The thing is, a stable ABI is provided by the distro, rather than the individual components. It's upto to the distro to stick to components with the same ABI, or port the old ABI to newer versions.
To illustrate, RHEL 4 is still shipping kernel 2.6.9. If a vendor targets RHEL 4 (in the same way they'd target Solaris 8), then their driver will almost certainly continue to work with all future kernels shipped for RHEL 4.
Of course, where that breaks down is the market demands that vendors like nVidia need to support RHEL 3, RHEL 4, RHEL 5, Fedora 8 through 10, not to mention various Ubuntu, Debian, Gentoo and Slackware releases. And some of those releases explicitly don't provide a stable ABI, because they're meant as places to progress the technology. But it'd be the same if you were a developer within Sun using daily snapshots of Solaris n+1 too, and that's what distros like Fedora and non-LTS Ubuntu should be fairly compared with.
The kernel would also end up with multiple different ways of doing the same thing, but no-one ever knowing if they can be removed for fear that some obscure driver hasn't been updated to use the current way.
Windows aims to provide binary compatibility, at the cost of complexity within the OS. UNIX, and Linux doubly-so, aims for source compatibility and improved architectural simplicity at the cost of some administrative complexity, aka 'Worse is Better' (http://en.wikipedia.org/wiki/New_Jersey_style).
Re:Different, minimal cards for different purposes
on
U-Turn On UK ID Cards
·
· Score: 1
Theoretically possible, but I suspect that eventually, there would be a diverse range of card readers out there and no way for a card user to easily determine whether they are a) official and b) uncompromised. See also chip and PIN.
I've believed for a while now that "winrot" and general perceived operating system instability are most often caused by hard drives in the beginning stages of failure. I think it's an underrated cause of random crashes, and boot errors such as "missing c:\windows\system32\hal.dll, etc" I wish the hardware vendors (Dell, Gateway, Apple, etc) would take more responsbility and be quicker to blame the drive (and replace it), instead of blindly having the end user run the recovery routine. Performing the recovery only papers over the underlying problem by temporarily rebuilding the file system.
Yes and no. Read errors can develop at any time and potentially take out key OS files when they do. However, this is expected behaviour for a hard disc and is a good reason to have backups. Reinstalling the OS will have the side effect of testing every block that's written, and forcing it to be re-allocated by the drive firmware if it has failed (a better strategy would be to write to every block on the disc before doing the reinstall). Once all failed blocks have been re-written (and re-allocated), the drive is pretty much as good as new, as long as there are still sufficient spare blocks left for future failures and re-allocations.
The short story is that read errors develop, even on healthy drives, but a drive that develops write errors should be returned under warranty or binned.
Re:Different, minimal cards for different purposes
on
U-Turn On UK ID Cards
·
· Score: 1
Just because some people are foolish is no reason to demand that everyone acts like a fool.
I made no such demand, or even suggested such.
I was merely making the point that I've met Real People who don't worry as much about privacy as you or I, and are looking forward to trading what privacy they have right now for a little convenience, and that any proposed system should be able to included their (ill-considered) desires as well.
Re:Different, minimal cards for different purposes
on
U-Turn On UK ID Cards
·
· Score: 1
If I was to be forced to comply with an identity card/database scheme, your proposal is roughly how I'd want it to work; an arbitrary number of distinct identities which I, the citizen, can separate or combine as I see fit. Like you, I'd be happy with the management overhead that comes with having several distinct identities (it's no worse than at the moment, after all).
On the other hand, there are people who really are looking forward to just being able to carry around one piece of plastic that serves as an all-purpose ID card. I think they're crazy, and I expect you do too, but I'm not sure it's my place to deny them the right to jeopardise their identity in that way. The only argument I can see for doing so would be to avoid compromise and loss of confidence in larger systems.
"What business is Harley Davidson in? Not Motorcycles.... What business is McDonald's in? Business process."
But if they dropped their respective core businesses, where would they end up?
It can be surprising, but some companies have, over the course of their existence evolved into something far different from their original core business. A good example of this is ITT. Sometimes a look at the numbers reveals that a corporation's true core business is completely different from what it's known for. I wonder how Harley Davidson's margins on their bikes compare with those from their merchandise and/or licensing deals, how the margins on McDonald's food sales compare with the profits from the sale of its unwanted real estate and so on.
"My point is that if Apple were a computer company, they'd be making the crappy margins that Dell and the other PC makers are making."
