It's a fair bet your program would work on Windows 2000 and Windows Vista. Yet Windows Vista will "tax" your system more to achieve exactly the same result. This is my point - the operating system is gobbling more and more resources that should be used by your applications without giving the user anything in return. In this sense, we are moving backwards.
A few things the OS will do that it didn't used to (I know nothing about vista, I've been a linux guy since winme):
Window transparency - when used properly, this can give a huge increase in productivity, since it's pretty much the same effect as adding another monitor.
Audio mixing in software - again, very useful if used properly
More effective search functions
Plus, of course, all the necessary things to support that fast hardware properly - faster timers, the address space gymnastics necessary to use large amounts of RAM on a processor architecture not really designed for it, filesystems that can handle the larger-but-relatively-slower disks we use today. Ultimately there's not that much for the OS to do, and it certainly shouldn't be grabbing all the processor cycles, but OSes *have* improved.
As I said before, to simplify the licensing of linux.
It seems to me that the license is changing EXCLUSIVELY in the hope that some future developer will be too lazy or ignorant to check the origin of the code, and his changes will then go under the GPL because he simply didn't know they could go under the BSD license.
Anyone who doesn't realise that would know nothing about licensing anyway. One of two things will be happening: a) The linux tree will simply be a copy of the BSD tree, in which case any patchers will be redirected to the BSD folks, who will tell them it needs to be BSD licensed, or b) the linux tree will be maintained separately, in which case any patches that the BSD folks want, they can ask the authors for. If it was merely a case of not realising, the author will be happy to relicense under BSD.
I think that sucks, and if you don't see why it sucks, you're an idiot.
Ah yes, the crux of internet debate. Go on, why does it suck, in practice?
When you combine multiple licenses, you HAVE to abide by all the licenses, so it becomes the most restrictive of the combination.
Umm, no. The whole point of dual-licensing is you only have to abide by *one* of the licenses.
That's what Theo is miffed about. And I can understand it entirely. I think the GPL crowd would be miffed if, due to the dual licensing, the BSD crowd took it as ok to remove GPL attribution and put it somewhere as purely BSD code.
No they wouldn't, and I suspect that happens all the time with e.g. BSD code incorporated into OSX or other proprietary software.
For an older, well-supported one that hasn't yet shown cracks the way MD5 and SHA1 have, try RIPEMD160. If your system supports more modern ones, Whirlpool is looking good, though it isn't so mature yet.
It's a bit like a good street artist contributing something to society to listen to and enjoy, with a friendly request to donate something if they like it.
Theo would do well to listen to the "friendly" part. Street artists with his attitude would be looking at a pretty empty hat.
BSD code can be modified but the BSD license cannot be removed from the modified file either.
Unless you have the permission of the copyright holder. Such as them giving you the software under another license which allows you to modify it, like, say, the GPL.
1) how this cause stands up when something like Nortan AV "accidentally" gets blocked
They'd have a hard time claiming that was done in "good faith".
2)IANAL, but shouldn't this cover DRM? It falls under otherwise objectionable at the very least (filthy too, IMHO)
Does it come under "material" though? It's not as if the DRM is in a few bits on the end you can knock off, you have to extract the data out of the DRMed format it's in.
Because changes from the Linux team can't go into the BSD tree anymore (GPL license is incompatible). So the Linux team can't actually change the code in their tree, or they'll have to do a full integration when the BSD team makes a change.
But that'd be the case anyway; OpenBSD people don't sync their code from external sources. If there was a really important change they could submit a patch to the BSD folks...just like they can now.
How will it make it any harder to manage the code? Any changes will need to be imported from BSD anyway, since it's highly unlikely that the BSD people have write access to the linux tree.
There shouldn't be a problem, but of course no such country exists, and if it did the MAFIAA would bribe enough congressmen to send a few gunboats their way until things changed.
So you are telling me that if I have 4Gb and no program is running the system should be using only 500 mb? No way... if the computer is idle, the OS better use the rest for, I don't know, indexing, caching, compacting, optimizing, or whatever. Only USE IT and give it back when needed.
But when one needs RAM one tends to need it now. If the OS has grabbed all the RAM and when I want to use it I have to wait for a few gigs to be written out to disk, this is bad. When idle, the main priority of the OS should be that the system is ready to go as quickly as possible for when I want to actually use it.
This new memory management was introduced for Vista and it was about freaking time somebody though about this....
I'm sure it's been thought of before; it's a bad thing to do for the reason above. In fact some OSes e.g. VMS would actually do the opposite - if it's idle, there is an option to pre-emptively swap out the running processes, because unlike network bandwidth, RAM is something that takes a long time to switch between applications. So if it's likely that the next thing the OS will have to do is load a new process, the optimal thing for the OS to do is make sure there's enough free space for the process to go into.
