol acquisitions? Or would a judge say that they clearly intended to to facilitate such acquisitions? What if there were a law stating that giving a child instructions on how to obtain alcohol was, while not the same as providing that child with alcohol, still a crime? Such laws do exist in many countries with regard to copyright infringement, after all...
"Your favorite artist" sees just about zilch from CD sales, unless they're totally independent. If you want to support them, go see them in concert.
If "your favorite artist" is signed to a major label, or to an imprint of a major label, then he/she/they is/are in permanent debt slavery. Neither album sales nor concert ticket sales nor t-shirt sales nor anything else will remedy that; the outlay for concert tours comes from the label just like the outlay for recording, album production, distribution, etc.
You can't tell me that Apple doesn't have a record of the songs I purchased over the years.
They do have such a record, and a little Googling reveals lots of stories of people who had hard drive crashes, etc., and who talked to Apple's iTunes support, and who were given access to re-download everything they'd previously bought.
Because the moment your phone can provide an Internet connection to your laptop is the moment the bottom falls out of pricey "business on the go" plans which include a plug-in card for your laptop.
Read the qmail source code sometime. Every time the author wants to assure himself that data has been written to the disk, it calls fsync.
Except even that's not enough, and risks data loss. Consider the following from the OS X man page for fsync:
Note that while fsync() will flush all data from the host to the drive (i.e. the "permanent storage device"), the drive itself may not physically write the data to the platters for quite some time and it may be written in an out-of-order sequence.
Specifically, if the drive loses power or the OS crashes, the application may find that only some or none of their data was written. The disk drive may also re-order the data so that later writes may be present, while earlier writes are not.
This is not a theoretical edge case. This scenario is easily reproduced with real world workloads and drive power failures.
For applications that require tighter guarantees about the integrity of their data, Mac OS X provides the F_FULLFSYNC fcntl. The F_FULLFSYNC fcntl asks the drive to flush all buffered data to permanent storage. Applications, such as databases, that require a strict ordering of writes should use F_FULLFSYNC to ensure that their data is written in the order they expect. Please see fcntl(2) for more detail.
This is almost certainly done in the name of performance, just as at the FS level. Which raises the question of whether, down the road, further "performance improvements" will create a need for F_FULLFSYNC_NO_REALLY_I_MEAN_IT, F_FULLFSYNC_JUST_WRITE_THE_DATA_ALREADY, etc.
As has been pointed out to you, there were other sites reporting the same.
Well, no actually. What happened was that one site looked at the new headphones and found there was a chip in there. Instead of coming to the most obvious conclusion -- that supporting the custom functionality of the headphones would logically require some sort of hardware in there -- that outlet (iLounge) jumped straight to OH NOES A DRM CHIP.
Everybody else who reported this "story" went solely on that, building increasingly purely speculative reports about third-party manufacturers potentially getting DMCA notices. Notably, all of them also are, in terms of credibility, largely the online equivalent of the tabloids you see at the grocery store checkout proclaiming that Elvis and space aliens are consulting with Obama to fight the recession. Actually credible tech news outlets (e.g., those who actually do basic fact-finding before they publish) either didn't touch this "story" or reported something different than the prevailing DRM line. See Ars Technica's article -- which actually talked to relevant sources to get information -- for an example of how this type of reporting should be handled. A choice quote:
We're no fans of DRM here at Ars--in many cases it simply makes life more difficult for honest paying customers to use the things they pay for. But iPod accessory maker Scosche told Ars that the chip does not use authentication.
Amazing what happens when you actually follow basic common-sense procedures before running a story...
Slashdot Mirror
Well, the iPhone, iPod touch and Mac desktops and laptops run the same OS in the same sense that, say, a web server and an Android phone both run "Linux". Which is to say that the core bits are either the same or very similar, but supplemented in each case by different sets of drivers, userland applications, etc.
And yes, the version of the OS that runs on phones and iPods can multitask; any time one of those devices is powered on it's running multiple applications. Apple just doesn't distribute third-party applications which continue running when not active. They claim it's to suit the fact that those are resource- and battery-constrained devices, a claim which you are free to believe or disbelieve at your pleasure.
You're too late, there's already prior art.
I once experimented with producing summaries that people like you might appreciate. Turns out they don't work so well.
But the GPL isn't the arbiter of what is and is not a derivative work; the copyright laws of your country are. In the US, it's far from clear that dynamic linking creates a derivative work under the law (static linking doesn't come into play because it triggers copyright law in other ways). And if a court were to rule that dynamic linking does not create a derivative, the GPL would pretty much be toast.
You do know that there's an approved and free Skype application available, right?
And if you'd take five minutes to actually read the article you'd see that the FCC is pretty clearly aware of that, and that several of their questions are implicitly but still quite clearly saying "we know you've already approved VoIP stuff, so we're wondering why those apps were OK and Google's app isn't".
