"Right now IPv6 is *not* backwards-compatible with IPv4 because, unless to translate it to IPv4 (using whatever method you choose) you can't connect to IPv4 devices."
Compare it with NAT'ing. A private 192.168 v4 address cant talk directly to yahoo.com. You need either a proxy or a NAT router that will translate your 192.168 address into a public internet address.
"If you move to IPv6, but you still need to use an IPv4 IP"
Someone, somewhere, needs to use an IPv4 IP (obviously, as your question assumes that yahoo wont respond to ipv6). You dont.
Your gateway could do it. A routable protocol translator could do it (route packets with the::ffff:0:0 address to the translator that merely strips the extra bits, just like you route public packets through your NAT gateway). A proxy could do it (all of which are available in one form or another).
So if you're saying that nothing inbetween is allowed to touch the packets, then, right, you cant use IPv6 to access yahoo.com. That would include IPv4 NAT too tho.
If, on the other hand, you mean that a machine can have IPv6 only, and that the network structure can be set up to transparently translate IPv6 packets to IPv4, then yes, that's possible.
It is, however, easier and possibly more reliable to keep the old NAT structure in place for v4-only accesses while gaining the benefits of direct host addressing that v6 gives you by running a transition dual-protocol network in the near future (thus giving you the ability to, for example, ssh directly to 2002:your.public.ip:yournetmask+MAC to reach separete private machines via IPv6 behind your single public IPv4 address).
Yer olde FC product salesman has a much better commission, as FC products have far, far higher margins than ethernet products. Therefore the FC saleman buys you better lunches and invites you to seminars with more free booze, while displaying his company produced graphs over how cutting edge lab FC hardware vastly outperforms iSCSI served by a PC from last century.
In your booze addled state you find this reasonable, and refrain from using google or performing actual tests on comparable (priced or performance) hardware for yourself.
Ok, to be slightly more fair, iSCSI overhead may have been an issue at some point in time. Today I'd say it's negligable for most scenarios (much like hardware-vs-software raid). CPU performance and optimized network stacks simply make it a non-issue (it's not like FC does magic fairy dust things, you've got to use those multicores and your apps not multithreaded anyway, so...).
And if you have an FC salesman yakking about it, throw ATA over Ethernet in his face and suggest that you're going to buy your future storage from Coraid.
Personally I've gotten close to max theoretical throughput (95-98MB/sec) with iSCSI over gigabit on cheap-ass COTS hardware so I'm less than impressed with our very expensive corporate SAN.
I max out at 960Mb/s with iSCSI over gigabit $15 realtek cards with a $150 dlink switch. With out of the box iSCSI enterprise target software on Linux, to a client running OpeniSCSI (eh, or whatever it is that's shipped in RedHat by default). Over substandard cabling, on top of that. (Fer sure, by then the iSCSI server has cached the data in-mem, but anyway.)
So I'd really have to wonder what anyone failing to get that is running. I hope they're not paying for it.
Sure, non-cached performance against the IDE and SATA disks backing it isnt much better than 400-600Mb/s, but since I started using iSCSI for my home SAN I have come to the conclusion that the oft-repeated mantra about overhead and latency related to IP and ethernet is FC SAN salesman driven utter shite.
"Ok, let's say I have a really progressive ISP, I'm using an expensive custom router (not the $50 routers every household in America has), and the entire network chain from ISP to my house is IPv6."
A) you dont need a really progressive ISP to support v6, the 6to4 anycast route will route your v6 packets on many ISPs.
B) You dont need an expensive custom router, any PC can do the 6to4 translation interface and run radvd. And if you insist on using a dedicated router, most of those $50 routers can run OpenWRT or similar v6 capable software.
"If, like you claim, IPv6 were backwards-compatible, then I should have no problem reaching www.yahoo.com."
