The "advanced methods" available in GNOME don't even compare to those in KDE. For example, there is no gconf key for putting the menu on top, so you can use regular panel hide methods with it.
That's Subversion 0.8, not Subversion 0.8x. The 'x' implies a series of ten. So Subversion 0.8x means the eighties versions: 0.81, 0.82, 0.83, 0.84, etc. Subversion only ever got into the thirties: 0.31, 0.32, 0.33...
Oh god. Cut the melodrama. Corporations and the government are not the ones to blame. It is the people. We are a free country. We elect our leaders, and nobody can claim that these elections are somehow fraudulent.
We the people put these guys into power, and we can choose to put them out of power. We don't, because we don't want to. We like the status quo. Our Constitution gives corporations the power to sell us what we want to buy, nothing more. The fact that we buy it is our fault, and ours alone.
Personally, I blame populism. The founding fathers were on to something --- not everyone should be able to vote. Today, money is no longer closely associated with education, so the standards for deciding who is fit to vote would probably need to change. In either case, a dictatorship of the meritorious is preferable to a dictatorship of the masses.
I think the point is not so much that there is an equivilency between what the US did in Iraq and what Germany did so much as there is a similarity. In both cases, soldiers were not fighting to protect their country, but to invade a foreign one. In both cases, they killed civilians in the process.
And apologizing for the comparison would be asinine. When someone compares to things it does not mean they are identical. It means that they share some characteristics. Unless the set of attributes of American soldiers is completely disjoint from the set of attributes of Nazi soldiers, then there *are* valid comparisons to be made between the two.
The *last* thing in the US that we need right now is something good for nationalism.
We don't need another wave of morons feeling superior to everyone else just because some very great people live (and lived) on the same 3.5 million square mile land mass as they do.
My knowledge of the Civil War is pretty limited, so Gettysburg was probably a bad example. However, I do know quite a bit about the European wars, and my comment about the Seven Year's War being pretty bloodless is true.
Youth in the US also consider the American genocide of the Native Americans to be "merely" historic. And rightly so --- what their ancestors did is not their fault.
Now the scary thing is that modern Americans still believe a lot of the "God blessed America" nonesense that allowed them to conceience displacing an entire race of people.
Actually, technology has increased the civilian and military costs of war greatly. Compare the relative civility of the Seven Years War in Europe to the outright carnage of Vietnam. Before machine guns, entire squadrons of troops wouldn't get mowed down at once. Before air-dropped bombs, cities weren't devestated in preparation for a ground-war. In the first Iraq war, hundreds of thousands of non-combatents were killed as a result of the infrastructure damage caused by US bombs. In the civil war, civilian casualties were much more limited --- for example, only one civilian died at Getteysburg.
Of course, things are swinging back a bit now. Better sattelite technology has made air-strikes more accurate. However, we are much closer to the levels of carnage we had during WWII than what we had during the Revolutionary War.
PS> I'd like to see those statistics. War is ravaging the world right now. Europeans, who are used to calling a century with only half a dozen wars a "peaceful" one might not think so, but relative to the long-term peaces we've had before in other parts of the world, today is by no means peaceful.
There's a great line in Tannenbaum's "Modern Operating Systems" that goes something like:
Of course RAID won't protect your data if the earth opens up and the computer falls 100 meters into a fissure. Its kind of hard to recover from that condition in software.
What's the point of doing things both ways? Basically, you need a generalized composite system to be useful. Now, if you can do anti-aliasing on top of that generalized composite system, why bother adding more stuff into the protocol specifically for AA text?
Well, a graphics system can't really make that kind of assumption. There are legitimate reasons *not* to want text to be anti-aliased. For example, anti-aliasing on many CRTs often results in headache-inducing fuzzy output compared to well-hinted monochrone rendering.
Also, its not just lines and text that the server is supposed to draw. What about transparent PNGs (or custom pixmaps)? People want to composit those too. What about using different anti-aliasing algorithms? For example, the anti-grain-geometry folks have a gamma-correct AA algorithm that looks a lot nicer than the one hard-coded into the Windows GDI.
Unless you want to start making a whole lot of assumptions about what your apps will be doing, and build a whole lot of policy into the server, you can see why it might be a better idea to just have a generalized compositing mechanism and build everything on top of that.
Well, its not really a limitation of X that it doesn't do that. The freedesktop.org X server already does that. I had KDE running with it the other day, and aside from issues with unoptimized resize code (the fd.o xserver is pre-alpha as of yet) it worked just fine.
