Well, Apple contribute to and make use of the TrustedBSD project; e.g. OpenBSM is derived from code Apple released, and the MAC framework is found on both FreeBSD and OS X.
I'd hope we're going to see BIOS support for GPT partitions before too long. One more doubling and we're right flat against the limit on MBR partition sizes, so.. what, a year to go?
Lossless has practical advantages which go beyond "it sounds closer to the original". 128kbps is ok for outside use? Right; but I don't want to make that from a 256kbps file, I'd rather make it from a lossless source, and in a few years maybe I'd like to switch to 64kbps HE-AAC, and encode that from my original source too. In 5 years maybe I'll want to use Vorbis 2 or WavPack Hybrid.. and I can do that too. And I can keep doing this until I die or I lose my data outright.
That 256kbps AAC is never going to be of much use other than as a 256kbps AAC.
No doubt, downloads will cost 1GBP, rather than 1USD, when they finally make it to the UK. I tried paying for an album at Magnatune in GBP instead of USD recently, in a vague effort to give them more money (maximum charges are $18 or £10; £10 ~= $20.30 at current exchange rates). Imagine my dispair when they cheerfully said I'd been charged a shade over $17. Makes a change from $10 == £10 most retailers seem to like.
Just because you can achieve it in a lab doesn't mean you can achieve it in a cheap, reliable, mass-produced device right this second. 1Tb/in^2 with a few million dollars of lab equipment != 1Tb/in^2 in a $40 consumer drive. I expect they've had variations on this sort of head for a few years, at $10 million a pop, and are now announcing that they're actually ready to start mass producting a version you might be able to actually buy.
I incrementally dump(8) my important few hundred GB to a rotated pair of 500G drives. I could do with an extra hot-swap caddy and something to store a drive + caddy in when it's not online, but it seems to work pretty well.
My home server lives in one of these, but standalone and 5.25" bay hot-swap SATA racks are pretty common.
One day, in the dim and distant future, I hope to see FastCGI supported by mod_proxy[_balancer]. Sadly, the module in Apache trunk seems to have pretty much died. *sulk*.
Actually, scratch that. What I want to see in the dim and distant future is a PHP HTTP SAPI module, so it can run its own webserver and I can proxy or not as needed.
I'm not denying that Steam doesn't have some good aspects, but the app itself is a pile of junk. It's slow, buggy ("Delete your ClientRegistry.blob. Again."), awkward to use, and lacks a lot of basic functionality I'd hope a download service like it would have by now.
Stardock's content delivery system, Stardock Central, is much closer to what I want out of such a service; if Stardock are down or my Internet connection is funky, it doesn't freeze for 120 seconds waiting for the connection to time out; if a patch is available, it can tell me about the exact version I'm about to get, let me checkpoint my current install, and if the patch goes horribly wrong I can rollback; it doesn't lock up and busyloop for no apparant reason; it looks and acts like an actual native application.
I'm pretty sure all the major browsers do some guessing these days, since there are a lot of misconfigured servers out there; CSS, JS, images, even HTML end up being served as text/plain or application/octet-stream, and people expect them to work.
In Opera it can be configured from opera:config under User Prefs -> Trust Server Types. I can't find an equivilent in Firefox.
Perhaps you should look up the meaning of "disbelief" and "belief" in that same dictionary. Maybe you should look up the word "or" too, it helps when you're using big words not to skip learning what the smaller ones mean.
The best you can do is believe he doesn't exist (or think you know he doesn't exist), and if that's the case then you now have a religious belief, so atheism becomes a contradiction in terms
Er, what definition of atheist are you working from? Theism means "the belief in the existence of one or more divinities or deities"; atheism means being without that. How does not believing in [gG]od(s) contradict the term?
Considering it a religious belief is amusing, but ultimately meaningless. You could just as well believe in Odin and Thor, fairies and elves, or Time Cube and claim anyone who didn't believe in them were just as religious as you, but it wouldn't say a thing about the relative merits of either side, nor would it do the word "religious" any favours.
