He's definitely from the generation that exhibited stress as heart attacks, so that's completely plausible. But I'm not too convinced that it couldn't be a fake. It's not like the government has lost their chance to look tough, and we've heard evidence that Lay expected at the beginning for Bush to bail him out. Considering everything the Bush administration has done, I think this being a fake is not something that couldn't happen, but rather something that probably didn't happen.
I suspect that they'll extend it to allow other forms of payment over time, but the real difference is that they don't hold your money like PayPal does; they act as an intermediary in transactions between financial accounts at other institutions.
It's possible that they'll start doing more of the PayPal features over time. There's no reason they couldn't do person-to-person transfers by accepting a credit card or check and doing direct deposit into the recipient's bank account. But I think they'll avoid actually storing users' money themselves, like PayPal does.
I'd generally agree that Linus matters to Linux about as much now as he has at any point since about '97. He hasn't really dictated direction since then, or before, because it met his needs, and the point of open source is that the direction comes from the needs of user-developers. It seems to me that these days, he understands a lot of the tricky sections, and explains issues when they arise, but his most important role is adjusting the pace of merges. It's a secondary effect; the subsystem maintainers actually make the decisions, but he's behind the scenes (along with Andrew Morton; they seem to act as checks on each other), providing the policy guidelines on what goes where when.
On the other hand, if you ask what would happen if Linus didn't pay any attention to Linux for a month or two, the answer has to be that almost everything would be the same, except that people would fumble around debugging a bit more, and no 2.6.x release would happen in that period. This is pretty clear, because he took about that long away from Linux when he wrote git, and there was hardly any effect.
Of course, I still don't think it makes sense to count him as not mattering, because he's still capable of singlehandedly writing revolutionary software if he feels inspired, and he's also got the following to popularize it.
The real first release was earlier
on
Quake is 10
·
· Score: 1
Some time before (end of February, according to Wikipedia), they'd released a test version, which had the engine working, only some of the graphics set up (the grenade launcher looked like a wooden stick), and three deathmatch levels. Anyone who still didn't know whether their computer could handle Quake on June 23rd was clearly out of the loop. And they'd also already missed the best deathmatches, because the test levels were much better for deathmatch than the actual released levels, which were more designed for single player. I was actually disappointed when Quake came out and all the qtest server disappeared.
Actually, the middle east is about the best part of Asia to get involved in a land war in. You just shouldn't try to occupy it afterwards. In most of the rest of Asia, you'll get wiped out long before you've occupied the territory.
Microcontroller-based systems are definitely popular these days, so it's clearly not dead. Pick up a DigiKey catalogue and look at just how many different microcontrollers they stock and sell in unit quantities; somebody's got to be using them. And you are finding postings for jobs requiring a university degree, which means there are people working at that level.
I'd never heard of a tech school program specializing in microcontrollers before, and there's definitely a substantial amount of amateur work in the area these days. I'd guess that the postings you're seeing are trying to eliminate people without formal training, and don't realize that there's anything else to include as an option other than a university degree. Have your students tried applying for jobs that ask for a university degree? (Half the time, job postings ask for things that they don't actually expect to get, just so that they can turn down people they don't want for being unqualified instead of for less clear reasons; there's the classic demand for longer experience in something than it's existed.)
Re:What about the traditional non-gamer crowd?
on
Wii-mote In Action
·
· Score: 4, Insightful
The Wii is really designed to interest the non-gamers. The point of the controller is basically to let games have people make movements that they actually make in real life, rather than traditional gaming, where the player is actually sitting still holding a small box and moving their fingers slightly. It lets them have a ping-pong game where, instead of using a bunch of knobs and buttons to play ping-pong, you play ping-pong to play ping-pong. This is obviously likely to be more popular with non-gamer ping-pong players.
I think Brain Age reflects this shift in thinking; you say things by saying things, and write by writing. When you're reading aloud, it's just like you're holding a small hardcover book and reading it. I think the goal of the Wii is to expand the concept of having the player action match the character action beyond fingertips and voice.
Another hint that it's not targetted at gamers: there's little hardware difference between the Gamecube and the Wii aside from the controller, especially compared to the difference between the Xbox and 360 or PS2 and PS3. The Gamecube is therefore already essentially sufficient for what they want the hardware to do, aside from the limited interaction with the controller.