As a general rule you wont have a problem as you'd use one of the transition methods. If, for some obscure reason, you want to run a v6 only network (in which case I'd have to question your migration strategy), there's a host of v4 access methods ranging from proxies to transport relay translators. Like I said, the v4 address space is a subset of the v6 space in several ways, and the translation is trivial.
"1) You still need a public IPv4 address right? I thought we were running out of those?"
You need _one_ public IPv4 address. Even the largest corporations in the world would need no more than _one_ public IPv4 address. Do the math. With every single connection point to the public internet on one single address each we would have enough addresses to tide us over the migration with room to spare.
"If you have a public IPv4 address and you use NATs/proxies, you might as well stick with private IPv4"
Unless you actually want to access the machines behind that public IP address. Which is one of the nice things you can do with v6.
"All this said and done, has anyone here on/. actually upgraded a network to be IPv6 compliant"
Yep. The actual upgrade was trivial; configure radvd to point at outside 6to4 gateways and everything on the inside autoconfigured, and then set up firewalling. The firewall was the biggest difficulty as the GUI tools were lacking, but writing ipv6 non-NAT rules is trivial compared to getting v4 NAT rules right.
"and what can you tell us about real world experience."
Great. I no longer have to do multi-stage ssh jumps between NAT'ed networks. I can scp files directly to the destination machine. There are lots of small improvements that make life easier.
"IPv6 is just a bad idea until it's entirely backwards-compatible with IPv4."
Say what? IPv6 is both back and forwards and sideways compatible with IPv4, I've seen so many log messages with v4 encoded in v6 that I dont even notice these anymore: [::ffff:192.168.10.13].
"Right now, IPv6 is pointless"
I find it perfectly pointful as it allows me to ssh directly into behind-NAT machines. It also makes it much easier to use scp, ftp, or even nfs mounts between NAT nets, completely eliminating the need for complex tunneling or multi-stage jumps.
IPv6 is both practical and stable enough (and easy!) to deploy widely today in many situations. I'd suggest that a lot of the lagging in implementation is largely due to laziness or lack of technical research.
A more apt comparison would be the student getting a ballpoint pen and being told to use the fountain pen that will regularly place inkblots on the paper, forcing him to start over.
I mean, what kind of workers would we end up with in the future if we teach kids to adopt more effient means of production. I mean, sheesh, soon we'd be using tractors instead of manually plowing the fields. How could we keep employment up?
"It would be interesting to offer that as a service to people that want to know exactly what it is to which we're subjecting people in the name of counter-terrorism."
Oh, I'm sure you can find it as a service already, in the right places.
There is, however, a vast abyss between various forms of consensual torture and nonconsensual torture. To accomplish the same level of terror instilled in the victims at gitmo you'd have to basically grab someone off the street and pretend you were a real bunch of psycho killers. They cannot know that you do not intend to permanently harm them, or the torture might not be 'effective'. (Of course, it still wont be effective for extracting truth, but at least it'll get them saying whatever you want.).
"odds are they might not be so accommodating of Bush's policies afterwards."
Frankly, I find it revolting that anyone can condone such practices either way. I have nothing but contempt for the disgusting maggots who have tarnished the US image with their abhorrent vile barbarism.
'yet they're held as the epitome of a "good" company.'
Say what? By whom? Self-rating hardly counts, and most of the time outsiders mentioning "dont be evil" seems to be mostly in sarcastic references to the failure to live up to the proclaimed motto.
So, I was wondering exactly how he managed to make that business idea unprofitable. Until I ran across this part of the article; "Nonetheless they gave MySpace a mountain of money to give it a run."
Oh. I see.
Ok, here's a hint for future venturers in the music sales business (or, in fact, any business): dont hand mountains of money to any joker you run across. As soon as you hand over those mountains you're the one who's taking the loss.
"is to constantly fight about it amongst ourselves."
It's called competition. And evolution.
And yes, in the long term, it actually will do the trick.
