By Puritanical American ethics, yes, drugs are *so* much more evil than pirated music. But in a sane World, we realize that in and of itself, Marijuana is no more dangerous than tobacco and alcohol. Some have at least claimed its a jumping off point for harder drugs, and while that may have some truth to it, I wonder if that would stiil be the case if it were legal. If you had to obtain Tobacco illegally, I bet it too would be a jumping off drug. And yes, it is legal, but stealing is not, so this isn't so strange. That's like saying America is weird becaus the RIAA is suing left and right and tobacco is legal, and that's far more evil than pirated music. Of course, maybe there is something to be said for that argument after all:)
I don't think that is a good comparison. Napster, FastTrack, Scourc Exchange, etc. are not marketed as non-infriging, the implications are clear that it's intended primary use is against copyright. The RIO/BetaMax comparisons aren't really valid here. A more valid comparison would be protocols such as ftp, http, and SMB. Combined with search engines, the first two *can* be used to the same purposes, but its clear that the *intent* of these protocols is not to fly in the face of copyright. Windows file sharing protocol seem like a strange thing to bring up? IIRC, before Napster or Scour Exchange, Scour had a client that did mostly the same thing as Napster, but less conveniently and used the SMB protocol as its method of operation. So we can see cases where all three have been used to infringe on copyright, but we don't see anyone making the stupid suggestion of shutting down these protocols.
I think in these cases of the RIAA and MPAA going after *for-profit* companies whose primary business is copyright infringement, RIAA and MPAA are in the right. It is in cases like DeCSS that I think MPAA and the RIAA are in the wrong (very wrong).
How could they do any better than Playstation2 did this early? The only way they may be selling more Gamecubes than Sony sold Playstation 2s is by having more available... Playstation 2s completely sold out for a long time... I would think it would be quite some months before you could make a call like that... Since I can go to the corner K-Mart and pick up a Gamecube, I find it hard to believe that they are doing better now than Sony did last year when there were no Playstations 2s to be had moments within hitting the shelves.
Besides, the X-Box and Gamecube are only marginally better hardware-wise, if at all, yet the game selection is horrid thusfar. I predict it will be like the original Playsation all over again, first to market does better even in the face of superior consoles that follow, through sheer numbers of available titles.
But if the Gamecube truly *is* more popular, we'll have to wait until about April of next year to see for sure..
But the original suggestion was to run only one cable in the wall and not put the hub/switch on the floor. With this solution, your core switch would stay just as small. Your suggestion has them running four cables again, and this is still inconsistant with the other ways in which the switch is evaluated in the review, against other 4-port switches.
Of course, the bottom line for the target customers (businesses), having a tidier appearance can be worth more than the money saved by using a standalone switch.
But Windows NT is *kinda* that deal. Granted it is not OpenBSD based, but the NT kinda rose from the ashes of VMS. Not linux compatible, but still a sort of *nix aimed directly at competitors.
I'm interested in running OpenBSD for my NAT gateway, though I'm left with a lingering issue..
Does OpenBSD include any support for decent irc connection tracking like what is available in iptables for linux? I have people behind the gateway that use DCC within IRC, and without good connection tracking, I'm not sure how to go about securely allowing one or more people to use IRC and have DCC work.
Everything else I plan on using this system for (software RAID, NIS+, samba PDC and fileserver, NFS) seems to be fine, but this one little nitpick of mine may keep me off of OpenBSD.
Also, how is the raid implementation as far as moving the array from one openbsd install to another, and is there any semblance of lvm there? The volume management stuff w/ resizable partitions would be nice, but by no means necessary..
Well, there is SSH, and if you want, you could have them use MindTerm, which has an attempt at a sFTP to local FTP proxy built in. Never got it to work, but it could be just the ticket to letting Windows users use their favorite FTP client to access sFTP....
http://www.mindbright.se/mindterm/ is mindterms url, check it out.
On the other hand, putty has both pscp, and a psftp in development, that also provides this functionality.