It was only back in February that people in the Linux community were pointing out that the OSDL organization wasn't actually particularly useful to the community, beyond funnelling corporate money to a few worthy individual developers. They had a list of requests for things OSDL could do to actually be useful, and they also, for the first time, got a community representative on the board of directors.
If you actually look at the OSDL's stated mission, it's all about attracting corporate interest to Linux, not about actually getting any open source development done directly. It's still a valuable function, but if Oracle wants to interact with the community (like, for example, pushing Ubuntu's kernel patches through the review process and into the mainline kernel), OSDL isn't going to be particularly useful, assuming that Oracle has employees who're active in the community (like, for example, Randy Dunlap).
That was the theory. But in practice, if Y was even, the kernel was obsolete, while if Y was odd, the kernel was broken. Except, of course, 2.even.0, which was actually stable, but broke compatibility with the previous kernel that worked. And occasionally, 2.even was kept up-to-date because nobody could use 2.odd for development, because it didn't work at all. You could tell that the old model didn't actually work, because no distribution shipped any kernel that used that model; they all shipped 2.even with an arbitrary set of patches (generally hundreds) from 2.odd and elsewhere. With the new model, distros are shipping kernels with only a few patches, and those patches are getting merged upstream.
The stable kernels aren't remotely on the bleeding edge; they contain only features which have been tested over the past three months, after being filtered out of the bleeding-edge development as being things that have already stabilized and stand a good chance of being proven in three months. It's effectively very similar, except the development series isn't left known-broken and the stabilization process happens on a quick schedule, with stuff that isn't ready pushed off to the next cycle rather than delaying the current cycle. Also, the version numbers change by less (development gets -mm, -rc, or -git; stable series change the third digit by one instead of the second by two; and bugfix releases change the fourth digit instead of the third).
In this case, the new syscall is because a specific situation can be optimized compared to using the existing functions, but the more efficient function only works at all for certain special (but important) cases. In this case, the optimization is that copying data from outside of the program to outside of the program is more efficient if the data doesn't have to go through the program; obviously, this can't be used for the common case where the program is trying to use the data. The case that it helps a lot is when a program is sending data from the disk to the video card, or network to the disk, etc. With the new syscall, the data doesn't have to be copied into the program's address space (or the program's page tables changed to bring it in, which generally costs a TLB flush).
Programs that want to be backwards compatible can rely on getting an error result of ENOSYS when they try, and can then fall back to using the traditional method.
Amazon already also has partnerships where they set up transactions with other businesses (e.g., their whole used-book system). In some places (e.g., Boston) grocery chains are still doing online orders and deliveries. (It has a reasonable ROI if you already have the food storage; delivery people and a web site to get the segment of the market that just wants food to show up.) So the next step is for Amazon, after you've specified your location, to offer fresh food if they've got a partner in the area. I don't see it leading to a loss for either business, relative to the status quo, so it's a perfectly plausible move.
That sounds a lot like Magnatune. Okay, so John Buskman isn't an artist himself, but his wife is, and, in any case, the artists get a good deal: they retain copyright and (non-exclusive) licensing rights, and they get half of the purchase price (after credit card processing fees; and the whole charge for getting a manufactured CD goes to the printing and shipping people). Of course, artists mostly have to show up with the recording ready. But advertizing, distribution, and so forth are covered in the half that Magnatune gets, rather than being charged against the artist's share.
Furthermore, there seems to be a lot of communication among the Magnatune artists; a bunch of the folk/world groups consist largely of solo artists in various combinations. So it is, in a sense, a coop; but the hassle of being an online business, attracting attention, figuring out what makes a good deal, and so on still requires a few people interested in some essentially non-musical effort.
I may only hang out in civilized places like the linux-kernel mailing list (where the standards for flames have been set by Al Viro when he had a more stressful job, and nothing anybody can now write remotely compares...), but my experience is that anyone who tells you "RTFM" doesn't actually know anything and should be ignored. Don't tell these people off; it's just feeding the trolls. Anyone worth listening to will tell you exactly which documentation to read (and it's worth getting this citation, even if it's not helpful, because then you can quote the part that doesn't answer the question but is supposed to, and everybody has the same context for the discussion). It's not actually too uncommon that you'll have read the readme, the install guide, and man pages, and none of these are actually up-to-date; the thing you actually have to read is an obscure Wiki only mentioned on the mailing list info page. Or the version of the readme file that's... wait for it... fixed in CVS.