The free software community, through dissent and conflict, becomes infinitely adaptable to any and all niches. Compare with monolithic entities like Microsoft with much stronger direction; when they decide to go down a number of dead ends they end up with products like Vista, with no fallbacks, unable to fill new niches like the low-end sub $200 pc's.
I'll take dissent before cooperation for the sake of shutting up any day.
"we can look also to the streets of London, Chicago, Paris or anywhere else there is a true (and far too often, growing) lower working class."
These too will gain from the OLPC through the (already demonstrated) commercial pressure to produce things like the Asus eee and the Walmart PC. Like I said, I'd attribute a large part of the sudden new existence of the sub $200 computer segment to the first step taken by the OLPC. Large industry players like Microsoft certainly had no intention of such a segment ever existing, considering their utter incapability to adapt their product line to it.
"Will a computer be the best aid to get the working poor out of their negative situation?"
Of course not. But the working poor will be better off with computers and educational materials within reach rather than educators _and_ materials out of reach. It would be even better with quality educators, but at least now the road to self-learning isnt quite so expensive and steep. Not a way for everyone, but a way for a few.
Also dont underestimate the value of computerbased training; educational games and other tricks can stretch even a mediocre educator over more pupils and leave more of them with useful knowledge. Dont hand them Maple, Office and Maya, the best of us want to get paid to even touch those apps. Instead hand them knowledge games, language games, math games, chemistry games, writing games, typing games.
Still, while I think this project has many such extra advantages, the greatest justification still comes down to this; it's not about charging them for a new toy, it's about replacing current costs of education material with lower costs for the OLPC instead.
"After all, this is about protecting the only market the US still has the upper hand and that generates more revenue internationally than it costs."
Mmm, no. Tricking _other_ countries into recognizing intellectual monopoly rights generates more revenue. Implementing more monopoly rights yourself merely makes your country less competetive, and strengthens the rights of _other countries_ to exact revenue from _you_.
"So what's left is content and patents."
Yeah, well, guess who's gonna own the monopoly rights of that content and those patents? Lets just say that the growing economies arent so dim they havent realized they too can get monopoly rights in the US.
Realize this: Intellectual 'property' is, and always has been, a covert distributed taxation scheme.
Saying enforcing IP 'protects jobs' is no different than saying 'raising and enforcing taxes protects jobs'. Give someone the right to exact taxes from some part of the economy and there's no limit to how large expenses they can create and how many workers they can employ. That does not equal competetive and efficient free market economy.
Considering that the OLPC isnt intended for demographics who have no food, people like Dvorak would be that reason...
There's a large and emerging segment of 'semi-poor people' who have food and most other necessities, but for whom educational material is a significant cost. One of the main points of the OLPC is to cut educational material costs while creating a load of other capabilities.
Personally I think the OLPC is already a huge success; I'd attribute the interest in it as a large part of the driving force for low-cost laptops such as ASUS Eee.
"Yeah, sometimes Dvorak's nothing more than a grumpy old man who rants."
Yep. Definitely for what it's worth. Still, it's important not to misread the google report; IIRC, while failures werent necessarily preceeded by SMART warnings, when SMART did warn there was a fair likelyhood of impending failure. Not enough to merit immediate replacement for google or someone else with massive redundancy (40% or something chance of failure within a short time period), it was definitely enough to merit migrating the disk to junk-disk for the average person.
"This "workaround" will surely make every Unix-guy rotate in his grave."
Not really. Well, the cron job is a bit ugly (a more elegant solution would be to trigger the script via udev or something, but I can understand if someone's reluctant to figure udev out).
"Go to the root of the problem and just tell the harddrive to not go into sleep and be done with it."
That isnt the root of the problem. The root of the problem is that the harddrive and driver doesnt recover gracefully after powerdown. Turning off powerdown is just a workaround hack, and an expensive one at that as it drains power, costs money for the power and causes extra wear on the drive.
