Much of the internet is built on a model of reasonably open trust. This proved to not be a mistake, but a particularly galling one, which has required patch after patch.
The problem, as I see it, occurred starting in about the mid 90s. At this point, what the internet actually was, was clear to all. Making assumptions of trustworthiness in 1985 was still quite reasonable: it was possible that all meaningful internet connections were to continue to be monitored for bad behavior manually and actioned when appropriate. It wasn't what was happening, but it wasn't lunacy.
In the mid to late 90s, once the majority of the really gullible things were beaten out of everything, things appeared to be kinda looking up- we were at least on the correct trajectory. Queue another massive overdose of functionality. The early versions of IE would just run any link as appropriate. You could provide a link to C:\windows\notepad.exe, and clicking it would run notepad. Or a deltree on your C drive. Unix land, while not as degenerate, was still busy taking URLs as commands, browsing all over the root filesystem, and generally behaving like amateur hour. Every new tech that got added was riddled with security problems that were reasonable obvious, and they were still adopted at absolute lightning speed.
Technologies were obsoleted almost as fast as it took them to hammer out their bugs. The idea of passing code from server to client caught on, but unlike the prior iterations of this, there was no reason to actually TRUST the server- sure, you might trust microsoft.com, but do you trust $RANDOM_ADDRESS.net?
Something like SPECTRE wouldn't even be that interesting if the underlying assumption wasn't that you were downloading and running code everywhere you pointed a browser to.
The security overlay on all of this can be heavy at times. It is also frequently misguided, which makes much of the ire. See pretty much anything related to passwords for a great example of something that doesn't buy much security at the cost of a massive amount of usability (and goes backwards if it starts asking what school you went to, and then gives access to anyone who can guess that, a fact you cannot change). Even automated systems like SSL can ultimately be mangled by someone dedicated to the task.
Overall, much of the security burden is based around some bad choices early on, but almost everything that weighs us down now is a result of continuing to make bad choices.
It's not the anonymous subject's assessment of forks that is silly, it's that it was the ONLY hypnagogic element that managed to pass the filters of both Horowitz (the relaying researcher) and Oberhaus (the article writer), and get reported in the article. A million ideas, but the Critical Theory one is the one that bubbles up?
This is being peddled as a way to increase creativity and by extension help the world in some fashion, hence its write-up on motherboard and article on slashdot, and presumably also hence its funding at MIT. Everyone reading this is hoping it lets someone solve a problem of a thousand generations, not conflate Roman utensils with out-of-fashion actions of certain select European states in the nineteenth century.
The article allllllmost gets through itself without going into politics, but out of the all the examples in the infinite noosphere this machine is meant to probe, the one that made the cut was the phrase "forks are colonialism". - “I asked him about it when he woke up,” Horowitz told me. “He said, ‘Oh at home I eat food with my hands and here I have a sharp, cold metal instrument that I use to stab the food that goes into me. I guess it has a colonial energy.’ -
The idea of hypnagogia exploitation is interesting, but if all we're going to get is postmodern poetry fragments that decry 2400 year old eating instruments, maybe our thought leaders should stick to microdosing LSD for ideas.
Maybe you know a trick I don't with XFCE. When I set that thing up to be vertical and then open like a dozen programs, they basically become little icons on the taskbar. If I do the same thing horizontally, I can read their names next to their icons. Having the text, which flows left to right in English, on a vertical bar seems to work very poorly compared to a horizontal one.
My normal setup features one vertical monitor, and on that guy I have plenty of screenspace at the bottom for a horizontal taskbar. But when I'm just on a landscape oriented monitor, I have to choose between legible title text and slightly superior space usage. I usually go with the former.
Why do phones always get blamed for supporting portrait mode, when the real issue is that pretty much no playback site properly handles it? I can put it on a vertically oriented monitor, and it will play it back as: A huge black bar for the top third of the monitor A tiny black bar in the middle third left of the monitor The actual portrait mode thing in the middle of the middle third A tiny black bar in the middle third right of the monitor A huge black bar for the bottom third of the monitor
This is ludicrous. A portrait mode video on a portrait mode monitor should be full screen or very nearly such. Certainly it works on a monitor (or a phone) if you manage to get the mp4 locally.
Portrait is a little lame because it has less form factors that support it, but the big problem is pretty much lack of ability to get it to display properly on portrait display, which absolutely exist.
