The difference between what CBS reporters did in their college newspapers and what bloggers do is thet bloggers get feedback, and (unless they're control freaks) you can watch that too...
everything which is being done _is_ being done on Unix
If it's using interfaces that are common to multiple UNIX implementations, then it's being done in UNIX. If it's using interfaces that only exist in HFS+, then it's specific to Mac OS X. Like, jails are a FreeBSD feature right now. The fact that FreeBSD is UNIX doesn't mean that jails are a feature of UNIX.
Everything you said here is completely 100% wrong.
Tell Apple then, because it's 100% based on what they've written. Things like "It's a completely new search technology that is tightly integrated with a fundamental part of the OS: The file system." and "These abilities build on the already impressive capabilities of the journaled HFS+ file system.".
If it's built into the vnode layer, as I suggested, then you should be able to use Spotlight on files in UFS, ISO-9660, or even FAT partitions. Can you?
It's tied to the FS in that it hooks in to find out when files are changed so that it can try to update its metadata store.
And that goes back to my original message. The file system is the wrong place to do it. The layer above the file system that goes "here's a system call, here's a file object, look up the file system for that object, tell it to do the operation" is a much better place. The FreeBSD kqueue mechanism allows this kind of efficient monitoring at a file level, it would need to be extended... maybe in EVFILT_VNODE or adding EVFILT_VNODE_ALL to watch all open vnodes...
For something like Spotlight, this isn't that big a deal. When you move data out of the system, you don't care, because Spotlight is all about the data in the system. When you bring it in, you have the spotlight daemon (or whatever Apple calls that component) dig through the file in the background and regenerate the metadata based on what it finds there.
That could include: the file name, the contents of the file, embedded info in the file (id3 tags), the contents of archives (dmg files, zip files, and so on) and metadata files inside them (README, *.plist, *.NIB, LICENSE, Install, Makefile, etc).
Since most applications aren't Spotlight-aware, Spotlight already has to depend to a large extent on metadata in the files, and will probably always have to.
Where does it come from in OS X? It comes from the application that creates the file, from the file name, from the finder info, and from reading the file itself. In UNIX, the finder info isn't there, but everything else is... and there's already a huge database of file type information in/etc/magic that's a lot more reliable than finder info.
On OS X, existing applications that aren't spotlight-aware will just create files the way they do now. Some of them provide finder info and some don't. Some of them add resources Spotlight can use, others don't. New applications can call the API to add more information to the metadata.
On UNIX, much the same thing can be done. That's the easy part, really. It's potentially a lot of code, but it's not essentially much different from tools like glimpse/harvest and file/grep/netpbm/sox/...
See, the core of Spotlight isn't mysterious and spooky, and it doesn't need to be built into the file system... it's the code that takes all the file information and creates indexes that it can use for searching. Because it's aware of file creation and deletion, file moving, name changes, and so on... it can do so much more efficiently than tools like glimpse/harvest. Because it's aware of file types, it can do a more complete job than glimpse. But almost everything it's doing could be done just as easily and efficiently on UNIX... IF there was a component that could tell when file changes were being made so it could update its database instead of having to regenerate it by an exhaustive search of the file system at some period that's infrequent enough it doesn't interfere with normal use.
he notes that the problem is largely with C/C++ and mostly because of the buffer overflow problems.
Most of the security problems that really turn into a bear with Windows aren't buffer overflows. They're layering problems. Windows doesn't have a strong distinction between different layers, it doesn't really have any internal security boundaries. It's got a complex privilege model that's wide open to privilege boosting, and applications have to be granted far too many privileges to do their normal operations... and because privileges can't be associated with applications that means a user has to be given all the privileges ANY application he uses will ever need. On top of that, "security zones" mean that if you can trick some component (the HTML control, of course) into thinking you're in the right zone it'll grant you full "local user" privileges and let you run any damn executable or script you want.