Looks like there's a fairly good solution at NSLU2-Linux. Sounds like it might handle the reattachment better.
That said, while I initially liked USB attached disks, I've later found the issues with lack of SMART and other features over USB to be a showstopper for any serious use (ie, anything beyond a replacement for burning DVD's for sneakernet transmission). I'm no longer particularly surprised when the level of 'working' of such devices is found to be relative.
"And any system will eventually get compromised somehow -- competitors flooding each other's patent comments, etc etc etc .."
Not any system. The trick is to cap the system so the playing parties are forced to cooperate to maximize the equitability of the system.
The basic problem today is that the parties involved in the system are not the parties paying for the system. The players all benefit from maximizing the ease with which you get patents.
What you could do instead would be to acknowledge that the economy is already paying for the system, formalize that payment (any patent fees paid anywhere for anything), create a total budget for payouts within the patent system and then let the players self-sanitize. Make it easier to get patents? Then everyone will get less. As the system would no longer provide a monoply, but just a payout, the litigation issue would disappear, and the damage to compound innovation would go as well.
You could fiddle with the numbers to no end (get paid by level of investment? number of products using the invention?), but as long as there's an actual budget then everyone involved will be much more interested in creating a fair system.
"Because patents pretty much force people to innovate."
Approximately the same way banning the use of anything new force people to innovate. How about we outright ban the use of technology? By that theory, that should really get the innovation going.
Slashdot Mirror
"Right now IPv6 is *not* backwards-compatible with IPv4 because, unless to translate it to IPv4 (using whatever method you choose) you can't connect to IPv4 devices."
::ffff:0:0 address to the translator that merely strips the extra bits, just like you route public packets through your NAT gateway). A proxy could do it (all of which are available in one form or another).
Compare it with NAT'ing. A private 192.168 v4 address cant talk directly to yahoo.com. You need either a proxy or a NAT router that will translate your 192.168 address into a public internet address.
"If you move to IPv6, but you still need to use an IPv4 IP"
Someone, somewhere, needs to use an IPv4 IP (obviously, as your question assumes that yahoo wont respond to ipv6). You dont.
Your gateway could do it. A routable protocol translator could do it (route packets with the
So if you're saying that nothing inbetween is allowed to touch the packets, then, right, you cant use IPv6 to access yahoo.com. That would include IPv4 NAT too tho.
If, on the other hand, you mean that a machine can have IPv6 only, and that the network structure can be set up to transparently translate IPv6 packets to IPv4, then yes, that's possible.
It is, however, easier and possibly more reliable to keep the old NAT structure in place for v4-only accesses while gaining the benefits of direct host addressing that v6 gives you by running a transition dual-protocol network in the near future (thus giving you the ability to, for example, ssh directly to 2002:your.public.ip:yournetmask+MAC to reach separete private machines via IPv6 behind your single public IPv4 address).
Well, basically, this is how it works:
Yer olde FC product salesman has a much better commission, as FC products have far, far higher margins than ethernet products. Therefore the FC saleman buys you better lunches and invites you to seminars with more free booze, while displaying his company produced graphs over how cutting edge lab FC hardware vastly outperforms iSCSI served by a PC from last century.
In your booze addled state you find this reasonable, and refrain from using google or performing actual tests on comparable (priced or performance) hardware for yourself.
Ok, to be slightly more fair, iSCSI overhead may have been an issue at some point in time. Today I'd say it's negligable for most scenarios (much like hardware-vs-software raid). CPU performance and optimized network stacks simply make it a non-issue (it's not like FC does magic fairy dust things, you've got to use those multicores and your apps not multithreaded anyway, so...).
And if you have an FC salesman yakking about it, throw ATA over Ethernet in his face and suggest that you're going to buy your future storage from Coraid.
Personally I've gotten close to max theoretical throughput (95-98MB/sec) with iSCSI over gigabit on cheap-ass COTS hardware so I'm less than impressed with our very expensive corporate SAN.
