Mod parent up. The domain system was meant to be strictly hierarchical, with most entities registering domains at the third or even the fourth level.
The perversion of system into a quasi-flat namespace centered around second-level domains came about because the planned high-level directory service for the Internet never materialized. These days most people use search engines to do much the same thing, albeit at the cost of standardization and protocol / vendor independence.
Judging from Apple's page, all the gestures are the same as the ones used by the current MacBook Pro trackpad. Apple didn't even bother to change the preferences panel (screenshot). Apple's claim to fame is keeping things simple as possible for consumers, and adding gestures that aren't supported by any other Apple device would tend to run against that.
Seeing as Apple has never bothered to provide official Windows driver support for their peripherals, I wouldn't hold my breath on that, either. Somebody will probably rip the appropriate bits out of Boot Camp if and when they update it to support the pad, however (this has been done before for other Apple devices).
The auto refresh is there because the article is a "live blog" which received real-time updates as the press conference progressed. Now that the conference is over, Engadget appears to have forgotten to turn it off, but removing the &refresh=60 from the end of the URL string and reloading the page should fix it.
[...] the secret ballot is such a terrible analogy for this petition that it's incredible anyone would bring it up if they have any clue whatsoever as to how secret ballots actually work, and how much effort is made to verify that people in secret ballot situations don't vote twice, and their identity matches who they say they are, and they are actually legally allowed to vote.
For all we know the names on this petition are "Donald Duck" repeated 100,000 times, or the names of closet gays (also known as Bible Believing Christians) and their minor children.
Publishing the names serves the good and useful purpose of validating that the signers are who they say they are, and that they are adults living in the State of Washington, as opposed to shills from out of state, minors, or fictional characters.
There is absolutely nothing to prevent the same checks from being applied to the signature gathering process. If all signatories provide their address along with their name, it is trivial to verify that they are indeed eligible citizens of the jurisdiction in which the petition is being circulated.
The agency responsible for verifying the signatures could even opt to contact a random sample of the signatories and confirm their participation, if they so chose. I would point out, however, that doing so goes beyond the measures taken to ensure the integrity of actual ballots. When I request an early ballot and vote by mail, the state has no way to confirm that the person who called for the ballot, filled it out and returned it was actually me.
The only practical reason for making petition signatures public is to allow them to be verified by any interested third party. The problem with this is that, for some people, "verification" involves breaking kneecaps. Why should we hold petitions to a higher standard than the elections that follow them?
That's just your speculation and not supported by the facts. Why are they dumping so much money into free HTML5 dev tools and support for HTML5 apps in Webkit if their goal is to prevent cross platform apps?
Apple's move to support HTML5 is commendable, but don't doubt for an instant that they also have an interest in breaking the back of Flash. Adobe has been less than stellar in supporting the Mac lately and Apple wants to be sure they aren't beholden to them in the future.
Also, why would taking steps that make cross platform development in and of itself help Apple instead of hurt them. Generally, breaking interoperability only helps when you have dominant market share, otherwise it hurts the bottom line. So basically, your hypothesis has no support.
Apple's mindshare with casual users is tremendous. The big fight right now is not among Apple, Android, RIM, Nokia et al, but rather between Apple and Android. The latter two have traditionally targeted the business and prosumer markets and have mature offerings there. Android and the iPhone OS are both new, emerging platforms targeting home users, and in that market Apple's position is enviable. They may recently have slipped behind Android on total sales, but they still have the next gen iPhone coming down the pipe and all reports indicate that it will close the hardware gap that emerged with the release of the Nexus One.
Your argument that Apple's exclusion of Flash is based exclusively on concern about application quality is a little naive, I think. If applications developed with Adobe's tools are lower in quality than the native offerings, consumers will recognize that and opt not to use them. Saying that they will instead turn around and blame Apple for their shortcomings is a little twisted. Additional choices are a win for consumers here, the only one who stands to lose is Apple. Denying that their actions here are not motivated at least in part by unadulterated self-interest is disingenuous.
Another reason Apple is so dead-set against using Adobe Flash on their iPhone/iPads is because they would lose their exclusive development platform of XCode on their custom Mac hardware. And if Apple is anything, they are a hardware company.
I'm trying hard to understand what you're trying to say here. Apple makes piles of money selling iPhones. They make next to nothing selling Macs to iPhone developers.
