Having written a few Qt/KDE themes, and perused in depth dozens more, I have to admit that Qt is mostly the same way.
But to be honest, the reason is to prevent bloat. Caching pixmaps uses a lot of memory. For a checkbox, there would be six pixmaps cached (foreground and background palettes for active, inactive and disabled states). This lack of caching doesn't cause a problem for simple controls like checkboxes, because redraws are uncommon.
On some themes, I can definitely see a flicker on slower machines, while on others that appear to be more complex, I see none. For example, Liquid is a pretty complex theme with blended gradients everywhere, but it makes extensive use of caching, so it has a smoother redraw than some simpler themes.
Of course, if this becomes a problem for non-cached themes, it's simple to fix, because of the way Qt is architected.
But all the kiddies are screaming for fully alpha blended vector graphics rendered on the fly by the GPU. Since they scream louder than the rest of us, they're they one's that get heard.
I've come to the conclusion that GNU-Darwin is a hoax! Oh, there may be some software there you can download, but it's all part of the trappings for an incredibly elaborate chicanery. I am here now to unmask the perpetrator of this hornswoggle. This has gone on long enough!
Proclus (otherwise known as Michael L. Love, of no relation to Ransom that I know of) started this scheme in an effort to illustrate absurdity by being absurd. His aim is to ridicule Free Software by taking the ideas of the FSF to a absurd extreme. His ultimate goal, I fear, is to completely discredit GNU. Look at the evidence! All of it points to either a madman or a cunning conniver.
GNU-Darwin has been an ardent defender of digital liberties, and it is a platform for digital activism.
Here Proclus is deliberately asserting that GNU-Darwin is a propaganda tool. This isn't a platform to write documents, edit videos or administrate servers, "it is a platform for digital activism." The sole purpose of GNU-Darwin is to spread an ideology. An ideology that Proclus has subverted towards his own diabolical ends.
The tools of such action include, but are not limited to; boycotts, blockades, community response, de-branding, labor solidarity, whistle-blowing, etc.
Here we see how he out-stallmans Stallman. While RMS may have some firm and inflexible opinions on everything from Bush to the local dogcatcher, he keeps all of his non-software ideology out of GNU, and on a separate personal website. The point of GNU is Free Software and Free Software only. The point of GNU-Darwin is everything imaginable. It's not about the software, it's about "labor solitarity." Huh? Is Proclus going to take down his site as a sympathetic protest the next time the AFL/CIO goes on strike?
Once your credibility to engage is established, then the mere threat of such action is often sufficient to produce results.
Fortunately, GNU-Darwin has no credibility. They can boycott, blockade and de-brand all they want, and no one will care. Remember, this is all an elaborate and intricate hoax! The aim was never to get credibility, but to discredit GNU.
It is important to realize that the software freedom status of GNU-Darwin was tenuous before the change to the APSL, so we were ready to cut our losses, and remove GNU-Darwin from the internet, if necessary.
This is a prime example of "kindergarten activism". Proclus threatens to hold his breath until he gets his cookie. But since Apple wasn't paying attention, he has to call attention to his self-serving breath-holding, hoping that we'll feel compassion for his almost-sacrifice. How noble!
In general the effects of such action are ambiguous by nature, because the rulers certainly have an interest in denying the influence of democratic power.
This is a classic example of radical pseudo-think. He's telling you in advance that he can't point to any concrete results of his political activism, and the reason he can't is because he's oppressed by the ruling class. This is the crux of his ideology. Upon this all his hopes ride. You either laugh at this point and go on with life, or you stretch your credulity too far and you fall victim to his rantings. Don't go there!
Our third year commenced with the initiation of the The Free Darwin campaign which was followed by the antiwar blackout in March.
With this antiwar blackout, he's obviously playing along with the Free Software protests against European patents. But he went too far this time. Protests against "europatents" were justified because they were inextricably linked with software. But the war in Iraq, no matter how vile, has nothing whatsoever to do with software.
This is part of his insidious scheming. But engaging in this blackout, and widely announcing it on each and every forum he can possibly find (kuro5hin, slashdot, osnews, etc), he's discrediting the entire Free Software Movement. Real Free Software developers,
other than that he put together a version of Darwin that has a bunch of GNU software.
If I wanted GNU software, I could get it from Fink or OpenDarwin instead. Frankly, the only point of GNU-Darwin, as far as I can tell, was to put "GNU" in front of the name.
A prime motivation to develope SVG-based UI imagery is to leverge the vector-processing abilities of current videocards, which are currently utilized only by videogames.
This worked for Apple with Quartz, since Apple is in control of the hardware. (actually, Quartz does a lot of prerendering, IIRC). But KDE is not in control of the user's hardware. This is a huge obstacle to overcome.
In case you do not realize it, KDE is not limited to Linux on x86 platforms. Don't forget the BSDs and commercial Unices. And don't forget the PPCs, Sparcs and Alphas. Even on the "Lintel" platform, you still have a wide variety of graphics cards, some of which do not have decent Linux drivers. Are there even any 3D accelerated NVidia or ATI drivers for LinuxPPC?
If I'm not playing video games or rendering 3D animations, I don't need a $500 video card with attached HooverMatic fan. That's because everything else is 2D. I don't need my 2D desktop any faster, because it's so fast now I wouldn't notice the difference.
Why do you think that good GUI development consists of doing screen mockups before actual code is written?
Good GUI development starts with a GUI mockup. This is not the same thing as a screen mockup. It's far different. And this mockup is just the *start*, not the middle or the end. Good usability designers do NOT use photoshop.