Sure, it seems like a reliable solution. But it doesn't work when you need it the most, like when all the air has been sucked out of the office. What are you going to use then?
The average screener wouldn't pay any attention, but they got the screening manager in there. The the screening manager probably wouldn't recognize the name, either, but the guy's behaving very strangely. (He claims to have mailed his driver's license to his destination, which is his home, for no reason that he feels like providing. So he's not going to be able to drive home from the airport when he gets there. And he's not going to have his license for a few days. And it couldn't be an accident.) Plus, there's a reporter hanging around outside, peering in the window. So it's obviously some sort of publicity stunt. The manager doesn't want to get caught doing the wrong thing, so he calls someone higher up to get an official ruling on the situation, so he's not left explaining his actions, whatever he does, and the official does a quick background check on the guy's claimed identity, and try to ID him, in case he's someone who should be prevented from flying (like a DHS agent testing no-fly-list effectiveness), and it's obvious from there what the guy is up to, and what they should do. I think it's reasonable to guess that whoever managers call (note that the article says the manager made a cell phone call while they had him) when there's a situation is at least clever enough to check Google, if not something more effective.
If he'd said that his wallet had been stolen, or that he'd accidentally left it in his luggage, which he'd already checked, or something logical, I'd buy that they'd use their standard procedure. But this situation just screams "you'd better figure out what's really going on here."
(And, in the story you cite, the security people did ID Senator Kennedy, and the computer told them not to let him fly. The computer didn't say that it's okay to let him fly if he's actually the US Senator, so they're not going to risk their jobs by second-guessing it.)
Also, it probably doesn't hurt that he's on the DHS's privacy advisory committee, according to the name on his ticket, and his employer (the Cato Institute) is a well-known think tank and has a high-res photo of him on their website.
What do you want to want to bet the security supervisor's phone call went:
"Any idea why a Jim Harper might be trying to fly without an ID? He says he mailed it home instead of carrying it with him."
"Jim Harper... like the Jim Harper? White, brown hair, balding, thin widow's peak, short beard and mustache, grey eyes?"
"Uh, yeah, that sounds like him."
"He's on the DHS privacy committee. Make sure he's not sneaking a fake bomb on the plane or anything, but don't keep him from flying. And don't let on that you know who he is."
I mean, the people checking him out probably had access to hard-to-fake photographic identification than anything he could possibly be carrying himself. And a quick Google reveals that his new book is about "How Identification is Overused and Misunderstood". It seems obvious that, if he attracts enough attention that somebody looks him up, he'll be given exactly the treatment that he got: pretend to ignore his identity and check him sufficiently thoroughly that you'd catch it if he had anything prohibited.
I think the correct balance of unlockables is to make the things you unlock be pointless, and have the things you do be interesting. I think Katamari Damacy did it right: there's the basic track of unlocking levels by playing each level reasonably well; there are "eternal" modes available by playing certain levels really well (same level, no time limit); then there's useless stuff that you get by being fast (shooting stars) or thorough (cousins) or obsessive (all of the items in the inventory). The only content you miss past eternal modes is the brief text descriptions of objects. The things you get by playing obsessively don't contribute to replay value; rather, if you're playing the game obsessively anyway, due to the replay value (you find the game actually entertaining), you find some new things.
Personally, I can't see how Microsoft can release a new version of their flagship product, and expect to get away with ignoring significant innovations made in competing implementations. (I.e., "ensure solubility")
The problem isn't even forks, anyway. It's forks that don't heal over time. It's okay that each distribution is somewhat different, because the differences between distros now are not greater than they were 5 years ago, and they are smaller in the ways that matter for portability of applications.
Furthermore, the stuff that's available everywhere in patched versions is generally under the GPL, which means that the patches from one distro are available to other distros if they happen to want them.
There was a case recently where a girl with a transplanted heart couldn't stay on the anti-rejection drugs, and they were able to remove the transplanted heart and restart her original heart, which had recovered while she wasn't using it. A Google search for "organ removed" finds a bunch of stories about it.
To be fair, that group also contains a number of ODF supporters. Or rather, a large portion of the membership of that group, unsatisfied with the available technology, came up with ODF. So it's entirely reasonable that Microsoft also join that group, so that their format gets at least one vote, and there's somebody to answer questions about it.