Well, each stream seems to have a maximum rate it can spew out dots; if you exceed that, they back up. If you can spew out 1,000 dots from each stream in a minute but you've got 10,000 to actually spew through it, it's going to take 10 minutes doing it.
I just ran it through 10,000 Apache requests. After a minute and a half or so it stopped spewing dots from most of the graphs other than the "Content" ones, which spewed for about 8 minutes. In all those logs (about 60 seconds of activity) took 6 minutes 22 seconds CPU time on a 1.66GHz Core Duo Mac Mini.
Most of that time seems to have been spent drawing dots at maximum speed spewing out of the "Content" lines; maybe they need to increase speed in response to higher request rates so it's not waiting for them to clear the screen so much?
What's the point of selling FLAC? You can just get the raw 44.1 Khz samples from the CD itself. Erm, because I can have a FLACed album in 12 minutes flat, while the CD will take anything from 2 days to 3 weeks to arrive, and will then demand effort ripping and finding space to store the hilariously oversized and redundant backup media?
And, FLAC doesn't play in mp3 players like iPod etc. So fucking what? FLAC turns into ANY OTHER FORMAT. I can turn it into Apple Lossless, or MP3 or AAC or Vorbis or WMA or ATRAC or anything else I have an encoder for, and I can do it as often as I want to as many different formats and bitrates as I want. I can pick different formats for my 4G iPod, my 60G laptop, what I might stream over the Internet or my crappy WiFi, and anything else I might want in future, ever.
Unless you're selling music incredibly cheaply, FLAC or WavPack or similar are the only formats I'm interested in buying music as.
Even sci-fi authors like Kim Stanley Robinson have included disaster scenarios when contemplating this technology, but irl nobody ever discusses the massive dangers. Er, you don't think the possible dangers haven't been discussed, modeled, etc? Here, have a paper I found in about 30 seconds. See page 10:
Below about 43 km, vterm is below 10 m/s. This confirms what was stated in Edwards (2000b, Sec. 10.9): the ribbon will reach the ground at a very low velocity, and there will be no impact damage due to the ribbon falling.... [pg 11] For a 20 T elevator, with a breaking point of 130 GPa, the maximum tension that can be achieved anywhere along the ribbon before it breaks is 1 MN. The Fate of a Broken Space Elevator corresponding force is 1.1 kN (about 110 kg) at 1 m of height, ten times more at 100 m of height. A building (100 m) should be undisturbed by this force, and a person (1 m) may be trapped, but should not be hurt.... [pg 13] After an intense deceleration phase the ribbon falls slowly to the ground at less than 1 m/s In future, either learn to troll better, or learn to use Google. And perhaps update your mental model of "teh scientists!1!111!" to take into account the likelihood that they're rather less dense than you
Anyone could have told you S754 was a dead-end upgrade path in 2003 when it was launched; it's a low end/mobile socket with no dual channel memory, and as I recall it was known or at least reliably postulated even then that dual core was going to be for the higher end sockets only. I paid the extra for S939 and my system's still pretty decent. It's not nVidia's fault you're still on AGP (low end PCIe cards aren't expensive, and for a server you can probably get away with a freebie PCI card) or that everything's moved to DDR2 (blame AMD and the market in general phasing out DDR).
The dearth of PCI-E IO cards is a bit annoying, though, as is the almost complete lack of higher end PCI sockets on even quite expensive boards; hardly anyone's even supporting 66MHz PCI. I expect we'll see better mid to low end PCIe support as time goes on though.
Erm, no, you don't need to be root to exploit these attacks, you just need to be in a situation where some security is provided by syscall wrappers. Such wrappers are sometimes used to provide an additional layer of security for applications and services; providing additional limits to what files can be read and modified, what executables can be ran, where sockets can be bound and so forth, and importantly, logging when attempts to breach these policies are attempted. You're exactly one buffer overflow away from such an exploit being a problem, since it makes that layer useless against a clueful attacker.