On the server side, there's all these spooky connections between application services and network services, so that you can't keep the system from leaving listening ports into important services open, and you can't firewall them off unless you want to shut down native network support completely.
THIS is the problem with Windows security. It's not just that it's a monoculture, it's a culture with security flaws baked into the APIs that can't be fixed without breaking applications.
The brain behind Spotlight is Dominic Giampaolo, the same guru that wrote the fantastic BeFS for BeOS.
Which explains why it's tied to the filesystem rather than using a general hook at the vnode layer to allow the same functionality to be implemented regardless of the filesystem in use. Having the filesystem support it would make it more efficient on HFS+ but it should be possible on UFS, ISO 9660 CDs, or even over NFS or SMB.
In fact, the way it's described... with one metadata store per filesystem rather than per file, and user-level metadata provided by applications... this is something that FreeBSD or Linux could implement right now, over any file system: all they would need would be a mechanism for the vnode layer to send messages to a usermode daemon that tracked inode operations (eg, creation, deletion, maybe mode changes or date changes, and renames) in a name-inode database (any database, including Postgres or MySQL) and updated any associated metadata in the background.
This could be done with negligable slowdown for file operations: the index can be updated asynchronously, because it can always be recreated in the background after a crash, so the vnode operation won't ever have to wait for the daemon to respond... and changes to the metadata are all in userspace.
However, this will only affect users who have Windows Scripting Host enabled and certain ActiveX controls, according to MessageLabs.
If only Microsoft would back out of this insistence on making the browser a completely general web applications framework with the ability to provide full access to local resources.
Microsoft: split the HTML rendering engine out of the web client components, and get rid of the "security zones" hacks. You've been trying to come up with a design that lets you do this safely for over seven years now, and never succeeded in holding off attackers for more than a few weeks at the most... it's time to admit that even all the brilliant people at Microsoft (and you have some bloody amazing blokes over there) won't be able to make it work. Please consider that you may have been mistaken.
The first multi-user online "Dungeon" wasn't bartle's MUD. It would have either been Empire on Plato, or the People's Computer Company's "Public Caves", both from the late '70s. The latter wasn't concurrent multi-user, but operated as a bullten board rather than a chat system (as did most online sustems at the time), but the interactions between people were very similar to the ones on MUDs.
The democrats have been scaring seniors for years now about social security and medicare, and they havent done crap.
Keeping them solvent, and even producing a surplus, isn't "doing crap", so I guess they haven't crapped on them the way George Bush has. I'm not sure why not "doing crap" is supposed to be a bad thing in this context... perhaps I'm misunderstanding you?
Bush keeps flip-flopping on free trade. 2000, he's all for it. 2002, he imposes tariffs on steel. 2004, election year, he makes free trade an issue and gets rid of them again. 2006... who knows?
All you have to do is keep enough people scared enough about a problem, and they'll vote for you even if everything you have done about the problem has made it worse.
I'd like one Bush voter, just one, to do two things.
One, demonstrate that he has actually paid attention to the news and to what Bush has actually said and done. Anyone who isn't aware that Bush has acknowledged that there was no link between 911 and Saddam, and that there was no active or planned atomic/biological/chemical warfare program in Iraq, is ineligable for this poll.
Two, explain why they voted for Bush in rational, practical terms. What they believe Bush will do during the next four years that might actually improve the international or domestic situation. Perhaps some specific promise that they believe he has the will and ability to carry out. Keep the first part of your answer in mind here, if you believe he's going to follow through, can you point to similar promises he made in 2000 that he's followed through on?
Back in the '80s, my boss had one of the first PCs in the building with a hard disk. One day he asked me to copy some files off onto a floppy, so I put the floppy in the drive and typed "format", as I was used to doing...
C:>FORMAT Insert floppy into drive C: and hit return.
The rest is history. As was everything on the drive.
Why bother to couple GeForce GPUs with the Macs, if you aren't going to bother to utilize that power for games?