Slashdot Mirror
He's definitely from the generation that exhibited stress as heart attacks, so that's completely plausible. But I'm not too convinced that it couldn't be a fake. It's not like the government has lost their chance to look tough, and we've heard evidence that Lay expected at the beginning for Bush to bail him out. Considering everything the Bush administration has done, I think this being a fake is not something that couldn't happen, but rather something that probably didn't happen.
I suspect that they'll extend it to allow other forms of payment over time, but the real difference is that they don't hold your money like PayPal does; they act as an intermediary in transactions between financial accounts at other institutions.
It's possible that they'll start doing more of the PayPal features over time. There's no reason they couldn't do person-to-person transfers by accepting a credit card or check and doing direct deposit into the recipient's bank account. But I think they'll avoid actually storing users' money themselves, like PayPal does.
I'd generally agree that Linus matters to Linux about as much now as he has at any point since about '97. He hasn't really dictated direction since then, or before, because it met his needs, and the point of open source is that the direction comes from the needs of user-developers. It seems to me that these days, he understands a lot of the tricky sections, and explains issues when they arise, but his most important role is adjusting the pace of merges. It's a secondary effect; the subsystem maintainers actually make the decisions, but he's behind the scenes (along with Andrew Morton; they seem to act as checks on each other), providing the policy guidelines on what goes where when.
On the other hand, if you ask what would happen if Linus didn't pay any attention to Linux for a month or two, the answer has to be that almost everything would be the same, except that people would fumble around debugging a bit more, and no 2.6.x release would happen in that period. This is pretty clear, because he took about that long away from Linux when he wrote git, and there was hardly any effect.
Of course, I still don't think it makes sense to count him as not mattering, because he's still capable of singlehandedly writing revolutionary software if he feels inspired, and he's also got the following to popularize it.
Some time before (end of February, according to Wikipedia), they'd released a test version, which had the engine working, only some of the graphics set up (the grenade launcher looked like a wooden stick), and three deathmatch levels. Anyone who still didn't know whether their computer could handle Quake on June 23rd was clearly out of the loop. And they'd also already missed the best deathmatches, because the test levels were much better for deathmatch than the actual released levels, which were more designed for single player. I was actually disappointed when Quake came out and all the qtest server disappeared.
Actually, the middle east is about the best part of Asia to get involved in a land war in. You just shouldn't try to occupy it afterwards. In most of the rest of Asia, you'll get wiped out long before you've occupied the territory.
Microcontroller-based systems are definitely popular these days, so it's clearly not dead. Pick up a DigiKey catalogue and look at just how many different microcontrollers they stock and sell in unit quantities; somebody's got to be using them. And you are finding postings for jobs requiring a university degree, which means there are people working at that level.
I'd never heard of a tech school program specializing in microcontrollers before, and there's definitely a substantial amount of amateur work in the area these days. I'd guess that the postings you're seeing are trying to eliminate people without formal training, and don't realize that there's anything else to include as an option other than a university degree. Have your students tried applying for jobs that ask for a university degree? (Half the time, job postings ask for things that they don't actually expect to get, just so that they can turn down people they don't want for being unqualified instead of for less clear reasons; there's the classic demand for longer experience in something than it's existed.)
The Wii is really designed to interest the non-gamers. The point of the controller is basically to let games have people make movements that they actually make in real life, rather than traditional gaming, where the player is actually sitting still holding a small box and moving their fingers slightly. It lets them have a ping-pong game where, instead of using a bunch of knobs and buttons to play ping-pong, you play ping-pong to play ping-pong. This is obviously likely to be more popular with non-gamer ping-pong players.
I think Brain Age reflects this shift in thinking; you say things by saying things, and write by writing. When you're reading aloud, it's just like you're holding a small hardcover book and reading it. I think the goal of the Wii is to expand the concept of having the player action match the character action beyond fingertips and voice.
Another hint that it's not targetted at gamers: there's little hardware difference between the Gamecube and the Wii aside from the controller, especially compared to the difference between the Xbox and 360 or PS2 and PS3. The Gamecube is therefore already essentially sufficient for what they want the hardware to do, aside from the limited interaction with the controller.
It was only back in February that people in the Linux community were pointing out that the OSDL organization wasn't actually particularly useful to the community, beyond funnelling corporate money to a few worthy individual developers. They had a list of requests for things OSDL could do to actually be useful, and they also, for the first time, got a community representative on the board of directors.