Ftp is a really ugly and insecure protocol, it needs to die:) Even if it were secure, the whole mechanism sucks, can't easily be port-forwarded, and uses two ports. Those ports just makes it that much easier for a sniffer to differentiate between transefered data, and command data (i.e. passwords in plaintext).
Sure they could, if they could log your keystrokes, then they could log the data between the thumb/face reader and the computer. Then they can fake the input from the device, biometric approach defeated. The lesson here is that once they can into your system and interact with stuff like keyboard I/O, they can get pretty much any I/O your computer does, including biometric device I/O.
Interesting that slashdot made it to the white list. It may be that bad images may not make it through, but with some if the comments and ASCII art trolls, I'm surprised Disney was willing to risk it..
A lot of people are complaining they have to pay cash for WineX, not quite true. Only pre-packaged versions. The source is available under the terms of the Aladdin license. If this codebase is not complete, would someone please correct me. I personally maintain two different installs of Wine, the main branch of winehqs wine, checked out of CVS every so often, and the WineX branch, again from CVS ever so often. Why both? The WineX branch provides DirectX support, but other advances in the Wine project are slow to get into WineX, for example I can get the QuickTime player to install under standard Wine CVS builds, but not WineX.
So if you don't feel like paying, build the source if you can. Takes a long while, and if it fails, oh well, shell out a bit of cash for your software for once:)
Transgaming is a decent company, even if they don't bend over backwards to give away all their work for free. This is a bad expectation for Linux users to have, and when commercial Linux projects receive such criticism, it makes commercial vendors more and more hesitant about releasing linux products.
Well, truform I'll give, though in practice any "guessing" about what the game intended the geometry to be produces some problems (puffy characters come to mind). I think there are GeForce boards out there that can push dual monitors. As far as smoothvision goes, that is just a really hyped up supersampling AA solution, dog slow and even at highest quality doesn't do much better than nVidia w/ Antisotropic filtering, which is much faster. Now when they release the All-In-Wonder 8500DV, then you have some kick-ass features. That is what I'm holding out for, great 3D and Multimedia:)
That's good to hear, but has the gameplay changed significantly from just being able to basically slide left and right? The graphics in the screenshots still seem sub-standard to me.
All over the place, a board is marked as a clear winner, even if another board shows the *exact* same result. The archiving benchmark where MSI was top dog was interesting, what about the Epox which scored the exact same thing? Also, the other boards had less than 0.5% difference from the winner, I doubt that is statistically significant. How many runs were done? With the kernel compile, again, how many runs were done, in what order? The same config? I seriously doubt motherboards with the same chipset would have such a dramatic difference as 15 seconds on Kernel Compilation, given equivalent Disk I/O and CPU...
Did you actually fill all the banks? If so, did you encounter any problems? I'm shopping around and am looking for something that will be both decent performance and stable. I have heard all good things about the Shuttle board, but have also heard that the system can get flaky with all 4 banks filled due to memory timing issues.
Whether or not a chip is embedded has no bearing on how much CPU utilization is required for it to work. The video chipset on nForce boards, for example, is a GeForce2MX, a full-fledged chipset that does not heavily rely on the CPU. You can have dumb devices (i.e. WinModems) on PCI cards too. The complaint I have against embedded components is when they constrict upgrade paths (i.e. some motherboards with integrated graphics opt not to have an AGP slot). Additionally, I rarely find the provided integrated components sufficient, and don't want to pay the additional cost for something I plan on replacing immediately. Integrated is not the way to go for enthusiasts, typically, in the same way component stereo systems that can be mixed and matched are better for audiophiles than, say a bookshelf stereo.
I may be mistaken, but I have heard that most of these 'so-called' IDE raid controllers that motherboard manufacturer's put on their board are little more than typical IDE controllers. The manufacturer's get away with calling them RAID because they implement the RAID functionality in the Windows drivers. If this is true, I could care less about the inclusion of RAID controllers apart from having that many more ports on the MB. The only advantage is that you get a software RAID implementation for Win9x/ME, but under Linux, *BSD, and Win2k/XP there are provided facilities for doing software raid on arbitrary block devices. Additionally, I've heard the OS implementations are often both more efficient and more reliable than the drivers for these cheap IDE RAID controllers.