Like IBM? Define "computer company". Hell, define "computer". This can get interesting.
IBM isn't really a computer company either! For a while, that was their predominant business, but if you look over their history, their main business has been consulting and professional services. This shows in their retreat from the manufacture of PCs (now Lenovo) and hard discs (now Hitachi).
"The reasonable man adapts himself to the conditions that surround him... The unreasonable man adapts surrounding conditions to himself... All progress depends on the unreasonable man."
Yet another good reason to set up a private local mirror of your chosen distro(s') update repositories and configure all your hosts to use it as their only source of updates.
Then, even if your mirror pulls from a mirror operated by spamgang[TM], they don't get any visibility of the vulnerable software within your organisation, other than (maybe) the machine performing the mirroring.
1. Metallica killed Napster - Did you really expect to be able to freely download any song you like, as easily as you like, forever and ever? Metallica may have been the first major band to bring a lawsuit against Napster, but I woudln't of given Napster another three months before someone else stepped in. And what did we get out of Napster crashing? P2P programs that let you share WAY more than music. You could get games, movies, porno, AND music on the software that popped up after Napster got shut down. I'm probably the only one that thinks this, but I appreciate Metallica getting Napster shut down. Now I can get the same music AND files from torrents.
Napster getting sued was pretty much inevitable, but I find it grossly hypocritical for one of the first lawsuits to come from a band whose background was in tape-trading, and which built much of their popularity through the tape-trading network.
I've used the Heise offline update tool, and I'd certainly prefer its approach (i.e. downloading the hotfixes directly from microsoft.com) to putting my trust in a random person compiling a.iso image for me. Yes, I need to place a little bit of trust that Heise's tool isn't sabotaging the stuff it downloads, but they've earnt that trust from me over the last decade.
Whether American, Canadian or Islamic, they pointed out that a disproportionate share of engineers seem to have a mindset that makes them open to the quintessential right-wing features of "monism" (why argue where there is one best solution) and by "simplism" (if only people were rational, remedies would be simple).
Monism and simplism sound to me like the MBTI characteristics of J and NT types respectively. And, big surprise, lots of scientists/engineers are, or identify as xNTJs.
If this is a problem, write a wrapper script that uses ntpdate to immediately set the clock from the same NTP server as the servers you're trying to connect to. Not great for multi-user machines, but should be fine for desktops and other single-user machines with a modicum of care (e.g. making sure 'make' isn't running whilst you do so...)
I'm not sure what you mean by "outside" our physical Universe - the Universe is everything that physically exists (as opposed to things like "ideas"), by definition.
So, to ask physics/cosmology-oriented question; if parallel universes as postulated by Everett et al actually exist, are they part of our universe, or outside our universe? How can our understanding of physics investigate the physics of another universe in which we can make no observations, and whose physical constants may well differ from that of our own?
Now, if you accept that parallel universes may possibly exist, and our science may not be able to investigate them, is it really that much of a stretch to accept that some form of powerful entity may possibly exist outside our physical universe (which is the logical conclusion of following the common attributes of 'god' as defined by most world religions and believers), albeit with no guarantee of it being a personal or especially "loving" God as defined by the Abrahamic religions.
Show me that there's a purpose of things, then we can ask "Why?" If we discover that there must be a purpose, then science will also conceivably be able to find what that purpose is.
[...]
They are all nonsensical questions that may not have answers, because the things in the question don't exist in the first place.
Maybe there is a purpose to things, maybe there isn't. Someone with a firmer spiritual belief than I will be utterly convinced there's a purpose, but it won't be able to be explained by science, as science is only a useful tool for examining that which exists in our physical universe, and their purpose will almost certainly exist outside of it.
One contains all the things explained by the phrase "god did it". The other contains all the things explained by "science".
A long time ago, everything was in the god basket, and nothing at all was in the science basket. The weather? God did it. Pregnancy? God did it. Disease? God did it. Where does stuff come from? God did it.
Actually, you've got the same thought process as many of your opponents, and your baskets are straw men. Really, they should be labelled "Can only be explained through supernatural phenomena" and "Can be explained solely using understood and observable physical phenomena". The difference is subtle, but important. Science doesn't explain why things happen (God, the Flying Spaghetti Monster, sheer blind luck), only how. It's possible that gods exist, and act through the physical phenomena we understand today (i.e. evolution, physics and all the rest).
It may not have been an 'abnormally large current' just one that was too much for the build specifications of the PSU in use. Not all PSUs are created equal.