I max out at 960Mb/s with iSCSI over gigabit $15 realtek cards with a $150 dlink switch. With out of the box iSCSI enterprise target software on Linux, to a client running OpeniSCSI (eh, or whatever it is that's shipped in RedHat by default). Over substandard cabling, on top of that. (Fer sure, by then the iSCSI server has cached the data in-mem, but anyway.)
So I'd really have to wonder what anyone failing to get that is running. I hope they're not paying for it.
Sure, non-cached performance against the IDE and SATA disks backing it isnt much better than 400-600Mb/s, but since I started using iSCSI for my home SAN I have come to the conclusion that the oft-repeated mantra about overhead and latency related to IP and ethernet is FC SAN salesman driven utter shite.
"8Gb FC will be out long before 10Gb ethernet becomes reasonably priced."
You mean, 8Gb FC will be out long before 100 Gb ethernet becomes reasonably priced.
10 Gb ethernet is already reasonably priced (compared to FC).
"Ok, let's say I have a really progressive ISP, I'm using an expensive custom router (not the $50 routers every household in America has), and the entire network chain from ISP to my house is IPv6."
A) you dont need a really progressive ISP to support v6, the 6to4 anycast route will route your v6 packets on many ISPs.
B) You dont need an expensive custom router, any PC can do the 6to4 translation interface and run radvd. And if you insist on using a dedicated router, most of those $50 routers can run OpenWRT or similar v6 capable software.
"If, like you claim, IPv6 were backwards-compatible, then I should have no problem reaching www.yahoo.com."
As a general rule you wont have a problem as you'd use one of the transition methods. If, for some obscure reason, you want to run a v6 only network (in which case I'd have to question your migration strategy), there's a host of v4 access methods ranging from proxies to transport relay translators. Like I said, the v4 address space is a subset of the v6 space in several ways, and the translation is trivial.
"1) You still need a public IPv4 address right? I thought we were running out of those?"
You need _one_ public IPv4 address. Even the largest corporations in the world would need no more than _one_ public IPv4 address. Do the math. With every single connection point to the public internet on one single address each we would have enough addresses to tide us over the migration with room to spare.
"If you have a public IPv4 address and you use NATs/proxies, you might as well stick with private IPv4"
Unless you actually want to access the machines behind that public IP address. Which is one of the nice things you can do with v6.
"What I don't get is why the hell did they make a protocol that is not backwards compatible?"
It is.
"We'd all be already using IPv6 if IPv4 routers could move the data around."
They can.
"Why did the committee developing the standard not *add* the IPv6 headers to the end of the IPv4 headers?"
Yes, that would be 6to4.
"All this said and done, has anyone here on /. actually upgraded a network to be IPv6 compliant"
Yep. The actual upgrade was trivial; configure radvd to point at outside 6to4 gateways and everything on the inside autoconfigured, and then set up firewalling. The firewall was the biggest difficulty as the GUI tools were lacking, but writing ipv6 non-NAT rules is trivial compared to getting v4 NAT rules right.
"and what can you tell us about real world experience."
Great. I no longer have to do multi-stage ssh jumps between NAT'ed networks. I can scp files directly to the destination machine. There are lots of small improvements that make life easier.
"IPv6 is just a bad idea until it's entirely backwards-compatible with IPv4."
Say what? IPv6 is both back and forwards and sideways compatible with IPv4, I've seen so many log messages with v4 encoded in v6 that I dont even notice these anymore: [::ffff:192.168.10.13].
"Right now, IPv6 is pointless"
I find it perfectly pointful as it allows me to ssh directly into behind-NAT machines. It also makes it much easier to use scp, ftp, or even nfs mounts between NAT nets, completely eliminating the need for complex tunneling or multi-stage jumps.
IPv6 is both practical and stable enough (and easy!) to deploy widely today in many situations. I'd suggest that a lot of the lagging in implementation is largely due to laziness or lack of technical research.