Also, I think Mark Thomas is really streching say that this is a "huge bottleneck" for performance with X now. On my KDE desktop (2GHz P4), the only time I can actually see the redraw is occasionally on the window titlebar, and on HTML buttons in Konqueror. Other than that, moves and even resizes are butter-smooth. I realize that low-end machines might not redraw fast enough where the user can't notice it, but then again, those machines aren't also like to have the 100MB+ memory to spare just to keep window buffers loaded. I also realize that not all apps handle X11 Expose events as well as Qt-based apps do, and apps like OpenOffice redraw slowly enough where the user does notice visual artifacts in this situation.
That said, the freedesktop.org X server (which will also feature OpenGL acceleration) does use double-buffering. The feature will probably result in a net loss of performance (because there is now more overhead to manage those huge window buffers, less memory is available for apps, and the window buffers must be composited). However, GUI speed is all about perception, and double-buffering does make the GUI *feel* faster (well, more solid, anyway) by eliminating artifacting and flicker during resizes.
This is precisely the reason why X developers do not like the idea of a server-side toolkit. Server-side toolkits can help performance in some cases, but mostly in cases where performance is fine under X already. Where it doesn't help performance (unless you allow the download of arbitrary code) is for those widgets that actually need the performance, such as HTML renderers, word processors, vector graphics layout engines, etc.
The Windows UI is *not* faster than X11 + any arbitrary window manager. The Windows UI is faster than X11 + *some* window managers. For example, the Windows UI is definitely faster than GNOME on my machine. However, with KDE 3.2 (plus the aformentioned patches to fix some synchronization issues) KDE is about as fast as XP + Luna. Sometimes faster, sometimes slower. Under heavy load, Linux 2.6.x + XFree + KDE handles much better than XP. On my 2GHz P4, the only way I can tell a compile is finished is when the fans on my laptop stop screaming at me. Under XP, I can *feel* a compile finish.
You hear a lot of people bitch, but mostly because they do not understand it. People tend to do that. Hell, I did that myself, before people beat some sense into me and made me actually understand the system I was bashing.
X does have some problems. The set of actual problems with X, and the set of perceived problems with X are more or less disjoint.
X is not slow. X is not a memory hog. X's feature set is not outdated. It does, however, have a very weird way of handling color, has some protocol peculiarities, and the default interface library (xlib) makes it hard to write fast apps. Most of these things can be fixed without ditching X. For example, there is an extension called XFixes which fixes certain problems with the protocol. There is a new interface library called XCB which is more "low-level" and makes it easier for toolkit authors to identify potential performance issues. Freedesktop.org's new X server will incorporate compositing and OpenGL-acceleration, to make X competitive with Longhorn, all within the existing X framework!
Actually, building AA into the client isn't as hacky as you'd think. What Xft does is implement a generalized compositing mechanism in the X server. One of the many uses of this compositing mechanism is AA'ed text. Ergo, keeping the AA mechanism on the client is actually reasonable, because other uses of the compositing mechanism will also be in the client.
New users find the selection of different toolkits for X confusing and inconsistent both in appearance and behaviour. One standard toolkit will help with newbie usability greatly - though whether it will stand the test of time remains to be seen. Windows seems to be doing just fine with it's standards though, so I rather suspect the same will apply to Y. ----------- Bullshit! Complete and utter bullshit! If you think Windows has one toolkit, you obviously do not use Windows. On my Windows desktop at work, I have 3 toolkits loaded into memory at any given time:
- Windows XP common toolkit (MSIE, etc). - Microsoft Office toolkit - Microsoft.NET toolkit (Visual Studio)
If I start up media player, Trillian, Winamp, AIM, or MSN messenger, I get still more apps that, while they may use the same toolkit, look and act nothing like each other! Even MacOS X has three toolkits (Cocoa, Carbon, Classic) and two UI styles (Aqua, Metal).
The key is getting applications to look similar and behave similarly. Efforts like freedesktop.org are making fine progress on that front, *without* building the toolkit into the server.
Slashdot Mirror
The "advanced methods" available in GNOME don't even compare to those in KDE. For example, there is no gconf key for putting the menu on top, so you can use regular panel hide methods with it.
I knew it sounded like a porn movie...