Yes, it can definitely disable TSX functionality. Like when TSX launched with Haswell, but it was fucked up, so they disabled it with microcode. Or when they fixed the Haswell problem and launched it with Broadwell, but it was fucked up, so they disabled with microcode. Skylake, of course, fixed the Broadewell problem...
But now you're saying that TSX is the issue again? And that it needs to be disabled AGAIN? How many fucking chip generations do we have to go through before transactional fucking memory doesn't get patched out because OOPS it crashes the box or OOPS it gives double-super-ultra-root to enemy spies?
Are you SURE that TSX is the issue? I didn't see anything in the article about TSX being the problem, but I'm not really read up on this.
The red/blue split is such that the states can't agree on anything important in numbers worth worrying the federal government. The supreme court's fundamental supremacy is at this point not even questioned: of course they can strip away state level abortion laws based on individual-rights reasoning, while also upholding the right of the federal government to ban substances such as cannabis, a power nowhere enumerated. Each controversial ruling, whether you are in favor or opposed, has exactly one purpose, one common constant: the federal apparatus enforces homogeneity on the states.
Things that are still mostly within the power of the states force anyone who wants policy to deal more closely with those affected by said policy- it may be cheaper to buy off a governor, it may be easier to get a state constitutional amendment, it may be easier to get your message out, but the impact is more limited. There aren't many ways to offer pushback against a large powerful conglomeration, but the state-level breakdown does offer some hedge.
You also have experimentation: legal cannabis, for instance, has not imploded any states. This proves that a VAST number of the alarmist theories posited by anti-cannabis activists were overblown, and this will be taken into account by the voters, if by no one else. Most people have come around on the idea that cannabis should be legal in Colorado, but if you ask pro-legalization people about whether Utah should be allowed to keep cannabis illegal in a world where it no longer has "federally banned" status, you are going to get a variety of answers, some drawing from inalienable rights and others drawing from traditional separation of powers.
Also: the net neutrality executive order seems open to subversion by some shenanigans involving shell companies, but assuming that this doesn't occur or is punished if and when it does, it's absolutely the type of thing that your standard voting Republican should get behind.
- It doesn't affect what companies in the marketplace can do, it only affects how the state government makes its contracts. As such, it has no effect on the, err, "mom and pop ISPs". I have no idea if these actually are real, but they were certainly used by anti-NN apologists.
- It doesn't have any of the carrier reclassification or potential first amendment concerns of the actual FCC net neutrality that was put in place. This was also used to drag net neutrality through the mud, and was a legitimate concern- of course, if this had been the ACTUAL concern, the FCC could simply have offered clarification or policy that ameliorated these concerns.
This executive order doesn't offer the feeble handholds that anti-NN talking points could grasp for the FCC version. I bet more states could pull this trick or something similar, if they wanted.
Yes, of course the infrastructure monopolists know what they want. Using states as a test bed will actually produce some data, if anyone still cares about that. It's not a bad idea for getting data on internet regulation, and it is really coming into its own regarding cannabis regulation. If anyone is looking for information, it could inform future decisions.
But you are correct that right now, everyone has their heels dug in and it has nothing to do with data. The top level net neutrality overturn was a battle of one group of infrastructure monopolists versus content monopolists, and each backed a different horse in a high stakes presidential race. The fact that one of these is marginally better for our wallets is happenstance, nothing more.
> Adding encryption to cameras will just slow them down Yes, and if this is a problem you need to add enough hardware to the phone to accomplish the encryption within the needed read/write time of the camera. Given how little power this takes relative to modern processing power, and given the specialized solutions available in most modern chips, this is exactly what is being added.
> and cause compatibility problems with photo editing software. No, by the time software sees the files they are unencrypted. This is not a concern.
Phones tend to control the entire stack, so adding encrypting to them was mostly invisible to the user, except for the obvious passcode or password. Because cameras are meant to be much more functional, you would need to have a setting on them that allows for optional operation in encrypted mode, with the user being left with the hassle of extracting data from an encrypted SD card. This also means that there would need to be an agreed upon standard that all cameras and other things interfacing with their cards could use, which is probably the real reason this is taking awhile- there's no commonly understood standard for this sort of thing across filesystems in the real world.
Cameras getting encryption is a bit less important than phones getting encryption, as well- because many cameras don't store anything EXCEPT to removable media, whereas phones are a stack of everything all in one place. Still would be nice though.