You are right in that it would be possible to find a rotating frame of reference in which the total angular momentum of the universe is zero, but a rotating frame of reference is not as valid as a non-rotating one; and you can tell, because in a rotating one you observe the centrifugal and corrolis pseudoforces. So if when we take this particular frame the universe as a whole is rotating (and we can tell that by measuring the centrifugal "force", without needing a reference point outside the universe), or equivalently in a non-rotating frame we observe that the angular momentum of all the objects in the universe adds up to something other than zero, then the universe has a non-zero angular momentum.
Now that the driver is under GPL, enhancements made to the BSD-licensed version don't automatically propagate to it; they have to be manually incorporated
Erm, what? The GPL version could be automatically synced against the BSD one, the BSD license allows it.
Then it must have had some nondeterministic means of choosing the server, because I've been at LAN parties where every one of us bombed out at least once (two of those times being the player who had initially "hosted") in a series of games between the same machines. At no point did the loss of any machine cause the game to stop; it just carried on with the remaining players.
As far as there being a computer powerful enough to do so, yes. Most Core 2 duos and Core duos should be capable of 10 player (minus the problem of there only being support for 8-players...), at realtime speeds.
But presumably if one were serving it it would have to devote extra resources to communicating with all the others, compensating for lag, etc., no?
No, it really wouldn't be. If it was an actually good post which had been modded into oblivion by the GNAA, or worse, mac fanboys, who would see it to mod it up again?
This is great and all, but the way Microsoft and GPG used it, it has to be peer to peer. And each computer runs the sim. Which would be fine, if it weren't one of the most taxing games on a cpu currently existing. This would be fine in a homogeneous environment, such as the Xbox. However, PCs aren't. So if one person has a crappy computer, it will slow EVERYONE down.
I suspect that has a lot more to do with the developer than MS pressure, since the same is true of Total Annihilation from 10 years ago, and you'll thank them for it in a few years if my experience with TA is any guide; besides, if extra load was placed on one of the machines, would there be a computer anywhere that was powerful enough to host a 10-player game?
the way one frequently sees artificial "focus" in animé. And given the Ws' known fondness for the stuff, could this be where the inspiration came from?
From the article it sounds like each seat actually has a thin client, which would in effect reinstall the OS after each user/flight which is good from a security standpoint. But with access to a keyboard and USB hub, it still sounds a bit more vulnerable to abuse than a standard kiosk.
IME such kiosks are likely to be *more* vulnerable, since people tend to forget that they're computers. And such a thin client is as close to invulnerable as it gets; without hacking the server or opening up the box, the only possible thing I can think to do to them is a hardware keylogger, and it's hard to make something with a keyboard that isn't vulnerable to that.
No, the program I'm talking about was not made by borland; I can't promise anything from wine with your particular program, but it's probably worth a try (using winecfg to tell wine to emulate win95 rather than the default of winxp may help), depending of course on how much effort it would take you to use wine; on the speed front wine is more than capable of near-native (certainly better than 50%) speeds even with graphics-heavy programs.
Slashdot Mirror
Get help, now. Hint: people die in one of them.
A few things the OS will do that it didn't used to (I know nothing about vista, I've been a linux guy since winme):
Window transparency - when used properly, this can give a huge increase in productivity, since it's pretty much the same effect as adding another monitor.
Audio mixing in software - again, very useful if used properly
More effective search functions
Plus, of course, all the necessary things to support that fast hardware properly - faster timers, the address space gymnastics necessary to use large amounts of RAM on a processor architecture not really designed for it, filesystems that can handle the larger-but-relatively-slower disks we use today. Ultimately there's not that much for the OS to do, and it certainly shouldn't be grabbing all the processor cycles, but OSes *have* improved.
As I said before, to simplify the licensing of linux.
It seems to me that the license is changing EXCLUSIVELY in the hope that some future developer will be too lazy or ignorant to check the origin of the code, and his changes will then go under the GPL because he simply didn't know they could go under the BSD license.
Anyone who doesn't realise that would know nothing about licensing anyway. One of two things will be happening: a) The linux tree will simply be a copy of the BSD tree, in which case any patchers will be redirected to the BSD folks, who will tell them it needs to be BSD licensed, or b) the linux tree will be maintained separately, in which case any patches that the BSD folks want, they can ask the authors for. If it was merely a case of not realising, the author will be happy to relicense under BSD.
I think that sucks, and if you don't see why it sucks, you're an idiot.
Ah yes, the crux of internet debate. Go on, why does it suck, in practice?
Umm, no. The whole point of dual-licensing is you only have to abide by *one* of the licenses.
That's what Theo is miffed about. And I can understand it entirely. I think the GPL crowd would be miffed if, due to the dual licensing, the BSD crowd took it as ok to remove GPL attribution and put it somewhere as purely BSD code.