Also, they mention the Shuttle MB having 4 DIMM slots, and not going far beyond that. Does anyone have experience with having all 4 banks in use? I have heard theat the Shuttle MB has problems with memory timings when all 4 are in use. Anyone want to enlighten me?
I remember downloading it and thinking it was kinda neat. An OpenGL game written with linux in mind, and it's somewhat entertaining. While I think it above many other GPL games in many respects, I don't think it's so great as to be worth anything money wise when you look at the market out there.
The gameplay is pretty straightforward and boring. Doesn't even seem as interesting as, say, the snowboarding mini-game in Final Fantasy VII, which wasn't even meant to stand on its own. Snowboarding games and the like typically offer a significant deal more than TuxRacer, and for this reason if I was going to shell out cash for a game of this genre, I'd go with a good game.
For another thing, the graphics are not that spectacular. The scenery has some nice textures, but the characters and objects are simple gouraud shaded polygons, and even then the polygon count in the player models and how they are put together is now substandard. I understand that having a low polygon count helps performance, but companies like Square show how you can really have some decent looking graphics without complex geometry.
All in all it was a neat little game that kept my attention for a few minutes when I first got it. It's not on the level of any commercial competition in my opinion, considering games from 97 have roughly equivalent graphics and the gameplay is really boring and repetitive, with next to nothing to spruce it up.
I was under the impression that the corruption could be corrected by an fsck, and that while the ext structure was invalid, the data was left intact, that the issue was more with lock files, so an fsck would restore a valid inode table, and the actual data should still be intact, did I misunderstand this? I didn't ever actually use the beast, so I could be wrong...
I know that part of the Object Desktop package has been doing this for a long time now. http://www.stardock.com/ The package in question, IIRC, is DesktopFX. Neat toy, but nothing really that useful or newsworthy, more useful than the 'fake' alpha blending most commonly seen in nix (only blend with the root window, ignore all others). For nearly real alpha blending, you could use KDE and use mosfet's (www.mosfet.org) liquid theme and at least have alpha blending on the menus. The reason I say almost real, is that while it does blend against windows as well as the background, it only blends against the screen as it was when it was first drawn, if background changes, the liquid alpha-blended menu does not. Xrender hints at the ability to do true alpha blending w/ hardware help, but I haven't seen it actually used for anything except AA-text.
While Alpha-blended windows give nifty screenshots and initial "ooh" factor, people switch it back off in moments because it really makes programs harder to use in the long run. It's hard enough to make sure colors within an application always have text that is readable against the background without other applications lending their colors to further mess things up.
Well not quite, because you could claim it was a backup copy of a disc that was either lost or stolen. At which point it should be your word against theirs. The "ink" example is different, it is a presence of something that should not be there. The lack of being able to present a hologram is much different.
Case in point, I had, among other things, a CD binder from my residence stolen. Most things were lost for good, except a few things I had copied, and I now copy *everything* I buy and store it seperately. If someone catches me with my backup (legal), why should I be punished because I can't present the original? Granted in this case I could reference the police report from the robbery, but there are legitimate reasons why people may not be able to present the pretty holograms to prove they purchased the software, and this would be unfair.
They likely have something in development. They have done a pretty good job of catchup, the Radeon was leaps and bounds better than the Rage series, but still nowhere near equivalent GeForce performance.
Now the 8500 chips are showing themselves to be, with decent drivers, at least roughly equivalent to the high-end GeForce3 chips. Though perhaps not quite as feature-rich, performance is quite competitive with nVidia offerings. If they keep up this pace, they may very well not only catch up with nVidia both performance and feature wise, but also pass them.