If you buy 100 worth of bitcoins, and it goes up by 10%, and then you pay your 110 dollar bill, you have made 10 dollars. That counts as $10 of income, not $110 of income. If this was some longer term thing then a lower capital gain rate applies.
Yes, like those, except usable somewhere besides in person. You can't use any of gold, silver, or cash online- you have to use some intermediary, all of which have the disadvantages described in previous posts.
So no one should be allowed to gather high scores, and if anyone does so, we should just assume that all the high scores will be cheaters so it is pointless?
I don't know anything about the topic of the article, but I know that if you don't police high score tables, liars and cheaters will be the only ones at the top, without fail. Just because you don't see the point of a high score table doesn't mean there are none who do.
Eh, I can't imagine they would build it from Chinese parts if the explicit worry is Chinese interference. There's a trusted foundry project for a reason, and it's not because we can build stuff more efficiently than they can.
--- "Trump is totally, unbelievably fucked......when the adults are back in charge in Washington, there are going to be SERIOUS, life-changing repercussions for Putin and his friends.... This is going to be FUN to watch." ---
Good fucking grief, the party-rah-rah-rah in this. Your 2018 is going to be as disappointing as your 2016.
> They contain only the window title and a close button
This statement is only marginally true in GNOME, and a complete farce everywhere else. I have six window controls there in XFCE, as well as a title. Why would someone think that it only has a close button? Is it because that guy is full time GNOME? Maybe rethink the direction that led you there, instead of doubling down on breaking the interface harder.
In XFCE, the window controls are "pin to desktop" (the window stays in position when you change desktop), "window menu" (allows always on top/always on bottom, in addition to other window choices), maximize/unmaximize, minimize, roll up (shrinks the window to the title bar) and close. That's how a standard looking X anything looks in XFCE.
But of course, some of what I use are GNOME applications. These forcefully discard all this useful stuff. They replace the close button with a big gray "X" and everything else with stupid glyphs they copied out of whatever the worst version of windows they have is. Because Microsoft is doing it, they have a dark gray on light gray scheme, instead of a user-configurable one. And naturally, they don't have the pin options and other controls that I want.
I'm sure that as this disease spreads, it will not be done cleverly. If it hits LibreOffice, I'm sure it will detect that the computer DOES support the GNOME awfulness, and use that (yes yes, it could be done correctly and seamlessly and display the GNOME crap for the GNOME users, but seriously, do you think they will go that path?).
The funny part is that this is actually a debate between:
1- One giant gray double or triple height bar with uncustomizable icons and a limited set of useful menu controls. 2- Two normal height bars taking up less space on the screen, one obeying system standards that the user can configure if he so desires and containing window options, and the other obeying the logic of the application designer with all the options logically arranged.
The only reason there is wasted space in the first place is because of whatever bland-flavor-of-the-month UI choices GNOME is busy aping.
A cybercriminal is committing a crime that fundamentally relies on computers: if you took the computer away, there would be no crime. The topic of this is hackers who illegally deploy software on a computer that illegally destroys data on that computer and demands that you convert money to a computer currency and pay it to a network of computers (because they all use bitcoin- they just started denominating the ransom demands in dollars so people can actually pay the ransom, because if you meant to ask someone for two thousand dollars and demand 1 btc, they'll never pay).
In the other cases, the fact that a computer is involved is incidental. The actions could exist without computers, they would just be less efficient. The realworld equivalent of ransomware doesn't exist, because people don't blindly follow orders from mysterious actors in the same literal fashion that computers do.
Given how generally impossible it is to run a Hackintosh with an AMD CPU, I imagine those folks will just issue more patches to their stack of stuff required to get OS X up and running on AMD. I really feel that AMD Hackintosh users are a trivial minority of Hackintosh users, who are in turn a trivial minority of OS X users, who are in turn a reasonable minority of desktop/server/lappie users.
Basically, they'll do what they have always done to get OS X to run on AMD: hack at it awhile and hopefully beat something into shape.
Slashdot Mirror
Much of the internet is built on a model of reasonably open trust. This proved to not be a mistake, but a particularly galling one, which has required patch after patch.
The problem, as I see it, occurred starting in about the mid 90s. At this point, what the internet actually was, was clear to all. Making assumptions of trustworthiness in 1985 was still quite reasonable: it was possible that all meaningful internet connections were to continue to be monitored for bad behavior manually and actioned when appropriate. It wasn't what was happening, but it wasn't lunacy.