Apple wants to control the development platform because it locks developers into supporting only Apple's end user hardware, namely, the iPhone. Apple's tools do not effectively support cross platform development, so anyone who wants to make an app that runs on more than just the iPhone has to write substantial amounts of additional code to support the non-Apple devices. The result is many developers opting not to target devices other than the iPhone at all, which is good for Apple because users are now pressured to buy their hardware for the exclusive applications.
If Apple allows Adobe's development tools to be used, the distinction between Apple and non-Apple hardware is abstracted away, and there is no longer any reason for developers not to make their applications available on as many devices as possible. This invites competition that Apple naturally does not want.
They have to monetize their service somehow. This seems perfectly logical and reasonable to me.
Certainly no disagreement here. I just find it a bit irksome that Twitter is glossing over this move with platitudes about preserving the relevance of their service and fostering "innovation" when it's obvious to anyone that they're just acting to protect their business model.
Of course Twitter isn't going to allow advertisers to use their API for free. They will muzzle any attempts to do so in the name of preventing spam, then turn around and charge them for the privilege.
The original author's argument consists entirely of pillorying OpenBSD for its lack of any Extended ACL framework as a second line defense against security breaches. Posters in the comments section rightly point out that OpenBSD does indeed include other second line defenses like PID randomization, ASLR, and extensive support for chroots - some of which are still not supported by default in Linux distributions today. The OpenBSD maintainers' choice to focus on ensuring the quality of the first line application and kernel codebases merely represents a different approach, no doubt motivated in part by their small number of developers. An EACL framework is essentially a bolt-on solution to help contain applications that are poorly written and difficult to isolate - it is potentially easier to address the issues in the targeted applications than to devote additional effort to the security frameworks, especially when said frameworks represent an increase in the system attack surface unto themselves.
The author completely fails to address these counterarguments, dismissing approaches like chroots as overly simplistic and holding fast to the unsupported position that properly audited security policies are somehow inherently superior to properly audited kernel and application code. There may very well be a place for Extended ACL frameworks in OpenBSD, but the linked article is certainly not enough to convince anyone.
So, you support school choice, except for those people to whom you feel superior? Get bent.
Not at all. Of course the same choices must be available to everyone; anything else is oppression, plain and simple.
I was just pointing out that choice is not a perfect solution, though I do believe it is superior to all the other ones that have been tried so far. People will find a way to do damaging things no matter what the system, and I do not think it is helpful to pretend otherwise merely for the sake of argument.
And when you leave it up to parents to decide what to teach their kids, sometimes they will be wrong, too.
I support school choice in principle, but I also recognize that many people support it precisely because it would allow them to teach their children lies that pale in comparison to the mild distortions perpetrated by the Texas school board.
On the one hand, it seems anyone who's ever used a computer before in their life would half-way expect this sort of incompatibility to arise, given the drastically different natures of Google Docs and Office (Web based vs standalone app).
Lowering your expectations is a great way to ensure the underlying problem is never addressed . The fact that we still don't have dependable multi-vendor support for some of the world's most common document interchange formats over 15 years after they were first introduced is a bit sad, don't you think?
Google Docs has always struck me as a quick and easy way to get Word documents from anywhere.
Even if you only use Docs as a distribution system, its unreliable import / export conversion can be infuriating. Things as simple as line spacing or paragraph indentation frequently get broken, and I've yet to see an embedded object that didn't get converted to an uneditable image or just dropped without any notification. Exporting from Docs can easily reduce a professional looking document to careless looking garbage.
On the other hand, how often do the people Google is trying to cater to actually use these features? [...] And I've gotta say, not many of my office reports use fancy styles, or SmartArt. Charts occasionally, yes, but the rest of those items just strike me as "meh" and SmartArt particularly strikes me as "yeah, that was cool when I was seven."
SmartArt might be dispensable, but decent styles support is essential for all but the shortest and simplest of documents. Without it, anything but flat text quickly turns into an manageable soup of conflicting format attributes.
I dunno. It just doesn't seem to me like this is going to be a problem in common usage.
The fact that many large organizations are passing over Google Docs in favor of continued dependence on Microsoft's offerings is evidence to the contrary.
The early 1980s architecture is not well suited to today's demands.