If you actually look at the OSDL's stated mission, it's all about attracting corporate interest to Linux, not about actually getting any open source development done directly. It's still a valuable function, but if Oracle wants to interact with the community (like, for example, pushing Ubuntu's kernel patches through the review process and into the mainline kernel), OSDL isn't going to be particularly useful, assuming that Oracle has employees who're active in the community (like, for example, Randy Dunlap).
That was the theory. But in practice, if Y was even, the kernel was obsolete, while if Y was odd, the kernel was broken. Except, of course, 2.even.0, which was actually stable, but broke compatibility with the previous kernel that worked. And occasionally, 2.even was kept up-to-date because nobody could use 2.odd for development, because it didn't work at all. You could tell that the old model didn't actually work, because no distribution shipped any kernel that used that model; they all shipped 2.even with an arbitrary set of patches (generally hundreds) from 2.odd and elsewhere. With the new model, distros are shipping kernels with only a few patches, and those patches are getting merged upstream.
The stable kernels aren't remotely on the bleeding edge; they contain only features which have been tested over the past three months, after being filtered out of the bleeding-edge development as being things that have already stabilized and stand a good chance of being proven in three months. It's effectively very similar, except the development series isn't left known-broken and the stabilization process happens on a quick schedule, with stuff that isn't ready pushed off to the next cycle rather than delaying the current cycle. Also, the version numbers change by less (development gets -mm, -rc, or -git; stable series change the third digit by one instead of the second by two; and bugfix releases change the fourth digit instead of the third).
In this case, the new syscall is because a specific situation can be optimized compared to using the existing functions, but the more efficient function only works at all for certain special (but important) cases. In this case, the optimization is that copying data from outside of the program to outside of the program is more efficient if the data doesn't have to go through the program; obviously, this can't be used for the common case where the program is trying to use the data. The case that it helps a lot is when a program is sending data from the disk to the video card, or network to the disk, etc. With the new syscall, the data doesn't have to be copied into the program's address space (or the program's page tables changed to bring it in, which generally costs a TLB flush).
Programs that want to be backwards compatible can rely on getting an error result of ENOSYS when they try, and can then fall back to using the traditional method.
Amazon already also has partnerships where they set up transactions with other businesses (e.g., their whole used-book system). In some places (e.g., Boston) grocery chains are still doing online orders and deliveries. (It has a reasonable ROI if you already have the food storage; delivery people and a web site to get the segment of the market that just wants food to show up.) So the next step is for Amazon, after you've specified your location, to offer fresh food if they've got a partner in the area. I don't see it leading to a loss for either business, relative to the status quo, so it's a perfectly plausible move.
That sounds a lot like Magnatune. Okay, so John Buskman isn't an artist himself, but his wife is, and, in any case, the artists get a good deal: they retain copyright and (non-exclusive) licensing rights, and they get half of the purchase price (after credit card processing fees; and the whole charge for getting a manufactured CD goes to the printing and shipping people). Of course, artists mostly have to show up with the recording ready. But advertizing, distribution, and so forth are covered in the half that Magnatune gets, rather than being charged against the artist's share.
Furthermore, there seems to be a lot of communication among the Magnatune artists; a bunch of the folk/world groups consist largely of solo artists in various combinations. So it is, in a sense, a coop; but the hassle of being an online business, attracting attention, figuring out what makes a good deal, and so on still requires a few people interested in some essentially non-musical effort.
I may only hang out in civilized places like the linux-kernel mailing list (where the standards for flames have been set by Al Viro when he had a more stressful job, and nothing anybody can now write remotely compares...), but my experience is that anyone who tells you "RTFM" doesn't actually know anything and should be ignored. Don't tell these people off; it's just feeding the trolls. Anyone worth listening to will tell you exactly which documentation to read (and it's worth getting this citation, even if it's not helpful, because then you can quote the part that doesn't answer the question but is supposed to, and everybody has the same context for the discussion). It's not actually too uncommon that you'll have read the readme, the install guide, and man pages, and none of these are actually up-to-date; the thing you actually have to read is an obscure Wiki only mentioned on the mailing list info page. Or the version of the readme file that's... wait for it... fixed in CVS.
Sure, it seems like a reliable solution. But it doesn't work when you need it the most, like when all the air has been sucked out of the office. What are you going to use then?
Have you tried using the ruler thingy? It pops up a dialog box, but that's not what dialog boxes look like...