In the mid to late 90s, once the majority of the really gullible things were beaten out of everything, things appeared to be kinda looking up- we were at least on the correct trajectory. Queue another massive overdose of functionality. The early versions of IE would just run any link as appropriate. You could provide a link to C:\windows\notepad.exe, and clicking it would run notepad. Or a deltree on your C drive. Unix land, while not as degenerate, was still busy taking URLs as commands, browsing all over the root filesystem, and generally behaving like amateur hour. Every new tech that got added was riddled with security problems that were reasonable obvious, and they were still adopted at absolute lightning speed.
Technologies were obsoleted almost as fast as it took them to hammer out their bugs. The idea of passing code from server to client caught on, but unlike the prior iterations of this, there was no reason to actually TRUST the server- sure, you might trust microsoft.com, but do you trust $RANDOM_ADDRESS.net?
Something like SPECTRE wouldn't even be that interesting if the underlying assumption wasn't that you were downloading and running code everywhere you pointed a browser to.
The security overlay on all of this can be heavy at times. It is also frequently misguided, which makes much of the ire. See pretty much anything related to passwords for a great example of something that doesn't buy much security at the cost of a massive amount of usability (and goes backwards if it starts asking what school you went to, and then gives access to anyone who can guess that, a fact you cannot change). Even automated systems like SSL can ultimately be mangled by someone dedicated to the task.
Overall, much of the security burden is based around some bad choices early on, but almost everything that weighs us down now is a result of continuing to make bad choices.
It's not the anonymous subject's assessment of forks that is silly, it's that it was the ONLY hypnagogic element that managed to pass the filters of both Horowitz (the relaying researcher) and Oberhaus (the article writer), and get reported in the article. A million ideas, but the Critical Theory one is the one that bubbles up?
This is being peddled as a way to increase creativity and by extension help the world in some fashion, hence its write-up on motherboard and article on slashdot, and presumably also hence its funding at MIT. Everyone reading this is hoping it lets someone solve a problem of a thousand generations, not conflate Roman utensils with out-of-fashion actions of certain select European states in the nineteenth century.
The article allllllmost gets through itself without going into politics, but out of the all the examples in the infinite noosphere this machine is meant to probe, the one that made the cut was the phrase "forks are colonialism".
-
“I asked him about it when he woke up,” Horowitz told me. “He said, ‘Oh at home I eat food with my hands and here I have a sharp, cold metal instrument that I use to stab the food that goes into me. I guess it has a colonial energy.’
-
The idea of hypnagogia exploitation is interesting, but if all we're going to get is postmodern poetry fragments that decry 2400 year old eating instruments, maybe our thought leaders should stick to microdosing LSD for ideas.
Maybe you know a trick I don't with XFCE. When I set that thing up to be vertical and then open like a dozen programs, they basically become little icons on the taskbar. If I do the same thing horizontally, I can read their names next to their icons. Having the text, which flows left to right in English, on a vertical bar seems to work very poorly compared to a horizontal one.
My normal setup features one vertical monitor, and on that guy I have plenty of screenspace at the bottom for a horizontal taskbar. But when I'm just on a landscape oriented monitor, I have to choose between legible title text and slightly superior space usage. I usually go with the former.
Why do phones always get blamed for supporting portrait mode, when the real issue is that pretty much no playback site properly handles it? I can put it on a vertically oriented monitor, and it will play it back as:
A huge black bar for the top third of the monitor
A tiny black bar in the middle third left of the monitor
The actual portrait mode thing in the middle of the middle third
A tiny black bar in the middle third right of the monitor
A huge black bar for the bottom third of the monitor
This is ludicrous. A portrait mode video on a portrait mode monitor should be full screen or very nearly such. Certainly it works on a monitor (or a phone) if you manage to get the mp4 locally.
Portrait is a little lame because it has less form factors that support it, but the big problem is pretty much lack of ability to get it to display properly on portrait display, which absolutely exist.
"If you don't want that, don't get it."
Please link me to a nice 4:3 monitor I can buy, or a nice laptop with those proportions.
You will find that they only technically exist. It is very difficult to get any monitors that are not very short.
> is by disabling Intel's TSX functionality
fucking AGAIN? Really????
> (which I believe microcode can do)
Yes, it can definitely disable TSX functionality. Like when TSX launched with Haswell, but it was fucked up, so they disabled it with microcode.