A few years later I went to turn on my computer as usual and it wouldn't turn on. A bit of troubleshooting later and I realized that the PSU connector had burned itself into the motherboard power socket because something on the motherboard had randomly decided to short itself. Four of the pins had fried (in a distinctive pattern, see here and here)
What make/model of PSU were you using? Looks to me that the PSU's power connector couldn't cope with the current the board was pulling. Of course, that may or may not be down to a fault on the board, but seeing as you haven't told us anything about the PSU, I'm betting you were using a cheapo one that came with the case you were using.
Of course biofuel is a conversion; it's utilizing photosynthesis to capture solar energy (i.e. light) and turn it into carbohydrates (i.e. chemical potential).
Now, solar energy comes 'for free', but fertilizers, pesticides, harvesting and processing don't. As it stands, we can't feed the 6 billion people on this planet without chemical fertilizers. Furthermore, production of biofuels competes with food production. George Monbiot was one of the first people to writeabout this, and we can see effects in the rising price of food staples such as tortillas and pasta.
Slashdot Mirror
The thing is, a stable ABI is provided by the distro, rather than the individual components. It's upto to the distro to stick to components with the same ABI, or port the old ABI to newer versions.
To illustrate, RHEL 4 is still shipping kernel 2.6.9. If a vendor targets RHEL 4 (in the same way they'd target Solaris 8), then their driver will almost certainly continue to work with all future kernels shipped for RHEL 4.
Of course, where that breaks down is the market demands that vendors like nVidia need to support RHEL 3, RHEL 4, RHEL 5, Fedora 8 through 10, not to mention various Ubuntu, Debian, Gentoo and Slackware releases. And some of those releases explicitly don't provide a stable ABI, because they're meant as places to progress the technology. But it'd be the same if you were a developer within Sun using daily snapshots of Solaris n+1 too, and that's what distros like Fedora and non-LTS Ubuntu should be fairly compared with.
The kernel would also end up with multiple different ways of doing the same thing, but no-one ever knowing if they can be removed for fear that some obscure driver hasn't been updated to use the current way.
Windows aims to provide binary compatibility, at the cost of complexity within the OS. UNIX, and Linux doubly-so, aims for source compatibility and improved architectural simplicity at the cost of some administrative complexity, aka 'Worse is Better' (http://en.wikipedia.org/wiki/New_Jersey_style).
Theoretically possible, but I suspect that eventually, there would be a diverse range of card readers out there and no way for a card user to easily determine whether they are a) official and b) uncompromised. See also chip and PIN.
Yes and no. Read errors can develop at any time and potentially take out key OS files when they do. However, this is expected behaviour for a hard disc and is a good reason to have backups. Reinstalling the OS will have the side effect of testing every block that's written, and forcing it to be re-allocated by the drive firmware if it has failed (a better strategy would be to write to every block on the disc before doing the reinstall). Once all failed blocks have been re-written (and re-allocated), the drive is pretty much as good as new, as long as there are still sufficient spare blocks left for future failures and re-allocations.
The short story is that read errors develop, even on healthy drives, but a drive that develops write errors should be returned under warranty or binned.
I made no such demand, or even suggested such.
I was merely making the point that I've met Real People who don't worry as much about privacy as you or I, and are looking forward to trading what privacy they have right now for a little convenience, and that any proposed system should be able to included their (ill-considered) desires as well.
On the other hand, there are people who really are looking forward to just being able to carry around one piece of plastic that serves as an all-purpose ID card. I think they're crazy, and I expect you do too, but I'm not sure it's my place to deny them the right to jeopardise their identity in that way. The only argument I can see for doing so would be to avoid compromise and loss of confidence in larger systems.
And before that there was the Eiger Labs MPMan F10 and the Diamond Rio PMP300.
MythTV also works fine in the non-US parts of the world where DVB-T is pretty much standard for digital terrestrial broadcasts.
But if they dropped their respective core businesses, where would they end up?
It can be surprising, but some companies have, over the course of their existence evolved into something far different from their original core business. A good example of this is ITT. Sometimes a look at the numbers reveals that a corporation's true core business is completely different from what it's known for. I wonder how Harley Davidson's margins on their bikes compare with those from their merchandise and/or licensing deals, how the margins on McDonald's food sales compare with the profits from the sale of its unwanted real estate and so on.
"My point is that if Apple were a computer company, they'd be making the crappy margins that Dell and the other PC makers are making."