Because Apple's built an OS and GUI that take good advantage of it. Besides, the GeForce 5200 and Radeon 9200... the standard Apple GPUs... are pretty cheap and anemic.
who is really paying more? The Mac users or the Windows users?
The Mac users.
I'm a Mac user.
Yeh yeh, mumble PCI Express mumble mumble... but for maybe $700 I could put together a PC using off the shelf parts that might not be anywhere near a top of the line PC game machine (and I could build a G5 Powermac that'll knock its socks off) but it'll be at least as good a game machine as a G5 iMac or G4 eMac.
That's with something like an Athlon 64 3400+ (which seems to have a price-performance sweet spot), an ATI Radeon 9800, half a gig of RAM, good solid name-brand parts. Not exactly an enthusiast machine, but it'll play current games at a reasonable resolution with frame rates that won't make you choke.
Your eMac G4 has a built-in monitor, but it's not worth anything: it's a crummy 17" shadow mask with dots I can see from six feet away... without my glasses! It's got a Radeon 9200 with 32M, not a 9800 with 128M. It's got 1/4 the RAM. And the G4 might be a bit more efficient than the 3400+, but with almost an extra GHz on the AMD side I'm not sweating it. Price: about par. Winner: AMD, no question.
The iMac G5 has a 1.6 GHz G5... but it needs as long a pipeline as a P4 to get that speed, so I suspect the 3400+ is still faster. And the video is a *censored* nVidia 5200. I just upgraded my wife's game box from that to an ATI 9600 and it's like night and day. Price: almost twice as much. Winner: AMD.
Neither of these Macs are comparable to what's really a bottom of the line entry-level machine for a novice gamer.
For the price of an iMac, I can upgrade to a 9800 Pro, double the RAM to a full gig, and go with a 3700+.
To get a Powermac that's going to compare to that, well... let's see. Let's assume the 1.8 GHz G5 is comparable to the Athlon-64 3700+... I don't know if it is, but we'll give Apple the benefit of the doubt. After you bring the RAM up to 1GB, and upgrade to a 9800XT you're looking at a cool two grand.
From the article: PC gamers would might begin to believe that if they got a Mac--which wouldn't cost any more than their custom-built cold-cathoded LED-fanned gaming PC--they would be able to play games on their computer, ported from consoles like the Playstation 2 and the XBox, earlier than if they stuck with Windows.
You CAN build a PC that costs as much as a comparable Mac, but you'll have to work at it. For the extra $500-$700 I could get a PS2 and an XBox and a bunch of games. And there's a bunch of games that never come out on the consoles... those will still come out for the PC first.
Yeh? Care to tell my copy of VLC where the -ing hell it is?
main: no suitable decoder module for fourcc `WMV3'. VLC probably does not support this sound or video format.
(and WM Player on my Mac dropped all but about 12 random frames... yeh, it's only a G4/466 but sheesh, it plays DVDs full screen without complaining...)
Nice video, it'd be cooler without the eclipse and eclipse watchers distracting you from the clouds. All it needs is some Phillip Glass music (or maybe early Floyd) and a nice long loop...
My gripe with WMV is that Microsoft's media player for OS X sucks the eclipsed moon through quantum wormholes. Real, too... hey guys, there's no blkoody reason a media player can't use native widgets. Yeh, yeh, I know Apple started it with Quicktime 4, but there's no reason to copy a bad idea even if Apple does implicitly say it's OK this time.
Slashdot Mirror
So he wants to sit on the Bench with the other SCOTUSes and burn heretics?
Well you can just rock me to sleep tonight.
The difference between what CBS reporters did in their college newspapers and what bloggers do is thet bloggers get feedback, and (unless they're control freaks) you can watch that too...
Something I wrote about this 5 years ago...
everything which is being done _is_ being done on Unix
If it's using interfaces that are common to multiple UNIX implementations, then it's being done in UNIX. If it's using interfaces that only exist in HFS+, then it's specific to Mac OS X. Like, jails are a FreeBSD feature right now. The fact that FreeBSD is UNIX doesn't mean that jails are a feature of UNIX.