Rubbish. Its client server remote desktop architecture is perfectly suited to enterprise enviroments. Why not go find out why
Citrix et al have been playing catch up on Windows for years to try and do the same thing.
X development may have gotten easier for most developers in recent years, but at the end of the day most of that progress has been enabled by hacks within shims wrapped in frameworks all glossing over the horror that is the ICCCM. Given the inherent limitations of X (e.g. high bandwidth usage, deliberate lack of audio support), many developers would prefer to simply roll their own solutions rather than have to deal with all the legacy cruft. Apple's decision to create Quartz instead of adding new features to X is a great example of this.
My criticism of that style is that if justices can adequately decide cases without hearing arguments or asking questions, then why, according to him, do courts have proceedings?
My understanding was that he believes it is better not to distract himself with finding pointed questions to ask when he can more readily get to the core of an argument by simply listening. To him, the hearing is in fact too important to muddle with it by attempting a dialogue with the speaker.
I was mostly being a wag [...] I think his style is less good than the normal style. But, I wouldn't say I'm informed enough about it to criticize him strongly -- mildly, perhaps (as I have), but not strongly.
Our positions don't differ by much, then. Were I were in his position, I too would favor asking questions over sitting mute, but I am willing to overlook his behavior as simple personal preference.
I'm sure Thomas finds most of the questions he might have are eventually raised by another justice, anyway. I would definitely have a problem if all the justices on the court as passive as he is, however.
With this last part, you were referring to Thomas, right? the justice who famously decides the case before oral arguments are even made, and thus has no use to ever ask any questions?
Thomas has said that he refrains from asking questions because it allows him to devote the fullest possible attention to the oral arguments. If he desires to pick apart the positions presented he can do so equally well after the fact.
I am not a huge fan of Thomas' jurisprudence, either, but that doesn't mean I can't see the merit in his rationale here.
Actually, symmetric cryptographic algorithms like AES don't take much of a hit from quantum computers at all. Just double the key size, and the speed advantage of Shor's algorithm is completely negated. Contemporary asymmetric algorithms like RSA can be broken in polynomial time, but there are so-called post-quantum algorithms that cannot be broken efficiently by any known quantum algorithm, and do not require a quantum computer to use. Such algorithms are not theoretically guaranteed to be unbreakable like quantum key exchange cryptosystems are, but they are no more vulnerable to quantum computers than RSA is to deterministic computers today.
My father suffered from estropia as a child which was later surgically corrected. His eyes do not naturally focus together, so he does not have working depth perception and cannot see 3D movies. He can consciously adjust the convergence of his eyes so that they focus together, but it takes too much effort to maintain for long periods.
It doesn't seem to bother him, though. Talking to him, it seems that his terminal difficulty with parallel parking is far more annoying.
Sadly, there's no indication that the original Half-Life engine will be ported as well. That's a shame, especially since the GoldSrc engine had a mature, Quake-derived OpenGL renderer that just screams to be taken cross platform.
Valve may not have any financial interest in furthering such an old codebase, but there's nothing stopping them from giving the community access to it. People are still working on Doom, Hexen, and Quake today because id and Raven were thoughtful enough to open up their code. Valve is one of the last PC developers of real import these days; it's sad to see them let their Microsoft roots hold them back.
TECHNICALLY you can drop a running circuit into *PURE* water and nothing happens. Water isn't very conductive.
I'm fairly certain that if you tried this, the water would be rendered conductive by dissolving whatever contaminants you happen to have on the surface of the device and you'd still get a short. YMMV.
What is happening with the Stargate props now is a lot like what happened to Star Trek when they had the It's A Wrap! auction after Enterprise went off the air. At that point Trek properties had been in continuous development at Paramount for nearly 30 years, so the auction gave everyone a chance to finally get a good look at things that the different productions had been batting around and re-using for ages. A lot of fans with interest in the props really had a field day with that.
There are times I wish Gmail had never implemented conversation view. The presentation of email threads in a flat chronological progression encourages users to reply the last message in the thread regardless of which message(s) they are actually responding to.
Because of this, it's not uncommon to see threads with replies four or more levels deep when most of the participants are using Gmail. The traditional use of a tree view to track the flow of conversations becomes useless in such cases. Add in Gmail's (unalterable) behavior of placing quotes at the end of replies and you have the recipe for a very frustrating experience when participating in long discussions.