According to the release notes: glibc 2.3.2 w/NPTL, kernel 2.4, x.org 6.7. I think anything else is included.
The average screener wouldn't pay any attention, but they got the screening manager in there. The the screening manager probably wouldn't recognize the name, either, but the guy's behaving very strangely. (He claims to have mailed his driver's license to his destination, which is his home, for no reason that he feels like providing. So he's not going to be able to drive home from the airport when he gets there. And he's not going to have his license for a few days. And it couldn't be an accident.) Plus, there's a reporter hanging around outside, peering in the window. So it's obviously some sort of publicity stunt. The manager doesn't want to get caught doing the wrong thing, so he calls someone higher up to get an official ruling on the situation, so he's not left explaining his actions, whatever he does, and the official does a quick background check on the guy's claimed identity, and try to ID him, in case he's someone who should be prevented from flying (like a DHS agent testing no-fly-list effectiveness), and it's obvious from there what the guy is up to, and what they should do. I think it's reasonable to guess that whoever managers call (note that the article says the manager made a cell phone call while they had him) when there's a situation is at least clever enough to check Google, if not something more effective.
If he'd said that his wallet had been stolen, or that he'd accidentally left it in his luggage, which he'd already checked, or something logical, I'd buy that they'd use their standard procedure. But this situation just screams "you'd better figure out what's really going on here."
(And, in the story you cite, the security people did ID Senator Kennedy, and the computer told them not to let him fly. The computer didn't say that it's okay to let him fly if he's actually the US Senator, so they're not going to risk their jobs by second-guessing it.)
Also, it probably doesn't hurt that he's on the DHS's privacy advisory committee, according to the name on his ticket, and his employer (the Cato Institute) is a well-known think tank and has a high-res photo of him on their website.
What do you want to want to bet the security supervisor's phone call went:
"Any idea why a Jim Harper might be trying to fly without an ID? He says he mailed it home instead of carrying it with him."
"Jim Harper... like the Jim Harper? White, brown hair, balding, thin widow's peak, short beard and mustache, grey eyes?"
"Uh, yeah, that sounds like him."
"He's on the DHS privacy committee. Make sure he's not sneaking a fake bomb on the plane or anything, but don't keep him from flying. And don't let on that you know who he is."
I mean, the people checking him out probably had access to hard-to-fake photographic identification than anything he could possibly be carrying himself. And a quick Google reveals that his new book is about "How Identification is Overused and Misunderstood". It seems obvious that, if he attracts enough attention that somebody looks him up, he'll be given exactly the treatment that he got: pretend to ignore his identity and check him sufficiently thoroughly that you'd catch it if he had anything prohibited.
I used to have an AT machine, but an SO package just doesn't fit my preferences, if you know what I mean.
I think the correct balance of unlockables is to make the things you unlock be pointless, and have the things you do be interesting. I think Katamari Damacy did it right: there's the basic track of unlocking levels by playing each level reasonably well; there are "eternal" modes available by playing certain levels really well (same level, no time limit); then there's useless stuff that you get by being fast (shooting stars) or thorough (cousins) or obsessive (all of the items in the inventory). The only content you miss past eternal modes is the brief text descriptions of objects. The things you get by playing obsessively don't contribute to replay value; rather, if you're playing the game obsessively anyway, due to the replay value (you find the game actually entertaining), you find some new things.
Personally, I can't see how Microsoft can release a new version of their flagship product, and expect to get away with ignoring significant innovations made in competing implementations. (I.e., "ensure solubility")
But then, if two planes are headed for a collision, and one of them is upside down...
The problem isn't even forks, anyway. It's forks that don't heal over time. It's okay that each distribution is somewhat different, because the differences between distros now are not greater than they were 5 years ago, and they are smaller in the ways that matter for portability of applications.
Furthermore, the stuff that's available everywhere in patched versions is generally under the GPL, which means that the patches from one distro are available to other distros if they happen to want them.
There was a case recently where a girl with a transplanted heart couldn't stay on the anti-rejection drugs, and they were able to remove the transplanted heart and restart her original heart, which had recovered while she wasn't using it. A Google search for "organ removed" finds a bunch of stories about it.
To be fair, that group also contains a number of ODF supporters. Or rather, a large portion of the membership of that group, unsatisfied with the available technology, came up with ODF. So it's entirely reasonable that Microsoft also join that group, so that their format gets at least one vote, and there's somebody to answer questions about it.