And some disasters are larger than others. A hurricane, for example, can devastate huge swathes of land because even a "normal"-sized hurricane is hundreds of miles across and can chug along for days or even weeks. But even the biggest, nastiest, meanest tornadoes don't get much above a mile wide and typically blow out very quickly. And those are the rarest of the rare; the combination of short lifespan and small size means tornadoes tend to be extremely localized in the damage they do. And thanks to low population density, the vast majority of twisters do no more than churn up empty. So even though there are hundreds of tornadoes each year, the likelihood of any particular inhabited place being hit is, well, tiny.
Add in the fact that hardening a building isn't terribly difficult (my small-town employer here has a datacenter that can take on anything we're likely to see, and a lot of stuff we're not likely to see) and I just don't see an issue. Oh, and the proposed site is in Manhattan, KS, which already has a nuclear reactor (on the Kansas State University campus) and which got swiped by an EF4 tornado last year. Some houses got chewed up, the university's wind erosion lab was destroyed (irony!), buildings on campus got damaged, and... um, that was about it. The stuff that needs to be secure is secure.
Browser rendering engines? In my application UI? It's more likely than you think, especially if you use Firefox, or any other application built around a XUL runtime. How many CSS-only exploits you heard of for them?
If you can read it, there's a fun discussion on the French Wikipedia over this sort of thing, with the suggestion that instead of "Américains" (literally, "Americans") the French-language Wikipedia should use "États-uniens" (literally, "United-Statesians"). Which brings up some fun replies (my loose translation is supplied here):
My personal favorite:
I think there are multiple ways to take this.
One is whether you can answer a couple simple questions: where is this program's executable located, and where is its configuration located? Though in theory there are standards for this, in practice they're not followed or don't really solve the problem. The executable may end up in /bin, /sbin, /usr/bin, /usr/sbin, /usr/local/bin, /opt/bin, /opt/local/bin, etc., etc. The configuration may be in a dotfile in your home directory, it may be in multiple files in one or more directories, it may be site-wide in /etc and customizable (or not) on a per-user basis, etc., etc.
The only platform I've seen tackle this in anything approaching a consistent way is OS X, where you know that there's a directory where the applications live, and you can easily predict where the configuration will be (Library/Application Support/(name of application), either top-level for global config or in the user's directory for per-user config). Simply getting a consistent standard for this on Linux distros would be a big win.
Organization of files created by end users is tougher; various platforms have tried different approaches -- "My Documents", specific locations for different types of media, etc. -- and so far nobody's really managed to beat this, with the result that every user either ends up with their own ad-hoc organizational system or (more commonly) ends up with a single directory holding "UNTITLED-1", "UNTITLED-2", and so on. And this isn't really related to users being dumb or not knowing how to use a computer; it's due to the fact that (1) we still impose a physical metaphor (a desktop, with files organized into folders inside) even though it's increasingly irrelevant and (2) we still haven't noticed that managing individual pieces and collections of data is a tedious and repetitive task, and that automating such tasks is a big part of why we use computers in the first place.
Apple usually updates the iPod line in the fall, just in time to gear up for the holiday shopping season where iPod sales go through the roof.
This is what we call a false dilemma.
While my personal opinion is going to stay personal for the moment, you're neglecting a variety of factors which affect the cost of health care. Most obvious is the ability of large entities to negotiate better rates; I have personal experience of working in health insurance, and saw this first-hand. When you represent a large group of patients, negotiating with a large group of doctors and/or hospitals, you can easily work out average costs which are far lower than any individual would manage but which still let everyone involved remain profitable. The federal government, incidentally, tends to get better rates than even the biggest private insurers, because it represents a staggeringly large pool of patients through Medicare.
Other factors -- reducing the number of middlemen, simplified interoperability, etc. -- would doubtless come into play as well, and none of these are "socialized magic". They're well-known and long-recognized market forces.
(incidentally, the largest contribution to the cost of health care is the cost of malpractice insurance premiums, which do not in any way correlate to the size of malpractice damage awards -- costs related to malpractice suits in the US have grown at roughly the rate of inflation in recent years, but the premiums for malpractice insurance have absolutely skyrocketed over the same period, in some cases rising by eight to ten times the amount of the corresponding rise in costs. This has nothing to do with what happens in courtrooms, and much to do with what happens in stock exchanges)
Personally, I think the higher profit-per-unit is a better bet in the long run, since it's a bit easier to survive a downturn in the market (when you're not making much on each unit sold, you need to sell a lot more to stay afloat -- that's why, for example, Dell is hurting pretty bad right now).
I mistyped -- meant "profit", not "profit margin" (too many arguments too early in the morning).