"Now I know a large scale re-ip can be painful, but they have years to do it if they start now."
At that point it would be much less painful to just migrate to ipv6, but then you dont see them doing that either.
"There really is no difference here"
A more apt comparison would be the student getting a ballpoint pen and being told to use the fountain pen that will regularly place inkblots on the paper, forcing him to start over.
I mean, what kind of workers would we end up with in the future if we teach kids to adopt more effient means of production. I mean, sheesh, soon we'd be using tractors instead of manually plowing the fields. How could we keep employment up?
"It would be interesting to offer that as a service to people that want to know exactly what it is to which we're subjecting people in the name of counter-terrorism."
Oh, I'm sure you can find it as a service already, in the right places.
There is, however, a vast abyss between various forms of consensual torture and nonconsensual torture. To accomplish the same level of terror instilled in the victims at gitmo you'd have to basically grab someone off the street and pretend you were a real bunch of psycho killers. They cannot know that you do not intend to permanently harm them, or the torture might not be 'effective'. (Of course, it still wont be effective for extracting truth, but at least it'll get them saying whatever you want.).
"odds are they might not be so accommodating of Bush's policies afterwards."
Frankly, I find it revolting that anyone can condone such practices either way. I have nothing but contempt for the disgusting maggots who have tarnished the US image with their abhorrent vile barbarism.
"That worked for Gonzales. Maybe waterboarding could make you remember, tho."
Hmm, that brings the question, did we waterboard Gonzales? If not, why not?
'yet they're held as the epitome of a "good" company.'
Say what? By whom? Self-rating hardly counts, and most of the time outsiders mentioning "dont be evil" seems to be mostly in sarcastic references to the failure to live up to the proclaimed motto.
So, I was wondering exactly how he managed to make that business idea unprofitable. Until I ran across this part of the article; "Nonetheless they gave MySpace a mountain of money to give it a run."
Oh. I see.
Ok, here's a hint for future venturers in the music sales business (or, in fact, any business): dont hand mountains of money to any joker you run across. As soon as you hand over those mountains you're the one who's taking the loss.
"is to constantly fight about it amongst ourselves."
It's called competition. And evolution.
And yes, in the long term, it actually will do the trick.
The free software community, through dissent and conflict, becomes infinitely adaptable to any and all niches. Compare with monolithic entities like Microsoft with much stronger direction; when they decide to go down a number of dead ends they end up with products like Vista, with no fallbacks, unable to fill new niches like the low-end sub $200 pc's.
I'll take dissent before cooperation for the sake of shutting up any day.
"we can look also to the streets of London, Chicago, Paris or anywhere else there is a true (and far too often, growing) lower working class."
These too will gain from the OLPC through the (already demonstrated) commercial pressure to produce things like the Asus eee and the Walmart PC. Like I said, I'd attribute a large part of the sudden new existence of the sub $200 computer segment to the first step taken by the OLPC. Large industry players like Microsoft certainly had no intention of such a segment ever existing, considering their utter incapability to adapt their product line to it.
"Will a computer be the best aid to get the working poor out of their negative situation?"
Of course not. But the working poor will be better off with computers and educational materials within reach rather than educators _and_ materials out of reach. It would be even better with quality educators, but at least now the road to self-learning isnt quite so expensive and steep. Not a way for everyone, but a way for a few.
Also dont underestimate the value of computerbased training; educational games and other tricks can stretch even a mediocre educator over more pupils and leave more of them with useful knowledge. Dont hand them Maple, Office and Maya, the best of us want to get paid to even touch those apps. Instead hand them knowledge games, language games, math games, chemistry games, writing games, typing games.
Still, while I think this project has many such extra advantages, the greatest justification still comes down to this; it's not about charging them for a new toy, it's about replacing current costs of education material with lower costs for the OLPC instead.