In a way, the situation right now seems to me reminiscent of nVidia's situation when they brought out the RivaTNT chips. They had been playing catchup to 3dfx, and the Riva128 series would be analagous to ATI's Radeon chips, and the first TNT analogous to the 8500 chips. Then people thought nVidia's progress was impressive, but thought they probably would never quite catch up with 3Dfx, and we all know how that went..
Basically, I would say that since the release of the first GeForce chip that nVidia development has slowed down and not made nearly the significant strides seen between the Riva, TNT, and that ended with the GeForce, since the market didn't pressure them enough, and ATI is taking advantage and may take the crown in the somewhat-near future....
Note he was talking about IE, which has a distinct lack of any MDI mode. But on the subject, I like being able to middle click and have a document load in a background tab. I can't display two pages at the same time in MDI, have to do SDI then, but that is a much rarer case. If Opera can do pop-under MDI, and have a "maximize by default" so it resembles a tab interface, I'd be much more interested. Having the option of windowed MDI is nice, but the tab paradigm has advantages..
Of course the concept that those holograms and such could be legally used to prosecute piracy is scary to me. Why? Because this means that when asked, you must *prove* that you did not pirate the software, else, you are considered guilty. Completely discards the "innocent until proven guilty". Of course, software-piracy end-users could, under the strict definition of the concept, almost never be proven guilty, but the re-distibrutors could be..
No, pro has it too, it's just that the 'corporate' version does not, which is just a copy of pro legally cracked by MS:) Well, maybe crack's the wrong word, but the mere existence of this corp edition must be really appreciated by the warez scene, while the legal users just get annoyed and hassled more.
When I first tried out tabbed interface MDI in a web browser, instantly recognized how much better it was. Not like you're forced into it, by default it will act non-MDI, just like IE. With this, you save screen space, it's much easier to keep organized. Only down side is when you decide you need information from two tabs at once. Hopefully you recognize this well in advance and used windows, but still, it would be nice to instantly "detach" a tab into a new window, and attach again once you finish, preserving all the state info, but still, galeon's my favorite,. Mozilla has tabs, but the interface to them still sucks..
Slashdot Mirror
By Puritanical American ethics, yes, drugs are *so* much more evil than pirated music. But in a sane World, we realize that in and of itself, Marijuana is no more dangerous than tobacco and alcohol. Some have at least claimed its a jumping off point for harder drugs, and while that may have some truth to it, I wonder if that would stiil be the case if it were legal. If you had to obtain Tobacco illegally, I bet it too would be a jumping off drug. And yes, it is legal, but stealing is not, so this isn't so strange. That's like saying America is weird becaus the RIAA is suing left and right and tobacco is legal, and that's far more evil than pirated music. Of course, maybe there is something to be said for that argument after all :)
I don't think that is a good comparison. Napster, FastTrack, Scourc Exchange, etc. are not marketed as non-infriging, the implications are clear that it's intended primary use is against copyright. The RIO/BetaMax comparisons aren't really valid here. A more valid comparison would be protocols such as ftp, http, and SMB. Combined with search engines, the first two *can* be used to the same purposes, but its clear that the *intent* of these protocols is not to fly in the face of copyright. Windows file sharing protocol seem like a strange thing to bring up? IIRC, before Napster or Scour Exchange, Scour had a client that did mostly the same thing as Napster, but less conveniently and used the SMB protocol as its method of operation. So we can see cases where all three have been used to infringe on copyright, but we don't see anyone making the stupid suggestion of shutting down these protocols.
I think in these cases of the RIAA and MPAA going after *for-profit* companies whose primary business is copyright infringement, RIAA and MPAA are in the right. It is in cases like DeCSS that I think MPAA and the RIAA are in the wrong (very wrong).
How could they do any better than Playstation2 did this early? The only way they may be selling more Gamecubes than Sony sold Playstation 2s is by having more available... Playstation 2s completely sold out for a long time... I would think it would be quite some months before you could make a call like that... Since I can go to the corner K-Mart and pick up a Gamecube, I find it hard to believe that they are doing better now than Sony did last year when there were no Playstations 2s to be had moments within hitting the shelves.