Everything you said here is completely 100% wrong.
Tell Apple then, because it's 100% based on what they've written. Things like "It's a completely new search technology that is tightly integrated with a fundamental part of the OS: The file system." and "These abilities build on the already impressive capabilities of the journaled HFS+ file system.".
If it's built into the vnode layer, as I suggested, then you should be able to use Spotlight on files in UFS, ISO-9660, or even FAT partitions. Can you?
It's tied to the FS in that it hooks in to find out when files are changed so that it can try to update its metadata store.
... maybe in EVFILT_VNODE or adding EVFILT_VNODE_ALL to watch all open vnodes...
And that goes back to my original message. The file system is the wrong place to do it. The layer above the file system that goes "here's a system call, here's a file object, look up the file system for that object, tell it to do the operation" is a much better place. The FreeBSD kqueue mechanism allows this kind of efficient monitoring at a file level, it would need to be extended
For something like Spotlight, this isn't that big a deal. When you move data out of the system, you don't care, because Spotlight is all about the data in the system. When you bring it in, you have the spotlight daemon (or whatever Apple calls that component) dig through the file in the background and regenerate the metadata based on what it finds there.
That could include: the file name, the contents of the file, embedded info in the file (id3 tags), the contents of archives (dmg files, zip files, and so on) and metadata files inside them (README, *.plist, *.NIB, LICENSE, Install, Makefile, etc).
Since most applications aren't Spotlight-aware, Spotlight already has to depend to a large extent on metadata in the files, and will probably always have to.
but where will the metadata come from?
/etc/magic that's a lot more reliable than finder info.
Where does it come from in OS X? It comes from the application that creates the file, from the file name, from the finder info, and from reading the file itself. In UNIX, the finder info isn't there, but everything else is... and there's already a huge database of file type information in
On OS X, existing applications that aren't spotlight-aware will just create files the way they do now. Some of them provide finder info and some don't. Some of them add resources Spotlight can use, others don't. New applications can call the API to add more information to the metadata.
On UNIX, much the same thing can be done. That's the easy part, really. It's potentially a lot of code, but it's not essentially much different from tools like glimpse/harvest and file/grep/netpbm/sox/...
See, the core of Spotlight isn't mysterious and spooky, and it doesn't need to be built into the file system... it's the code that takes all the file information and creates indexes that it can use for searching. Because it's aware of file creation and deletion, file moving, name changes, and so on... it can do so much more efficiently than tools like glimpse/harvest. Because it's aware of file types, it can do a more complete job than glimpse. But almost everything it's doing could be done just as easily and efficiently on UNIX... IF there was a component that could tell when file changes were being made so it could update its database instead of having to regenerate it by an exhaustive search of the file system at some period that's infrequent enough it doesn't interfere with normal use.
he notes that the problem is largely with C/C++ and mostly because of the buffer overflow problems.
Most of the security problems that really turn into a bear with Windows aren't buffer overflows. They're layering problems. Windows doesn't have a strong distinction between different layers, it doesn't really have any internal security boundaries. It's got a complex privilege model that's wide open to privilege boosting, and applications have to be granted far too many privileges to do their normal operations... and because privileges can't be associated with applications that means a user has to be given all the privileges ANY application he uses will ever need. On top of that, "security zones" mean that if you can trick some component (the HTML control, of course) into thinking you're in the right zone it'll grant you full "local user" privileges and let you run any damn executable or script you want.
On the server side, there's all these spooky connections between application services and network services, so that you can't keep the system from leaving listening ports into important services open, and you can't firewall them off unless you want to shut down native network support completely.
THIS is the problem with Windows security. It's not just that it's a monoculture, it's a culture with security flaws baked into the APIs that can't be fixed without breaking applications.
The brain behind Spotlight is Dominic Giampaolo, the same guru that wrote the fantastic BeFS for BeOS.