Try taking a look at Clean's IO library to see how they handle a GUI in a purely functional language :)
But I thought the communist government was officialy athiest?
That's Subversion 0.8, not Subversion 0.8x. The 'x' implies a series of ten. So Subversion 0.8x means the eighties versions: 0.81, 0.82, 0.83, 0.84, etc. Subversion only ever got into the thirties: 0.31, 0.32, 0.33...
Ha ha. There never was a subversion 0.8x. The last release before the 1.0 release-candidate was 0.37 :)
It won't. Subversion is a traditional, centralized system. Linus wants a distributed system, which Bitkeeper is.
Oh god. Cut the melodrama. Corporations and the government are not the ones to blame. It is the people. We are a free country. We elect our leaders, and nobody can claim that these elections are somehow fraudulent.
We the people put these guys into power, and we can choose to put them out of power. We don't, because we don't want to. We like the status quo. Our Constitution gives corporations the power to sell us what we want to buy, nothing more. The fact that we buy it is our fault, and ours alone.
Personally, I blame populism. The founding fathers were on to something --- not everyone should be able to vote. Today, money is no longer closely associated with education, so the standards for deciding who is fit to vote would probably need to change. In either case, a dictatorship of the meritorious is preferable to a dictatorship of the masses.
I think the point is not so much that there is an equivilency between what the US did in Iraq and what Germany did so much as there is a similarity. In both cases, soldiers were not fighting to protect their country, but to invade a foreign one. In both cases, they killed civilians in the process.
And apologizing for the comparison would be asinine. When someone compares to things it does not mean they are identical. It means that they share some characteristics. Unless the set of attributes of American soldiers is completely disjoint from the set of attributes of Nazi soldiers, then there *are* valid comparisons to be made between the two.
Great post!
The *last* thing in the US that we need right now is something good for nationalism.
We don't need another wave of morons feeling superior to everyone else just because some very great people live (and lived) on the same 3.5 million square mile land mass as they do.
For a supposedly atheist country?
My knowledge of the Civil War is pretty limited, so Gettysburg was probably a bad example. However, I do know quite a bit about the European wars, and my comment about the Seven Year's War being pretty bloodless is true.
Youth in the US also consider the American genocide of the Native Americans to be "merely" historic. And rightly so --- what their ancestors did is not their fault.
Now the scary thing is that modern Americans still believe a lot of the "God blessed America" nonesense that allowed them to conceience displacing an entire race of people.
Actually, technology has increased the civilian and military costs of war greatly. Compare the relative civility of the Seven Years War in Europe to the outright carnage of Vietnam. Before machine guns, entire squadrons of troops wouldn't get mowed down at once. Before air-dropped bombs, cities weren't devestated in preparation for a ground-war. In the first Iraq war, hundreds of thousands of non-combatents were killed as a result of the infrastructure damage caused by US bombs. In the civil war, civilian casualties were much more limited --- for example, only one civilian died at Getteysburg.
Of course, things are swinging back a bit now. Better sattelite technology has made air-strikes more accurate. However, we are much closer to the levels of carnage we had during WWII than what we had during the Revolutionary War.
PS> I'd like to see those statistics. War is ravaging the world right now. Europeans, who are used to calling a century with only half a dozen wars a "peaceful" one might not think so, but relative to the long-term peaces we've had before in other parts of the world, today is by no means peaceful.
There's a great line in Tannenbaum's "Modern Operating Systems" that goes something like:
Of course RAID won't protect your data if the earth opens up and the computer falls 100 meters into a fissure. Its kind of hard to recover from that condition in software.
What's the point of doing things both ways? Basically, you need a generalized composite system to be useful. Now, if you can do anti-aliasing on top of that generalized composite system, why bother adding more stuff into the protocol specifically for AA text?
Well, a graphics system can't really make that kind of assumption. There are legitimate reasons *not* to want text to be anti-aliased. For example, anti-aliasing on many CRTs often results in headache-inducing fuzzy output compared to well-hinted monochrone rendering.
Also, its not just lines and text that the server is supposed to draw. What about transparent PNGs (or custom pixmaps)? People want to composit those too. What about using different anti-aliasing algorithms? For example, the anti-grain-geometry folks have a gamma-correct AA algorithm that looks a lot nicer than the one hard-coded into the Windows GDI.
Unless you want to start making a whole lot of assumptions about what your apps will be doing, and build a whole lot of policy into the server, you can see why it might be a better idea to just have a generalized compositing mechanism and build everything on top of that.