That so many people are finding this attack surprising is probably a sign that there isn't enough appreciation for the security implications of concurrent systems (even if they've been known by security conscious people for many years), and that we might be finding more attacks based on them in future as more people become aware of them.
"can't see why someone with the need for Sun's heavy iron to plop Windows on there just to run Exchange, for example"
Um, Sun do make some pretty cheapo systems too. It might not make a huge amount of sense to put Windows on a 32 core X4600M2 but I expect they sell 40x as many dual and single socket systems it'll feel perfectly at home on.
I found SS2 a lot creepier. I made my way through BioShock without a single big jumpy moment or any sense of wanting to get through a section quickly because it was fucking scary being there. SS2 had oodles of those. Maybe I'm just older and harder to scare.
That's not to say it's not a good game, it's just.. shallower than I would have liked.
From the Wikipedia article:
Observing a sufficient number of iterates (624 in the case of MT19937) allows one to predict all future iterates A proper cryptographically secure PRNG like Yarrow shouldn't be guessable like this, and neither should this RNG, though they may not be "more effecient and faster", just "better".
Slashdot Mirror
Something that doesn't work on most computers.
OK, so I guess that means they're competitive in at least some areas.
Well, Apple contribute to and make use of the TrustedBSD project; e.g. OpenBSM is derived from code Apple released, and the MAC framework is found on both FreeBSD and OS X.
I'd hope we're going to see BIOS support for GPT partitions before too long. One more doubling and we're right flat against the limit on MBR partition sizes, so.. what, a year to go?
Lossless has practical advantages which go beyond "it sounds closer to the original". 128kbps is ok for outside use? Right; but I don't want to make that from a 256kbps file, I'd rather make it from a lossless source, and in a few years maybe I'd like to switch to 64kbps HE-AAC, and encode that from my original source too. In 5 years maybe I'll want to use Vorbis 2 or WavPack Hybrid.. and I can do that too. And I can keep doing this until I die or I lose my data outright.
That 256kbps AAC is never going to be of much use other than as a 256kbps AAC.
Just because you can achieve it in a lab doesn't mean you can achieve it in a cheap, reliable, mass-produced device right this second. 1Tb/in^2 with a few million dollars of lab equipment != 1Tb/in^2 in a $40 consumer drive. I expect they've had variations on this sort of head for a few years, at $10 million a pop, and are now announcing that they're actually ready to start mass producting a version you might be able to actually buy.
I incrementally dump(8) my important few hundred GB to a rotated pair of 500G drives. I could do with an extra hot-swap caddy and something to store a drive + caddy in when it's not online, but it seems to work pretty well.
My home server lives in one of these, but standalone and 5.25" bay hot-swap SATA racks are pretty common.
One day, in the dim and distant future, I hope to see FastCGI supported by mod_proxy[_balancer]. Sadly, the module in Apache trunk seems to have pretty much died. *sulk*.
Actually, scratch that. What I want to see in the dim and distant future is a PHP HTTP SAPI module, so it can run its own webserver and I can proxy or not as needed.
(woohoo, Portal and Ep2 unlocking. Suck it, people who have to go to work today ;)
I'm not denying that Steam doesn't have some good aspects, but the app itself is a pile of junk. It's slow, buggy ("Delete your ClientRegistry.blob. Again."), awkward to use, and lacks a lot of basic functionality I'd hope a download service like it would have by now.
Stardock's content delivery system, Stardock Central, is much closer to what I want out of such a service; if Stardock are down or my Internet connection is funky, it doesn't freeze for 120 seconds waiting for the connection to time out; if a patch is available, it can tell me about the exact version I'm about to get, let me checkpoint my current install, and if the patch goes horribly wrong I can rollback; it doesn't lock up and busyloop for no apparant reason; it looks and acts like an actual native application.