Or when they fixed the Haswell problem and launched it with Broadwell, but it was fucked up, so they disabled with microcode.
Skylake, of course, fixed the Broadewell problem...
But now you're saying that TSX is the issue again? And that it needs to be disabled AGAIN? How many fucking chip generations do we have to go through before transactional fucking memory doesn't get patched out because OOPS it crashes the box or OOPS it gives double-super-ultra-root to enemy spies?
Are you SURE that TSX is the issue? I didn't see anything in the article about TSX being the problem, but I'm not really read up on this.
The red/blue split is such that the states can't agree on anything important in numbers worth worrying the federal government. The supreme court's fundamental supremacy is at this point not even questioned: of course they can strip away state level abortion laws based on individual-rights reasoning, while also upholding the right of the federal government to ban substances such as cannabis, a power nowhere enumerated. Each controversial ruling, whether you are in favor or opposed, has exactly one purpose, one common constant: the federal apparatus enforces homogeneity on the states.
Things that are still mostly within the power of the states force anyone who wants policy to deal more closely with those affected by said policy- it may be cheaper to buy off a governor, it may be easier to get a state constitutional amendment, it may be easier to get your message out, but the impact is more limited. There aren't many ways to offer pushback against a large powerful conglomeration, but the state-level breakdown does offer some hedge.
You also have experimentation: legal cannabis, for instance, has not imploded any states. This proves that a VAST number of the alarmist theories posited by anti-cannabis activists were overblown, and this will be taken into account by the voters, if by no one else. Most people have come around on the idea that cannabis should be legal in Colorado, but if you ask pro-legalization people about whether Utah should be allowed to keep cannabis illegal in a world where it no longer has "federally banned" status, you are going to get a variety of answers, some drawing from inalienable rights and others drawing from traditional separation of powers.
Also: the net neutrality executive order seems open to subversion by some shenanigans involving shell companies, but assuming that this doesn't occur or is punished if and when it does, it's absolutely the type of thing that your standard voting Republican should get behind.
- It doesn't affect what companies in the marketplace can do, it only affects how the state government makes its contracts. As such, it has no effect on the, err, "mom and pop ISPs". I have no idea if these actually are real, but they were certainly used by anti-NN apologists.
- It doesn't have any of the carrier reclassification or potential first amendment concerns of the actual FCC net neutrality that was put in place. This was also used to drag net neutrality through the mud, and was a legitimate concern- of course, if this had been the ACTUAL concern, the FCC could simply have offered clarification or policy that ameliorated these concerns.
This executive order doesn't offer the feeble handholds that anti-NN talking points could grasp for the FCC version. I bet more states could pull this trick or something similar, if they wanted.
Yes, of course the infrastructure monopolists know what they want. Using states as a test bed will actually produce some data, if anyone still cares about that. It's not a bad idea for getting data on internet regulation, and it is really coming into its own regarding cannabis regulation. If anyone is looking for information, it could inform future decisions.
But you are correct that right now, everyone has their heels dug in and it has nothing to do with data. The top level net neutrality overturn was a battle of one group of infrastructure monopolists versus content monopolists, and each backed a different horse in a high stakes presidential race. The fact that one of these is marginally better for our wallets is happenstance, nothing more.
> Adding encryption to cameras will just slow them down
Yes, and if this is a problem you need to add enough hardware to the phone to accomplish the encryption within the needed read/write time of the camera. Given how little power this takes relative to modern processing power, and given the specialized solutions available in most modern chips, this is exactly what is being added.
> and cause compatibility problems with photo editing software.
No, by the time software sees the files they are unencrypted. This is not a concern.
Phones tend to control the entire stack, so adding encrypting to them was mostly invisible to the user, except for the obvious passcode or password. Because cameras are meant to be much more functional, you would need to have a setting on them that allows for optional operation in encrypted mode, with the user being left with the hassle of extracting data from an encrypted SD card. This also means that there would need to be an agreed upon standard that all cameras and other things interfacing with their cards could use, which is probably the real reason this is taking awhile- there's no commonly understood standard for this sort of thing across filesystems in the real world.
Cameras getting encryption is a bit less important than phones getting encryption, as well- because many cameras don't store anything EXCEPT to removable media, whereas phones are a stack of everything all in one place. Still would be nice though.