Well, its not really a limitation of X that it doesn't do that. The freedesktop.org X server already does that. I had KDE running with it the other day, and aside from issues with unoptimized resize code (the fd.o xserver is pre-alpha as of yet) it worked just fine.
Also, I think Mark Thomas is really streching say that this is a "huge bottleneck" for performance with X now. On my KDE desktop (2GHz P4), the only time I can actually see the redraw is occasionally on the window titlebar, and on HTML buttons in Konqueror. Other than that, moves and even resizes are butter-smooth. I realize that low-end machines might not redraw fast enough where the user can't notice it, but then again, those machines aren't also like to have the 100MB+ memory to spare just to keep window buffers loaded. I also realize that not all apps handle X11 Expose events as well as Qt-based apps do, and apps like OpenOffice redraw slowly enough where the user does notice visual artifacts in this situation.
That said, the freedesktop.org X server (which will also feature OpenGL acceleration) does use double-buffering. The feature will probably result in a net loss of performance (because there is now more overhead to manage those huge window buffers, less memory is available for apps, and the window buffers must be composited). However, GUI speed is all about perception, and double-buffering does make the GUI *feel* faster (well, more solid, anyway) by eliminating artifacting and flicker during resizes.
This is precisely the reason why X developers do not like the idea of a server-side toolkit. Server-side toolkits can help performance in some cases, but mostly in cases where performance is fine under X already. Where it doesn't help performance (unless you allow the download of arbitrary code) is for those widgets that actually need the performance, such as HTML renderers, word processors, vector graphics layout engines, etc.
The Windows UI is *not* faster than X11 + any arbitrary window manager. The Windows UI is faster than X11 + *some* window managers. For example, the Windows UI is definitely faster than GNOME on my machine. However, with KDE 3.2 (plus the aformentioned patches to fix some synchronization issues) KDE is about as fast as XP + Luna. Sometimes faster, sometimes slower. Under heavy load, Linux 2.6.x + XFree + KDE handles much better than XP. On my 2GHz P4, the only way I can tell a compile is finished is when the fans on my laptop stop screaming at me. Under XP, I can *feel* a compile finish.
You hear a lot of people bitch, but mostly because they do not understand it. People tend to do that. Hell, I did that myself, before people beat some sense into me and made me actually understand the system I was bashing.
X does have some problems. The set of actual problems with X, and the set of perceived problems with X are more or less disjoint.
X is not slow. X is not a memory hog. X's feature set is not outdated. It does, however, have a very weird way of handling color, has some protocol peculiarities, and the default interface library (xlib) makes it hard to write fast apps. Most of these things can be fixed without ditching X. For example, there is an extension called XFixes which fixes certain problems with the protocol. There is a new interface library called XCB which is more "low-level" and makes it easier for toolkit authors to identify potential performance issues. Freedesktop.org's new X server will incorporate compositing and OpenGL-acceleration, to make X competitive with Longhorn, all within the existing X framework!
Actually, building AA into the client isn't as hacky as you'd think. What Xft does is implement a generalized compositing mechanism in the X server. One of the many uses of this compositing mechanism is AA'ed text. Ergo, keeping the AA mechanism on the client is actually reasonable, because other uses of the compositing mechanism will also be in the client.
Okay, compare Windows 2000 to Fluxbox. Guess which is faster? Like I said: not a fair comparison!
New users find the selection of different toolkits for X confusing and inconsistent both in appearance and behaviour. One standard toolkit will help with newbie usability greatly - though whether it will stand the test of time remains to be seen. Windows seems to be doing just fine with it's standards though, so I rather suspect the same will apply to Y.
.NET toolkit (Visual Studio)
-----------
Bullshit! Complete and utter bullshit! If you think Windows has one toolkit, you obviously do not use Windows. On my Windows desktop at work, I have 3 toolkits loaded into memory at any given time:
- Windows XP common toolkit (MSIE, etc).
- Microsoft Office toolkit
- Microsoft
If I start up media player, Trillian, Winamp, AIM, or MSN messenger, I get still more apps that, while they may use the same toolkit, look and act nothing like each other! Even MacOS X has three toolkits (Cocoa, Carbon, Classic) and two UI styles (Aqua, Metal).
The key is getting applications to look similar and behave similarly. Efforts like freedesktop.org are making fine progress on that front, *without* building the toolkit into the server.