No they wouldn't, and I suspect that happens all the time with e.g. BSD code incorporated into OSX or other proprietary software.
For an older, well-supported one that hasn't yet shown cracks the way MD5 and SHA1 have, try RIPEMD160. If your system supports more modern ones, Whirlpool is looking good, though it isn't so mature yet.
Theo would do well to listen to the "friendly" part. Street artists with his attitude would be looking at a pretty empty hat.
Unless you have the permission of the copyright holder. Such as them giving you the software under another license which allows you to modify it, like, say, the GPL.
They'd have a hard time claiming that was done in "good faith".
2)IANAL, but shouldn't this cover DRM? It falls under otherwise objectionable at the very least (filthy too, IMHO)
Does it come under "material" though? It's not as if the DRM is in a few bits on the end you can knock off, you have to extract the data out of the DRMed format it's in.
But that'd be the case anyway; OpenBSD people don't sync their code from external sources. If there was a really important change they could submit a patch to the BSD folks...just like they can now.
How will it make it any harder to manage the code? Any changes will need to be imported from BSD anyway, since it's highly unlikely that the BSD people have write access to the linux tree.
Why relicense it?
Answer: so that it's under the same license as the rest of the kernel, which makes managing linux licensing easier.
There shouldn't be a problem, but of course no such country exists, and if it did the MAFIAA would bribe enough congressmen to send a few gunboats their way until things changed.
But when one needs RAM one tends to need it now. If the OS has grabbed all the RAM and when I want to use it I have to wait for a few gigs to be written out to disk, this is bad. When idle, the main priority of the OS should be that the system is ready to go as quickly as possible for when I want to actually use it.
This new memory management was introduced for Vista and it was about freaking time somebody though about this....
I'm sure it's been thought of before; it's a bad thing to do for the reason above. In fact some OSes e.g. VMS would actually do the opposite - if it's idle, there is an option to pre-emptively swap out the running processes, because unlike network bandwidth, RAM is something that takes a long time to switch between applications. So if it's likely that the next thing the OS will have to do is load a new process, the optimal thing for the OS to do is make sure there's enough free space for the process to go into.
You are right in that it would be possible to find a rotating frame of reference in which the total angular momentum of the universe is zero, but a rotating frame of reference is not as valid as a non-rotating one; and you can tell, because in a rotating one you observe the centrifugal and corrolis pseudoforces. So if when we take this particular frame the universe as a whole is rotating (and we can tell that by measuring the centrifugal "force", without needing a reference point outside the universe), or equivalently in a non-rotating frame we observe that the angular momentum of all the objects in the universe adds up to something other than zero, then the universe has a non-zero angular momentum.
I hope you will take the equal morality displayed by these creations as evidence for the nonexistence of the soul.
Erm, what? The GPL version could be automatically synced against the BSD one, the BSD license allows it.
Isn't that the whole point of the BSD license, that anyone can redistribute it under (pretty much) any license they like?
Then it must have had some nondeterministic means of choosing the server, because I've been at LAN parties where every one of us bombed out at least once (two of those times being the player who had initially "hosted") in a series of games between the same machines. At no point did the loss of any machine cause the game to stop; it just carried on with the remaining players.
As far as there being a computer powerful enough to do so, yes. Most Core 2 duos and Core duos should be capable of 10 player (minus the problem of there only being support for 8-players...), at realtime speeds.
But presumably if one were serving it it would have to devote extra resources to communicating with all the others, compensating for lag, etc., no?
No, it really wouldn't be. If it was an actually good post which had been modded into oblivion by the GNAA, or worse, mac fanboys, who would see it to mod it up again?
I suspect that has a lot more to do with the developer than MS pressure, since the same is true of Total Annihilation from 10 years ago, and you'll thank them for it in a few years if my experience with TA is any guide; besides, if extra load was placed on one of the machines, would there be a computer anywhere that was powerful enough to host a 10-player game?
the way one frequently sees artificial "focus" in animé. And given the Ws' known fondness for the stuff, could this be where the inspiration came from?
IME such kiosks are likely to be *more* vulnerable, since people tend to forget that they're computers. And such a thin client is as close to invulnerable as it gets; without hacking the server or opening up the box, the only possible thing I can think to do to them is a hardware keylogger, and it's hard to make something with a keyboard that isn't vulnerable to that.
Erm, what?
Shrug, guess it's no good then; sorry.
No, the program I'm talking about was not made by borland; I can't promise anything from wine with your particular program, but it's probably worth a try (using winecfg to tell wine to emulate win95 rather than the default of winxp may help), depending of course on how much effort it would take you to use wine; on the speed front wine is more than capable of near-native (certainly better than 50%) speeds even with graphics-heavy programs.