The investigators would mount the disk image on another system and attempt to crack the password hash. Government entities can be expected to have a good set of rainbow tables for whatever algorithm Windows uses.
Slashdot Mirror
Mod parent up. The domain system was meant to be strictly hierarchical, with most entities registering domains at the third or even the fourth level.
The perversion of system into a quasi-flat namespace centered around second-level domains came about because the planned high-level directory service for the Internet never materialized. These days most people use search engines to do much the same thing, albeit at the cost of standardization and protocol / vendor independence.
Judging from Apple's page, all the gestures are the same as the ones used by the current MacBook Pro trackpad. Apple didn't even bother to change the preferences panel (screenshot). Apple's claim to fame is keeping things simple as possible for consumers, and adding gestures that aren't supported by any other Apple device would tend to run against that.
Seeing as Apple has never bothered to provide official Windows driver support for their peripherals, I wouldn't hold my breath on that, either. Somebody will probably rip the appropriate bits out of Boot Camp if and when they update it to support the pad, however (this has been done before for other Apple devices).
The auto refresh is there because the article is a "live blog" which received real-time updates as the press conference progressed. Now that the conference is over, Engadget appears to have forgotten to turn it off, but removing the &refresh=60 from the end of the URL string and reloading the page should fix it.
[...] the secret ballot is such a terrible analogy for this petition that it's incredible anyone would bring it up if they have any clue whatsoever as to how secret ballots actually work, and how much effort is made to verify that people in secret ballot situations don't vote twice, and their identity matches who they say they are, and they are actually legally allowed to vote.
For all we know the names on this petition are "Donald Duck" repeated 100,000 times, or the names of closet gays (also known as Bible Believing Christians) and their minor children.
Publishing the names serves the good and useful purpose of validating that the signers are who they say they are, and that they are adults living in the State of Washington, as opposed to shills from out of state, minors, or fictional characters.
There is absolutely nothing to prevent the same checks from being applied to the signature gathering process. If all signatories provide their address along with their name, it is trivial to verify that they are indeed eligible citizens of the jurisdiction in which the petition is being circulated.
The agency responsible for verifying the signatures could even opt to contact a random sample of the signatories and confirm their participation, if they so chose. I would point out, however, that doing so goes beyond the measures taken to ensure the integrity of actual ballots. When I request an early ballot and vote by mail, the state has no way to confirm that the person who called for the ballot, filled it out and returned it was actually me.
The only practical reason for making petition signatures public is to allow them to be verified by any interested third party. The problem with this is that, for some people, "verification" involves breaking kneecaps. Why should we hold petitions to a higher standard than the elections that follow them?
Billcutus can stay; he's practically an institution at this point.
As for Jobs, I think an image of him done up in the fashion of Big Brother from the "1984" ad would be more fitting.
That's just your speculation and not supported by the facts. Why are they dumping so much money into free HTML5 dev tools and support for HTML5 apps in Webkit if their goal is to prevent cross platform apps?
Apple's move to support HTML5 is commendable, but don't doubt for an instant that they also have an interest in breaking the back of Flash. Adobe has been less than stellar in supporting the Mac lately and Apple wants to be sure they aren't beholden to them in the future.
Also, why would taking steps that make cross platform development in and of itself help Apple instead of hurt them. Generally, breaking interoperability only helps when you have dominant market share, otherwise it hurts the bottom line. So basically, your hypothesis has no support.
Apple's mindshare with casual users is tremendous. The big fight right now is not among Apple, Android, RIM, Nokia et al, but rather between Apple and Android. The latter two have traditionally targeted the business and prosumer markets and have mature offerings there. Android and the iPhone OS are both new, emerging platforms targeting home users, and in that market Apple's position is enviable. They may recently have slipped behind Android on total sales, but they still have the next gen iPhone coming down the pipe and all reports indicate that it will close the hardware gap that emerged with the release of the Nexus One.
Your argument that Apple's exclusion of Flash is based exclusively on concern about application quality is a little naive, I think. If applications developed with Adobe's tools are lower in quality than the native offerings, consumers will recognize that and opt not to use them. Saying that they will instead turn around and blame Apple for their shortcomings is a little twisted. Additional choices are a win for consumers here, the only one who stands to lose is Apple. Denying that their actions here are not motivated at least in part by unadulterated self-interest is disingenuous.