Besides, the X-Box and Gamecube are only marginally better hardware-wise, if at all, yet the game selection is horrid thusfar. I predict it will be like the original Playsation all over again, first to market does better even in the face of superior consoles that follow, through sheer numbers of available titles.
But if the Gamecube truly *is* more popular, we'll have to wait until about April of next year to see for sure..
But the original suggestion was to run only one cable in the wall and not put the hub/switch on the floor. With this solution, your core switch would stay just as small. Your suggestion has them running four cables again, and this is still inconsistant with the other ways in which the switch is evaluated in the review, against other 4-port switches.
Of course, the bottom line for the target customers (businesses), having a tidier appearance can be worth more than the money saved by using a standalone switch.
But Windows NT is *kinda* that deal. Granted it is not OpenBSD based, but the NT kinda rose from the ashes of VMS. Not linux compatible, but still a sort of *nix aimed directly at competitors.
I'm interested in running OpenBSD for my NAT gateway, though I'm left with a lingering issue..
Does OpenBSD include any support for decent irc connection tracking like what is available in iptables for linux? I have people behind the gateway that use DCC within IRC, and without good connection tracking, I'm not sure how to go about securely allowing one or more people to use IRC and have DCC work.
Everything else I plan on using this system for (software RAID, NIS+, samba PDC and fileserver, NFS) seems to be fine, but this one little nitpick of mine may keep me off of OpenBSD.
Also, how is the raid implementation as far as moving the array from one openbsd install to another, and is there any semblance of lvm there? The volume management stuff w/ resizable partitions would be nice, but by no means necessary..
Well, there is SSH, and if you want, you could have them use MindTerm, which has an attempt at a sFTP to local FTP proxy built in. Never got it to work, but it could be just the ticket to letting Windows users use their favorite FTP client to access sFTP.... :) Even if it were secure, the whole mechanism sucks, can't easily be port-forwarded, and uses two ports. Those ports just makes it that much easier for a sniffer to differentiate between transefered data, and command data (i.e. passwords in plaintext).
http://www.mindbright.se/mindterm/ is mindterms url, check it out.
On the other hand, putty has both pscp, and a psftp in development, that also provides this functionality.
Ftp is a really ugly and insecure protocol, it needs to die
Sure they could, if they could log your keystrokes, then they could log the data between the thumb/face reader and the computer. Then they can fake the input from the device, biometric approach defeated. The lesson here is that once they can into your system and interact with stuff like keyboard I/O, they can get pretty much any I/O your computer does, including biometric device I/O.
Interesting that slashdot made it to the white list. It may be that bad images may not make it through, but with some if the comments and ASCII art trolls, I'm surprised Disney was willing to risk it..
A lot of people are complaining they have to pay cash for WineX, not quite true. Only pre-packaged versions. The source is available under the terms of the Aladdin license. If this codebase is not complete, would someone please correct me. I personally maintain two different installs of Wine, the main branch of winehqs wine, checked out of CVS every so often, and the WineX branch, again from CVS ever so often. Why both? The WineX branch provides DirectX support, but other advances in the Wine project are slow to get into WineX, for example I can get the QuickTime player to install under standard Wine CVS builds, but not WineX. :)
So if you don't feel like paying, build the source if you can. Takes a long while, and if it fails, oh well, shell out a bit of cash for your software for once
Transgaming is a decent company, even if they don't bend over backwards to give away all their work for free. This is a bad expectation for Linux users to have, and when commercial Linux projects receive such criticism, it makes commercial vendors more and more hesitant about releasing linux products.
Well, truform I'll give, though in practice any "guessing" about what the game intended the geometry to be produces some problems (puffy characters come to mind). I think there are GeForce boards out there that can push dual monitors. As far as smoothvision goes, that is just a really hyped up supersampling AA solution, dog slow and even at highest quality doesn't do much better than nVidia w/ Antisotropic filtering, which is much faster. Now when they release the All-In-Wonder 8500DV, then you have some kick-ass features. That is what I'm holding out for, great 3D and Multimedia :)
That's good to hear, but has the gameplay changed significantly from just being able to basically slide left and right? The graphics in the screenshots still seem sub-standard to me.