Like IBM? Define "computer company". Hell, define "computer". This can get interesting.
IBM isn't really a computer company either! For a while, that was their predominant business, but if you look over their history, their main business has been consulting and professional services. This shows in their retreat from the manufacture of PCs (now Lenovo) and hard discs (now Hitachi).
The original quote was by George Bernard Shaw:
"The reasonable man adapts himself to the conditions that surround him... The unreasonable man adapts surrounding conditions to himself... All progress depends on the unreasonable man."
Yet another good reason to set up a private local mirror of your chosen distro(s') update repositories and configure all your hosts to use it as their only source of updates.
Then, even if your mirror pulls from a mirror operated by spamgang[TM], they don't get any visibility of the vulnerable software within your organisation, other than (maybe) the machine performing the mirroring.
Napster getting sued was pretty much inevitable, but I find it grossly hypocritical for one of the first lawsuits to come from a band whose background was in tape-trading, and which built much of their popularity through the tape-trading network.
Lousy secondary source, I know, but Wikipedia says the titanium for the SR-71 was bought from the USSR
I've used the Heise offline update tool, and I'd certainly prefer its approach (i.e. downloading the hotfixes directly from microsoft.com) to putting my trust in a random person compiling a .iso image for me. Yes, I need to place a little bit of trust that Heise's tool isn't sabotaging the stuff it downloads, but they've earnt that trust from me over the last decade.
Monism and simplism sound to me like the MBTI characteristics of J and NT types respectively. And, big surprise, lots of scientists/engineers are, or identify as xNTJs.
As if "playing out" isn't already an annoying problem, it looks like we'll soon have to deal with "projecting out" too.
If this is a problem, write a wrapper script that uses ntpdate to immediately set the clock from the same NTP server as the servers you're trying to connect to. Not great for multi-user machines, but should be fine for desktops and other single-user machines with a modicum of care (e.g. making sure 'make' isn't running whilst you do so...)
So, to ask physics/cosmology-oriented question; if parallel universes as postulated by Everett et al actually exist, are they part of our universe, or outside our universe? How can our understanding of physics investigate the physics of another universe in which we can make no observations, and whose physical constants may well differ from that of our own?
Now, if you accept that parallel universes may possibly exist, and our science may not be able to investigate them, is it really that much of a stretch to accept that some form of powerful entity may possibly exist outside our physical universe (which is the logical conclusion of following the common attributes of 'god' as defined by most world religions and believers), albeit with no guarantee of it being a personal or especially "loving" God as defined by the Abrahamic religions.
[...]
They are all nonsensical questions that may not have answers, because the things in the question don't exist in the first place.
Maybe there is a purpose to things, maybe there isn't. Someone with a firmer spiritual belief than I will be utterly convinced there's a purpose, but it won't be able to be explained by science, as science is only a useful tool for examining that which exists in our physical universe, and their purpose will almost certainly exist outside of it.
One contains all the things explained by the phrase "god did it". The other contains all the things explained by "science".
A long time ago, everything was in the god basket, and nothing at all was in the science basket. The weather? God did it. Pregnancy? God did it. Disease? God did it. Where does stuff come from? God did it.
Actually, you've got the same thought process as many of your opponents, and your baskets are straw men. Really, they should be labelled "Can only be explained through supernatural phenomena" and "Can be explained solely using understood and observable physical phenomena". The difference is subtle, but important. Science doesn't explain why things happen (God, the Flying Spaghetti Monster, sheer blind luck), only how. It's possible that gods exist, and act through the physical phenomena we understand today (i.e. evolution, physics and all the rest).
It may not have been an 'abnormally large current' just one that was too much for the build specifications of the PSU in use. Not all PSUs are created equal.
What make/model of PSU were you using? Looks to me that the PSU's power connector couldn't cope with the current the board was pulling. Of course, that may or may not be down to a fault on the board, but seeing as you haven't told us anything about the PSU, I'm betting you were using a cheapo one that came with the case you were using.
Now, solar energy comes 'for free', but fertilizers, pesticides, harvesting and processing don't. As it stands, we can't feed the 6 billion people on this planet without chemical fertilizers. Furthermore, production of biofuels competes with food production. George Monbiot was one of the first people to write about this, and we can see effects in the rising price of food staples such as tortillas and pasta.
Sure, but you'll lose energy both in the conversion from waste heat/electricity/whatever to oil/hydrolysis, and the conversion back again.