Another reason Apple is so dead-set against using Adobe Flash on their iPhone/iPads is because they would lose their exclusive development platform of XCode on their custom Mac hardware. And if Apple is anything, they are a hardware company.
I'm trying hard to understand what you're trying to say here. Apple makes piles of money selling iPhones. They make next to nothing selling Macs to iPhone developers.
Apple wants to control the development platform because it locks developers into supporting only Apple's end user hardware, namely, the iPhone. Apple's tools do not effectively support cross platform development, so anyone who wants to make an app that runs on more than just the iPhone has to write substantial amounts of additional code to support the non-Apple devices. The result is many developers opting not to target devices other than the iPhone at all, which is good for Apple because users are now pressured to buy their hardware for the exclusive applications.
If Apple allows Adobe's development tools to be used, the distinction between Apple and non-Apple hardware is abstracted away, and there is no longer any reason for developers not to make their applications available on as many devices as possible. This invites competition that Apple naturally does not want.
They have to monetize their service somehow. This seems perfectly logical and reasonable to me.
Certainly no disagreement here. I just find it a bit irksome that Twitter is glossing over this move with platitudes about preserving the relevance of their service and fostering "innovation" when it's obvious to anyone that they're just acting to protect their business model.
Of course Twitter isn't going to allow advertisers to use their API for free. They will muzzle any attempts to do so in the name of preventing spam, then turn around and charge them for the privilege.
The original author's argument consists entirely of pillorying OpenBSD for its lack of any Extended ACL framework as a second line defense against security breaches. Posters in the comments section rightly point out that OpenBSD does indeed include other second line defenses like PID randomization, ASLR, and extensive support for chroots - some of which are still not supported by default in Linux distributions today. The OpenBSD maintainers' choice to focus on ensuring the quality of the first line application and kernel codebases merely represents a different approach, no doubt motivated in part by their small number of developers. An EACL framework is essentially a bolt-on solution to help contain applications that are poorly written and difficult to isolate - it is potentially easier to address the issues in the targeted applications than to devote additional effort to the security frameworks, especially when said frameworks represent an increase in the system attack surface unto themselves.
The author completely fails to address these counterarguments, dismissing approaches like chroots as overly simplistic and holding fast to the unsupported position that properly audited security policies are somehow inherently superior to properly audited kernel and application code. There may very well be a place for Extended ACL frameworks in OpenBSD, but the linked article is certainly not enough to convince anyone.
So, you support school choice, except for those people to whom you feel superior? Get bent.
Not at all. Of course the same choices must be available to everyone; anything else is oppression, plain and simple.
I was just pointing out that choice is not a perfect solution, though I do believe it is superior to all the other ones that have been tried so far. People will find a way to do damaging things no matter what the system, and I do not think it is helpful to pretend otherwise merely for the sake of argument.
And when you leave it up to parents to decide what to teach their kids, sometimes they will be wrong, too.
I support school choice in principle, but I also recognize that many people support it precisely because it would allow them to teach their children lies that pale in comparison to the mild distortions perpetrated by the Texas school board.
On the one hand, it seems anyone who's ever used a computer before in their life would half-way expect this sort of incompatibility to arise, given the drastically different natures of Google Docs and Office (Web based vs standalone app).
Lowering your expectations is a great way to ensure the underlying problem is never addressed . The fact that we still don't have dependable multi-vendor support for some of the world's most common document interchange formats over 15 years after they were first introduced is a bit sad, don't you think?
Google Docs has always struck me as a quick and easy way to get Word documents from anywhere.
Even if you only use Docs as a distribution system, its unreliable import / export conversion can be infuriating. Things as simple as line spacing or paragraph indentation frequently get broken, and I've yet to see an embedded object that didn't get converted to an uneditable image or just dropped without any notification. Exporting from Docs can easily reduce a professional looking document to careless looking garbage.
On the other hand, how often do the people Google is trying to cater to actually use these features? [...] And I've gotta say, not many of my office reports use fancy styles, or SmartArt. Charts occasionally, yes, but the rest of those items just strike me as "meh" and SmartArt particularly strikes me as "yeah, that was cool when I was seven."
SmartArt might be dispensable, but decent styles support is essential for all but the shortest and simplest of documents. Without it, anything but flat text quickly turns into an manageable soup of conflicting format attributes.