If you buy 100 worth of bitcoins, and it goes up by 10%, and then you pay your 110 dollar bill, you have made 10 dollars. That counts as $10 of income, not $110 of income. If this was some longer term thing then a lower capital gain rate applies.
> You mean like gold, silver, or cash?
Yes, like those, except usable somewhere besides in person. You can't use any of gold, silver, or cash online- you have to use some intermediary, all of which have the disadvantages described in previous posts.
So no one should be allowed to gather high scores, and if anyone does so, we should just assume that all the high scores will be cheaters so it is pointless?
I don't know anything about the topic of the article, but I know that if you don't police high score tables, liars and cheaters will be the only ones at the top, without fail. Just because you don't see the point of a high score table doesn't mean there are none who do.
Eh, I can't imagine they would build it from Chinese parts if the explicit worry is Chinese interference. There's a trusted foundry project for a reason, and it's not because we can build stuff more efficiently than they can.
Are you being a posterchild for divisiveness ironically, or is it just a 24/7 state of being for you?
Sigh.
--- ...when the adults are back in charge in Washington, there are going to be SERIOUS, life-changing repercussions for Putin and his friends.... This is going to be FUN to watch."
"Trump is totally, unbelievably fucked...
---
Good fucking grief, the party-rah-rah-rah in this. Your 2018 is going to be as disappointing as your 2016.
> They contain only the window title and a close button
This statement is only marginally true in GNOME, and a complete farce everywhere else. I have six window controls there in XFCE, as well as a title. Why would someone think that it only has a close button? Is it because that guy is full time GNOME? Maybe rethink the direction that led you there, instead of doubling down on breaking the interface harder.
Ok, try this:
chrome://settings/
Then use the slider to enable "use system titles and borders".
This works to get a regular header in chrome, chromium, and iridium in XFCE- does it work for you too?
In XFCE, if you press the system menu (one of the six standard window decorations), you have a resize options, no alt keys required!
In XFCE, the window controls are "pin to desktop" (the window stays in position when you change desktop), "window menu" (allows always on top/always on bottom, in addition to other window choices), maximize/unmaximize, minimize, roll up (shrinks the window to the title bar) and close. That's how a standard looking X anything looks in XFCE.
But of course, some of what I use are GNOME applications. These forcefully discard all this useful stuff. They replace the close button with a big gray "X" and everything else with stupid glyphs they copied out of whatever the worst version of windows they have is. Because Microsoft is doing it, they have a dark gray on light gray scheme, instead of a user-configurable one. And naturally, they don't have the pin options and other controls that I want.
I'm sure that as this disease spreads, it will not be done cleverly. If it hits LibreOffice, I'm sure it will detect that the computer DOES support the GNOME awfulness, and use that (yes yes, it could be done correctly and seamlessly and display the GNOME crap for the GNOME users, but seriously, do you think they will go that path?).
The funny part is that this is actually a debate between:
1- One giant gray double or triple height bar with uncustomizable icons and a limited set of useful menu controls.
2- Two normal height bars taking up less space on the screen, one obeying system standards that the user can configure if he so desires and containing window options, and the other obeying the logic of the application designer with all the options logically arranged.
The only reason there is wasted space in the first place is because of whatever bland-flavor-of-the-month UI choices GNOME is busy aping.
Is tether actually backed by USD? I just thought it was a giant scam.
A cybercriminal is committing a crime that fundamentally relies on computers: if you took the computer away, there would be no crime. The topic of this is hackers who illegally deploy software on a computer that illegally destroys data on that computer and demands that you convert money to a computer currency and pay it to a network of computers (because they all use bitcoin- they just started denominating the ransom demands in dollars so people can actually pay the ransom, because if you meant to ask someone for two thousand dollars and demand 1 btc, they'll never pay).
In the other cases, the fact that a computer is involved is incidental. The actions could exist without computers, they would just be less efficient. The realworld equivalent of ransomware doesn't exist, because people don't blindly follow orders from mysterious actors in the same literal fashion that computers do.
Given how generally impossible it is to run a Hackintosh with an AMD CPU, I imagine those folks will just issue more patches to their stack of stuff required to get OS X up and running on AMD. I really feel that AMD Hackintosh users are a trivial minority of Hackintosh users, who are in turn a trivial minority of OS X users, who are in turn a reasonable minority of desktop/server/lappie users.
Basically, they'll do what they have always done to get OS X to run on AMD: hack at it awhile and hopefully beat something into shape.