All over the place, a board is marked as a clear winner, even if another board shows the *exact* same result. The archiving benchmark where MSI was top dog was interesting, what about the Epox which scored the exact same thing? Also, the other boards had less than 0.5% difference from the winner, I doubt that is statistically significant. How many runs were done? With the kernel compile, again, how many runs were done, in what order? The same config? I seriously doubt motherboards with the same chipset would have such a dramatic difference as 15 seconds on Kernel Compilation, given equivalent Disk I/O and CPU...
Did you actually fill all the banks? If so, did you encounter any problems? I'm shopping around and am looking for something that will be both decent performance and stable. I have heard all good things about the Shuttle board, but have also heard that the system can get flaky with all 4 banks filled due to memory timing issues.
Whether or not a chip is embedded has no bearing on how much CPU utilization is required for it to work. The video chipset on nForce boards, for example, is a GeForce2MX, a full-fledged chipset that does not heavily rely on the CPU. You can have dumb devices (i.e. WinModems) on PCI cards too. The complaint I have against embedded components is when they constrict upgrade paths (i.e. some motherboards with integrated graphics opt not to have an AGP slot). Additionally, I rarely find the provided integrated components sufficient, and don't want to pay the additional cost for something I plan on replacing immediately. Integrated is not the way to go for enthusiasts, typically, in the same way component stereo systems that can be mixed and matched are better for audiophiles than, say a bookshelf stereo.
I may be mistaken, but I have heard that most of these 'so-called' IDE raid controllers that motherboard manufacturer's put on their board are little more than typical IDE controllers. The manufacturer's get away with calling them RAID because they implement the RAID functionality in the Windows drivers. If this is true, I could care less about the inclusion of RAID controllers apart from having that many more ports on the MB. The only advantage is that you get a software RAID implementation for Win9x/ME, but under Linux, *BSD, and Win2k/XP there are provided facilities for doing software raid on arbitrary block devices. Additionally, I've heard the OS implementations are often both more efficient and more reliable than the drivers for these cheap IDE RAID controllers.
Also, they mention the Shuttle MB having 4 DIMM slots, and not going far beyond that. Does anyone have experience with having all 4 banks in use? I have heard theat the Shuttle MB has problems with memory timings when all 4 are in use. Anyone want to enlighten me?
I remember downloading it and thinking it was kinda neat. An OpenGL game written with linux in mind, and it's somewhat entertaining. While I think it above many other GPL games in many respects, I don't think it's so great as to be worth anything money wise when you look at the market out there.
The gameplay is pretty straightforward and boring. Doesn't even seem as interesting as, say, the snowboarding mini-game in Final Fantasy VII, which wasn't even meant to stand on its own. Snowboarding games and the like typically offer a significant deal more than TuxRacer, and for this reason if I was going to shell out cash for a game of this genre, I'd go with a good game.
For another thing, the graphics are not that spectacular. The scenery has some nice textures, but the characters and objects are simple gouraud shaded polygons, and even then the polygon count in the player models and how they are put together is now substandard. I understand that having a low polygon count helps performance, but companies like Square show how you can really have some decent looking graphics without complex geometry.
All in all it was a neat little game that kept my attention for a few minutes when I first got it. It's not on the level of any commercial competition in my opinion, considering games from 97 have roughly equivalent graphics and the gameplay is really boring and repetitive, with next to nothing to spruce it up.
I was under the impression that the corruption could be corrected by an fsck, and that while the ext structure was invalid, the data was left intact, that the issue was more with lock files, so an fsck would restore a valid inode table, and the actual data should still be intact, did I misunderstand this? I didn't ever actually use the beast, so I could be wrong...