Oh, and it doesn't have any DRM.
Grr, that link should be opera:config#Trust%20Server%20Types -- Slashdot ate my #
I'm pretty sure all the major browsers do some guessing these days, since there are a lot of misconfigured servers out there; CSS, JS, images, even HTML end up being served as text/plain or application/octet-stream, and people expect them to work.
In Opera it can be configured from opera:config under User Prefs -> Trust Server Types. I can't find an equivilent in Firefox.
Perhaps you should look up the meaning of "disbelief" and "belief" in that same dictionary. Maybe you should look up the word "or" too, it helps when you're using big words not to skip learning what the smaller ones mean.
Er. I'm sure that quote block was closed in preview.. gah. Stupid bit flipping Interweb spirits!!!1
Considering it a religious belief is amusing, but ultimately meaningless. You could just as well believe in Odin and Thor, fairies and elves, or Time Cube and claim anyone who didn't believe in them were just as religious as you, but it wouldn't say a thing about the relative merits of either side, nor would it do the word "religious" any favours.
Well, each stream seems to have a maximum rate it can spew out dots; if you exceed that, they back up. If you can spew out 1,000 dots from each stream in a minute but you've got 10,000 to actually spew through it, it's going to take 10 minutes doing it.
I just ran it through 10,000 Apache requests. After a minute and a half or so it stopped spewing dots from most of the graphs other than the "Content" ones, which spewed for about 8 minutes. In all those logs (about 60 seconds of activity) took 6 minutes 22 seconds CPU time on a 1.66GHz Core Duo Mac Mini.
Most of that time seems to have been spent drawing dots at maximum speed spewing out of the "Content" lines; maybe they need to increase speed in response to higher request rates so it's not waiting for them to clear the screen so much?
Unless you're selling music incredibly cheaply, FLAC or WavPack or similar are the only formats I'm interested in buying music as.
Anyone could have told you S754 was a dead-end upgrade path in 2003 when it was launched; it's a low end/mobile socket with no dual channel memory, and as I recall it was known or at least reliably postulated even then that dual core was going to be for the higher end sockets only. I paid the extra for S939 and my system's still pretty decent. It's not nVidia's fault you're still on AGP (low end PCIe cards aren't expensive, and for a server you can probably get away with a freebie PCI card) or that everything's moved to DDR2 (blame AMD and the market in general phasing out DDR).
The dearth of PCI-E IO cards is a bit annoying, though, as is the almost complete lack of higher end PCI sockets on even quite expensive boards; hardly anyone's even supporting 66MHz PCI. I expect we'll see better mid to low end PCIe support as time goes on though.
Erm, no, you don't need to be root to exploit these attacks, you just need to be in a situation where some security is provided by syscall wrappers. Such wrappers are sometimes used to provide an additional layer of security for applications and services; providing additional limits to what files can be read and modified, what executables can be ran, where sockets can be bound and so forth, and importantly, logging when attempts to breach these policies are attempted. You're exactly one buffer overflow away from such an exploit being a problem, since it makes that layer useless against a clueful attacker.
That so many people are finding this attack surprising is probably a sign that there isn't enough appreciation for the security implications of concurrent systems (even if they've been known by security conscious people for many years), and that we might be finding more attacks based on them in future as more people become aware of them.
"can't see why someone with the need for Sun's heavy iron to plop Windows on there just to run Exchange, for example"
Um, Sun do make some pretty cheapo systems too. It might not make a huge amount of sense to put Windows on a 32 core X4600M2 but I expect they sell 40x as many dual and single socket systems it'll feel perfectly at home on.
I found SS2 a lot creepier. I made my way through BioShock without a single big jumpy moment or any sense of wanting to get through a section quickly because it was fucking scary being there. SS2 had oodles of those. Maybe I'm just older and harder to scare.
That's not to say it's not a good game, it's just.. shallower than I would have liked.
You'll probably find you lose some speed and accuracy with just one eye.