"After all, this is about protecting the only market the US still has the upper hand and that generates more revenue internationally than it costs."
Mmm, no. Tricking _other_ countries into recognizing intellectual monopoly rights generates more revenue. Implementing more monopoly rights yourself merely makes your country less competetive, and strengthens the rights of _other countries_ to exact revenue from _you_.
"So what's left is content and patents."
Yeah, well, guess who's gonna own the monopoly rights of that content and those patents? Lets just say that the growing economies arent so dim they havent realized they too can get monopoly rights in the US.
Realize this: Intellectual 'property' is, and always has been, a covert distributed taxation scheme.
Saying enforcing IP 'protects jobs' is no different than saying 'raising and enforcing taxes protects jobs'. Give someone the right to exact taxes from some part of the economy and there's no limit to how large expenses they can create and how many workers they can employ. That does not equal competetive and efficient free market economy.
"Of course there's no reason it can't be both."
Considering that the OLPC isnt intended for demographics who have no food, people like Dvorak would be that reason...
There's a large and emerging segment of 'semi-poor people' who have food and most other necessities, but for whom educational material is a significant cost. One of the main points of the OLPC is to cut educational material costs while creating a load of other capabilities.
Personally I think the OLPC is already a huge success; I'd attribute the interest in it as a large part of the driving force for low-cost laptops such as ASUS Eee.
"Yeah, sometimes Dvorak's nothing more than a grumpy old man who rants."
Yeah, well, no different this time.
"For what that's worth."
Yep. Definitely for what it's worth. Still, it's important not to misread the google report; IIRC, while failures werent necessarily preceeded by SMART warnings, when SMART did warn there was a fair likelyhood of impending failure. Not enough to merit immediate replacement for google or someone else with massive redundancy (40% or something chance of failure within a short time period), it was definitely enough to merit migrating the disk to junk-disk for the average person.
Oh, and of course, a few seconds of searching brought up this link: http://ubuntuforums.org/showthread.php?t=494673
Complete with udev support.
"This "workaround" will surely make every Unix-guy rotate in his grave."
Not really. Well, the cron job is a bit ugly (a more elegant solution would be to trigger the script via udev or something, but I can understand if someone's reluctant to figure udev out).
"Go to the root of the problem and just tell the harddrive to not go into sleep and be done with it."
That isnt the root of the problem. The root of the problem is that the harddrive and driver doesnt recover gracefully after powerdown. Turning off powerdown is just a workaround hack, and an expensive one at that as it drains power, costs money for the power and causes extra wear on the drive.
Looks like there's a fairly good solution at NSLU2-Linux. Sounds like it might handle the reattachment better.
That said, while I initially liked USB attached disks, I've later found the issues with lack of SMART and other features over USB to be a showstopper for any serious use (ie, anything beyond a replacement for burning DVD's for sneakernet transmission). I'm no longer particularly surprised when the level of 'working' of such devices is found to be relative.
"And any system will eventually get compromised somehow -- competitors flooding each other's patent comments, etc etc etc . ."
Not any system. The trick is to cap the system so the playing parties are forced to cooperate to maximize the equitability of the system.
The basic problem today is that the parties involved in the system are not the parties paying for the system. The players all benefit from maximizing the ease with which you get patents.
What you could do instead would be to acknowledge that the economy is already paying for the system, formalize that payment (any patent fees paid anywhere for anything), create a total budget for payouts within the patent system and then let the players self-sanitize. Make it easier to get patents? Then everyone will get less. As the system would no longer provide a monoply, but just a payout, the litigation issue would disappear, and the damage to compound innovation would go as well.
You could fiddle with the numbers to no end (get paid by level of investment? number of products using the invention?), but as long as there's an actual budget then everyone involved will be much more interested in creating a fair system.
"Because patents pretty much force people to innovate."
Approximately the same way banning the use of anything new force people to innovate. How about we outright ban the use of technology? By that theory, that should really get the innovation going.