Which explains why it's tied to the filesystem rather than using a general hook at the vnode layer to allow the same functionality to be implemented regardless of the filesystem in use. Having the filesystem support it would make it more efficient on HFS+ but it should be possible on UFS, ISO 9660 CDs, or even over NFS or SMB.
In fact, the way it's described... with one metadata store per filesystem rather than per file, and user-level metadata provided by applications... this is something that FreeBSD or Linux could implement right now, over any file system: all they would need would be a mechanism for the vnode layer to send messages to a usermode daemon that tracked inode operations (eg, creation, deletion, maybe mode changes or date changes, and renames) in a name-inode database (any database, including Postgres or MySQL) and updated any associated metadata in the background.
This could be done with negligable slowdown for file operations: the index can be updated asynchronously, because it can always be recreated in the background after a crash, so the vnode operation won't ever have to wait for the daemon to respond... and changes to the metadata are all in userspace.
Kerry lost largely on high voter turnout for those who opposed him on moral grounds, especially gay marriage.
Boy, that sure puts Bush in his place, all he had goin' for him was lyin' cheatin' and theivin'.
However, this will only affect users who have Windows Scripting Host enabled and certain ActiveX controls, according to MessageLabs.
If only Microsoft would back out of this insistence on making the browser a completely general web applications framework with the ability to provide full access to local resources.
Microsoft: split the HTML rendering engine out of the web client components, and get rid of the "security zones" hacks. You've been trying to come up with a design that lets you do this safely for over seven years now, and never succeeded in holding off attackers for more than a few weeks at the most... it's time to admit that even all the brilliant people at Microsoft (and you have some bloody amazing blokes over there) won't be able to make it work. Please consider that you may have been mistaken.
The first multi-user online "Dungeon" wasn't bartle's MUD. It would have either been Empire on Plato, or the People's Computer Company's "Public Caves", both from the late '70s. The latter wasn't concurrent multi-user, but operated as a bullten board rather than a chat system (as did most online sustems at the time), but the interactions between people were very similar to the ones on MUDs.
The democrats have been scaring seniors for years now about social security and medicare, and they havent done crap.
Keeping them solvent, and even producing a surplus, isn't "doing crap", so I guess they haven't crapped on them the way George Bush has. I'm not sure why not "doing crap" is supposed to be a bad thing in this context... perhaps I'm misunderstanding you?
bush if more for free trade than kerry
Bush keeps flip-flopping on free trade. 2000, he's all for it. 2002, he imposes tariffs on steel. 2004, election year, he makes free trade an issue and gets rid of them again. 2006... who knows?
Bush won the popular vote.
Terrifying, isn't it?
All you have to do is keep enough people scared enough about a problem, and they'll vote for you even if everything you have done about the problem has made it worse.
It's like watching gangrene spread.
I'd like one Bush voter, just one, to do two things.
One, demonstrate that he has actually paid attention to the news and to what Bush has actually said and done. Anyone who isn't aware that Bush has acknowledged that there was no link between 911 and Saddam, and that there was no active or planned atomic/biological/chemical warfare program in Iraq, is ineligable for this poll.
Two, explain why they voted for Bush in rational, practical terms. What they believe Bush will do during the next four years that might actually improve the international or domestic situation. Perhaps some specific promise that they believe he has the will and ability to carry out. Keep the first part of your answer in mind here, if you believe he's going to follow through, can you point to similar promises he made in 2000 that he's followed through on?
Back in the '80s, my boss had one of the first PCs in the building with a hard disk. One day he asked me to copy some files off onto a floppy, so I put the floppy in the drive and typed "format", as I was used to doing...
C:>FORMAT
Insert floppy into drive C: and hit return.
The rest is history. As was everything on the drive.
Why bother to couple GeForce GPUs with the Macs, if you aren't going to bother to utilize that power for games?