What I had in mind was a comparison like:
Which is, as far as I can tell, a reasonable pair of numbers to pull out of my ass (though nobody in the general public knows for sure how much Apple makes on hardware). This underscores the fact that you can make some pretty good money on hardware if you manage to market it correctly, and the fact that you don't have to be on everybody's home computer to be raking in the dough.
But is that their business model? Or is it a good business model compared to the one they currently have? Even setting aside the costs of supporting any old piece of hardware someone wants to run on, the difference between the OS business and the premium PC business is pretty significant. The former has a couple orders of magnitude more unit sales, but the latter has a couple orders of magnitude higher profit margins.
You're making the fatal mistake of assuming that, say, a court proceeding works according to the sorts of logical rules you're used to. This is by far the easiest way to end up on the wrong end of a court's decision.
If you work for a record company and you set out to create the impression that you're listening to and distributing bootleg copies of an album (ahead of its release date or not), I highly doubt that your boss will think it's a great prank if you pull the "honest, it was just a file of random data!" defense. More likely, you'll be thrown out on your ass. And if you do end up in legal proceedings, well, be careful about creating a reasonable facsimile of copyright infringement, lest the court impose a reasonable facsimile of judgment upon you. Judges have a funny sense of humor about that stuff sometimes.
Comments expressing alternate points of view are seemingly disappearing from the TechCrunch discussion thread. If TC feels it's necessary to delete comments which critique their reporting, perhaps we should be taking a closer look at that reporting...
You're looking at it the wrong way.
User listening data is not really that useful as a tool for filing lawsuits. It is useful as a tool for tracking and potentially identifying leaks. For example: suppose User X listened to a new album ten days before it was actually released, and is friends (on the site) with User Y who listened to it twelve days before the release date, and User Y is friends with User Z whose profile matches up with an intern at the studio. Odds are that User Z -- the intern -- is the source of the leak.
Well, actually you're making an important mistake by using an incorrect analogy.
If the phone book changed its title to "The Underage Alcohol Hookup Bay", changed its listings to a bunch of messages saying "ALCOHOL SOLD 2 N E 1 NO ID REQUIRED UNDER AGE A-OK!!!!!", etc., and then put a little label on the front saying "you kids be good and behave yourselves, we don't want any law-breaking in here", would they have any reasonable case for claiming innocence in any resulting underage alcohol acquisitions? Or would a judge say that they clearly intended to to facilitate such acquisitions? What if there were a law stating that giving a child instructions on how to obtain alcohol was, while not the same as providing that child with alcohol, still a crime? Such laws do exist in many countries with regard to copyright infringement, after all...
Australia.
Little-known fact: in Australia it is currently the year 2057. The International Date Line is funny like that.
If "your favorite artist" is signed to a major label, or to an imprint of a major label, then he/she/they is/are in permanent debt slavery. Neither album sales nor concert ticket sales nor t-shirt sales nor anything else will remedy that; the outlay for concert tours comes from the label just like the outlay for recording, album production, distribution, etc.
They do have such a record, and a little Googling reveals lots of stories of people who had hard drive crashes, etc., and who talked to Apple's iTunes support, and who were given access to re-download everything they'd previously bought.
Because the moment your phone can provide an Internet connection to your laptop is the moment the bottom falls out of pricey "business on the go" plans which include a plug-in card for your laptop.
Setting aside the performance implications of forcing fsync over and over, it's important to beat people over the head with the fact that fsync doesn't guarantee anything like what you think it does.
Except even that's not enough, and risks data loss. Consider the following from the OS X man page for fsync:
This is almost certainly done in the name of performance, just as at the FS level. Which raises the question of whether, down the road, further "performance improvements" will create a need for F_FULLFSYNC_NO_REALLY_I_MEAN_IT, F_FULLFSYNC_JUST_WRITE_THE_DATA_ALREADY, etc.
Well, no actually. What happened was that one site looked at the new headphones and found there was a chip in there. Instead of coming to the most obvious conclusion -- that supporting the custom functionality of the headphones would logically require some sort of hardware in there -- that outlet (iLounge) jumped straight to OH NOES A DRM CHIP.
Everybody else who reported this "story" went solely on that, building increasingly purely speculative reports about third-party manufacturers potentially getting DMCA notices. Notably, all of them also are, in terms of credibility, largely the online equivalent of the tabloids you see at the grocery store checkout proclaiming that Elvis and space aliens are consulting with Obama to fight the recession. Actually credible tech news outlets (e.g., those who actually do basic fact-finding before they publish) either didn't touch this "story" or reported something different than the prevailing DRM line. See Ars Technica's article -- which actually talked to relevant sources to get information -- for an example of how this type of reporting should be handled. A choice quote:
Amazing what happens when you actually follow basic common-sense procedures before running a story...