I know that part of the Object Desktop package has been doing this for a long time now. http://www.stardock.com/ The package in question, IIRC, is DesktopFX. Neat toy, but nothing really that useful or newsworthy, more useful than the 'fake' alpha blending most commonly seen in nix (only blend with the root window, ignore all others). For nearly real alpha blending, you could use KDE and use mosfet's (www.mosfet.org) liquid theme and at least have alpha blending on the menus. The reason I say almost real, is that while it does blend against windows as well as the background, it only blends against the screen as it was when it was first drawn, if background changes, the liquid alpha-blended menu does not. Xrender hints at the ability to do true alpha blending w/ hardware help, but I haven't seen it actually used for anything except AA-text.
While Alpha-blended windows give nifty screenshots and initial "ooh" factor, people switch it back off in moments because it really makes programs harder to use in the long run. It's hard enough to make sure colors within an application always have text that is readable against the background without other applications lending their colors to further mess things up.
Well not quite, because you could claim it was a backup copy of a disc that was either lost or stolen. At which point it should be your word against theirs. The "ink" example is different, it is a presence of something that should not be there. The lack of being able to present a hologram is much different.
Case in point, I had, among other things, a CD binder from my residence stolen. Most things were lost for good, except a few things I had copied, and I now copy *everything* I buy and store it seperately. If someone catches me with my backup (legal), why should I be punished because I can't present the original? Granted in this case I could reference the police report from the robbery, but there are legitimate reasons why people may not be able to present the pretty holograms to prove they purchased the software, and this would be unfair.
They likely have something in development. They have done a pretty good job of catchup, the Radeon was leaps and bounds better than the Rage series, but still nowhere near equivalent GeForce performance.
Now the 8500 chips are showing themselves to be, with decent drivers, at least roughly equivalent to the high-end GeForce3 chips. Though perhaps not quite as feature-rich, performance is quite competitive with nVidia offerings. If they keep up this pace, they may very well not only catch up with nVidia both performance and feature wise, but also pass them.
In a way, the situation right now seems to me reminiscent of nVidia's situation when they brought out the RivaTNT chips. They had been playing catchup to 3dfx, and the Riva128 series would be analagous to ATI's Radeon chips, and the first TNT analogous to the 8500 chips. Then people thought nVidia's progress was impressive, but thought they probably would never quite catch up with 3Dfx, and we all know how that went..
Basically, I would say that since the release of the first GeForce chip that nVidia development has slowed down and not made nearly the significant strides seen between the Riva, TNT, and that ended with the GeForce, since the market didn't pressure them enough, and ATI is taking advantage and may take the crown in the somewhat-near future....
Note he was talking about IE, which has a distinct lack of any MDI mode. But on the subject, I like being able to middle click and have a document load in a background tab. I can't display two pages at the same time in MDI, have to do SDI then, but that is a much rarer case. If Opera can do pop-under MDI, and have a "maximize by default" so it resembles a tab interface, I'd be much more interested. Having the option of windowed MDI is nice, but the tab paradigm has advantages..
Of course the concept that those holograms and such could be legally used to prosecute piracy is scary to me. Why? Because this means that when asked, you must *prove* that you did not pirate the software, else, you are considered guilty. Completely discards the "innocent until proven guilty". Of course, software-piracy end-users could, under the strict definition of the concept, almost never be proven guilty, but the re-distibrutors could be..
No, pro has it too, it's just that the 'corporate' version does not, which is just a copy of pro legally cracked by MS :) Well, maybe crack's the wrong word, but the mere existence of this corp edition must be really appreciated by the warez scene, while the legal users just get annoyed and hassled more.
When I first tried out tabbed interface MDI in a web browser, instantly recognized how much better it was. Not like you're forced into it, by default it will act non-MDI, just like IE. With this, you save screen space, it's much easier to keep organized. Only down side is when you decide you need information from two tabs at once. Hopefully you recognize this well in advance and used windows, but still, it would be nice to instantly "detach" a tab into a new window, and attach again once you finish, preserving all the state info, but still, galeon's my favorite,. Mozilla has tabs, but the interface to them still sucks..