Because Apple's built an OS and GUI that take good advantage of it. Besides, the GeForce 5200 and Radeon 9200... the standard Apple GPUs... are pretty cheap and anemic.
who is really paying more? The Mac users or the Windows users?
The Mac users.
I'm a Mac user.
Yeh yeh, mumble PCI Express mumble mumble... but for maybe $700 I could put together a PC using off the shelf parts that might not be anywhere near a top of the line PC game machine (and I could build a G5 Powermac that'll knock its socks off) but it'll be at least as good a game machine as a G5 iMac or G4 eMac.
That's with something like an Athlon 64 3400+ (which seems to have a price-performance sweet spot), an ATI Radeon 9800, half a gig of RAM, good solid name-brand parts. Not exactly an enthusiast machine, but it'll play current games at a reasonable resolution with frame rates that won't make you choke.
Your eMac G4 has a built-in monitor, but it's not worth anything: it's a crummy 17" shadow mask with dots I can see from six feet away... without my glasses! It's got a Radeon 9200 with 32M, not a 9800 with 128M. It's got 1/4 the RAM. And the G4 might be a bit more efficient than the 3400+, but with almost an extra GHz on the AMD side I'm not sweating it. Price: about par. Winner: AMD, no question.
The iMac G5 has a 1.6 GHz G5... but it needs as long a pipeline as a P4 to get that speed, so I suspect the 3400+ is still faster. And the video is a *censored* nVidia 5200. I just upgraded my wife's game box from that to an ATI 9600 and it's like night and day. Price: almost twice as much. Winner: AMD.
Neither of these Macs are comparable to what's really a bottom of the line entry-level machine for a novice gamer.
For the price of an iMac, I can upgrade to a 9800 Pro, double the RAM to a full gig, and go with a 3700+.
To get a Powermac that's going to compare to that, well... let's see. Let's assume the 1.8 GHz G5 is comparable to the Athlon-64 3700+... I don't know if it is, but we'll give Apple the benefit of the doubt. After you bring the RAM up to 1GB, and upgrade to a 9800XT you're looking at a cool two grand.
From the article: PC gamers would might begin to believe that if they got a Mac--which wouldn't cost any more than their custom-built cold-cathoded LED-fanned gaming PC--they would be able to play games on their computer, ported from consoles like the Playstation 2 and the XBox, earlier than if they stuck with Windows.
You CAN build a PC that costs as much as a comparable Mac, but you'll have to work at it. For the extra $500-$700 I could get a PS2 and an XBox and a bunch of games. And there's a bunch of games that never come out on the consoles... those will still come out for the PC first.
And it's another physorg dead-end. Rather than mirror it or anything, a little googling will find the original material. Here's The original spam-free press release and Professor Wang's home page with a full citation for the paper.
If you're running gentoo and you're not recompiling your kernel at boot you're just a poser!
Its already on Gnutella
:)
Someone want to point me to a decent gnutella proxy then?
video lan client has a great codec for wmv
Yeh? Care to tell my copy of VLC where the -ing hell it is?
main: no suitable decoder module for fourcc `WMV3'.
VLC probably does not support this sound or video format.
(and WM Player on my Mac dropped all but about 12 random frames... yeh, it's only a G4/466 but sheesh, it plays DVDs full screen without complaining...)
Nice video, it'd be cooler without the eclipse and eclipse watchers distracting you from the clouds. All it needs is some Phillip Glass music (or maybe early Floyd) and a nice long loop...
Yeh...
Trippy...
votes for automatic torrents
Can someone point me to a decent torrent proxy?
Hey, they didn't encrypt the movie.
My gripe with WMV is that Microsoft's media player for OS X sucks the eclipsed moon through quantum wormholes. Real, too... hey guys, there's no blkoody reason a media player can't use native widgets. Yeh, yeh, I know Apple started it with Quicktime 4, but there's no reason to copy a bad idea even if Apple does implicitly say it's OK this time.
[metal delenda est]