I dunno. It just doesn't seem to me like this is going to be a problem in common usage.
The fact that many large organizations are passing over Google Docs in favor of continued dependence on Microsoft's offerings is evidence to the contrary.
X development may have gotten easier for most developers in recent years, but at the end of the day most of that progress has been enabled by hacks within shims wrapped in frameworks all glossing over the horror that is the ICCCM. Given the inherent limitations of X (e.g. high bandwidth usage, deliberate lack of audio support), many developers would prefer to simply roll their own solutions rather than have to deal with all the legacy cruft. Apple's decision to create Quartz instead of adding new features to X is a great example of this.
My understanding was that he believes it is better not to distract himself with finding pointed questions to ask when he can more readily get to the core of an argument by simply listening. To him, the hearing is in fact too important to muddle with it by attempting a dialogue with the speaker.
Our positions don't differ by much, then. Were I were in his position, I too would favor asking questions over sitting mute, but I am willing to overlook his behavior as simple personal preference.
I'm sure Thomas finds most of the questions he might have are eventually raised by another justice, anyway. I would definitely have a problem if all the justices on the court as passive as he is, however.
Thomas has said that he refrains from asking questions because it allows him to devote the fullest possible attention to the oral arguments. If he desires to pick apart the positions presented he can do so equally well after the fact.
I am not a huge fan of Thomas' jurisprudence, either, but that doesn't mean I can't see the merit in his rationale here.
Actually, symmetric cryptographic algorithms like AES don't take much of a hit from quantum computers at all. Just double the key size, and the speed advantage of Shor's algorithm is completely negated. Contemporary asymmetric algorithms like RSA can be broken in polynomial time, but there are so-called post-quantum algorithms that cannot be broken efficiently by any known quantum algorithm, and do not require a quantum computer to use. Such algorithms are not theoretically guaranteed to be unbreakable like quantum key exchange cryptosystems are, but they are no more vulnerable to quantum computers than RSA is to deterministic computers today.
My father suffered from estropia as a child which was later surgically corrected. His eyes do not naturally focus together, so he does not have working depth perception and cannot see 3D movies. He can consciously adjust the convergence of his eyes so that they focus together, but it takes too much effort to maintain for long periods.
It doesn't seem to bother him, though. Talking to him, it seems that his terminal difficulty with parallel parking is far more annoying.
War was beginning...
(Obligatory humor: Somebody set up us the BIND).
Sadly, there's no indication that the original Half-Life engine will be ported as well. That's a shame, especially since the GoldSrc engine had a mature, Quake-derived OpenGL renderer that just screams to be taken cross platform.
Valve may not have any financial interest in furthering such an old codebase, but there's nothing stopping them from giving the community access to it. People are still working on Doom, Hexen, and Quake today because id and Raven were thoughtful enough to open up their code. Valve is one of the last PC developers of real import these days; it's sad to see them let their Microsoft roots hold them back.
TECHNICALLY you can drop a running circuit into *PURE* water and nothing happens. Water isn't very conductive.
I'm fairly certain that if you tried this, the water would be rendered conductive by dissolving whatever contaminants you happen to have on the surface of the device and you'd still get a short. YMMV.
Clearly, Old Europe hate us for our freedoms. Specifically, they hate us us for the freedoms we take with their citizen's personal information.
How shocking.
What is happening with the Stargate props now is a lot like what happened to Star Trek when they had the It's A Wrap! auction after Enterprise went off the air. At that point Trek properties had been in continuous development at Paramount for nearly 30 years, so the auction gave everyone a chance to finally get a good look at things that the different productions had been batting around and re-using for ages. A lot of fans with interest in the props really had a field day with that.
There are times I wish Gmail had never implemented conversation view. The presentation of email threads in a flat chronological progression encourages users to reply the last message in the thread regardless of which message(s) they are actually responding to.
Because of this, it's not uncommon to see threads with replies four or more levels deep when most of the participants are using Gmail. The traditional use of a tree view to track the flow of conversations becomes useless in such cases. Add in Gmail's (unalterable) behavior of placing quotes at the end of replies and you have the recipe for a very frustrating experience when participating in long discussions.
The investigators would mount the disk image on another system and attempt to crack the password hash. Government entities can be expected to have a good set of rainbow tables for whatever algorithm Windows uses.