In the same way that a photo of a sports car tells me nothing about the way it handles, a mere screenshot tells me little about the usability of an application. It might reveal some gross mistakes that were made, but not much else.
How much longer until we get a completely SVG user interface? Thats the main feature I want
Why? What does SVG give you, as a user, that the alternative doesn't? Positing a decent API, I can think of some benefits to the developer, but I can't think of any to the user.
First, it will be slower. Period. Your video display is bitmap based, while SVGs are vector based. It will always be faster to use a png than a svg with the current hardware. A vector based display would be different, but I just don't see any truecolor vector displays anywhere on the horizon.
Second, you won't notice any differences. The Crystal icon theme is touted as being SVG, but if you actually look at the icons, they're all pngs. A 32x32 SVG icon will look identical to the 32x32 png icon generated from it.
Okay, the scaling issue might make a difference. What if you wanted your icons to be 33x33 instead of 32x32. In that case SVG would be the way to go. But are you REALLY going to want this fine of a control over the sizes of your icons, buttons and doohickeys? But it would still be faster to render pngs from the SVGs on the fly. Unless of course you wanted fully dynamic scaling of everything. That would require a vector format, but that level of eyecandy is going to seriously impact the performance of your machine.
I also would like them to work on making it faster.
Well which is it! A dog slow SVG interface or a quick and snappy UI without SVG? You can't get both.
Why arent they trying to optimize the current code before adding new features?
You do realize that SVG is a new feature, don't you? Don't you?
How does the developer know what you use all the time? The obvious answer is to run some usability studies. The the unobvious results are that different people will use different things. What you use all the time is not going to be what I use all the time, or what Joe down the street uses all the time.
One quick example from GNOME. I just changed my wallpaper. Now the theme's color scheme doesn't match my wallpaper. How to change the color scheme without changing the widget or wm theme? In KDE this is simple. In GNOME this is an advanced option to themes. Huh?
Come on, look at these JuK screen shots and compare to iTunes screen shots.
How the hell can you determine "user experience" based on screenshots!?! I've never used iTunes, but I can guarantee you that it's outward visual appearance is completely irrelevant to its usability. Don't get me wrong, I'm sure it's very usable. But usability is not related to appearance. My Dodge Neon is every bit as usable as my neighbor's Plymouth PT Cruiser. Which should come as no surprise since the everything's the same between those vehicles except for the styling.
JuK is very usable, despite its lackluster appearance. It's probably the most usable multimedia application I've every used. It's just not pretty.
I think you're right. M$ is in the enterprise because M$ is what the decision makers use at home. Redhat is in a lot of enterprises because Redhat is what a lot of IT decision makers use at home.
At my work we had a lot of small group servers running Redhat because the guy in charge of setting them up ran Redhat elsewhere. He's gone now. These are slowly getting converted over to FreeBSD because the people who inherited them run FreeBSD. Now there's this new guy who is bitching that we should really be running Windows XP instead. Fortunately we won't, because these are 100-400MHz machines with no budget for replacement or licensing.
Redhat may be making all of its money with Redhat Enterprise, but all of its advertising comes from plain old free-beer Redhat Linux.
Ever seen a simple UML diagram for any real world solution? I haven't. UML wasn't designed to be simple. It was designed to that you could design complex software without worrying about the trivial details of implementation.
Unfortunately, even the UML advocates tend to forget this. I remember a Rational saleman giving a presentation at my work. "Look how easy this is," he said, as he brought up the UML diagram for Rose itself. Ick!
Don't confuse "complexity" with "details". This is the mistake that the author is making. Take for example a web server. The http protocol is complex no matter what language you write one in. With C you have to worry about buffer overflows. With Ruby you don't, but you still have to worry about invalid URLs. With C you're stuck with the drudgery of for loops. With Ruby you don't have to wade through that, but you still have to manage your server threads or processes. The complexity is still there, even though you're no longer worrying over the trivial details.
Building a house is a complex task, whether you forge all your own nails or buy them in bulk.
"Software should be as easy to edit as a PowerPoint presentation," Simonyi asserts.
When's the last time you saw a quality PowerPoint presentation? I've seen them, but they're rare. Presentations from people who don't know how to communicate effectively are lame as Visual Basic programs from people who don't know how to program. The style takes precedence over the actual substance.
Complexity is not something that needs to be hidden away. Software is complex. Using software is a complex activity. Writing software is more complex still. You cannot manage that complexity by imagining that it is not there. The way to manage it is to recognize that it exists.
It doesn't matter if you use C, Java, VB or Ruby, the complexity is still there. The advantages of high level languages is not that they hide away the complexity, but rather that they enable you to manage the complexity by taking care of the details.
Take any book on software development. Not programming, but development. How much time is spent on implementation? Not much. For a good project, 90% or more of the time is spent analyzing, specifying, designing and testing. This is the HARD part of developing. Give me complete specs, a valid design, and a top-notch QA group, and I could code just about anything. All that other stuff is there to MANAGE the complexity.
I've seen what Microsoft offers to make things easy. They're solutions to complexity is to ignore it, which is the wrong approach. And thus we end up with crap presentations, crap documents, and crap VB programs. It's not because these tools are crap in-and-of-themselves, but simply because they lead the user to disregard the existing complexity.
Good point, but I'll go one further. Maybe the government shouldn't be in charge of information about the government. How accurate do you think a report by the DoJ about the DoJ is going to be anyway?
A friend at school actually completed one. We took it out to the launch pad, and launched. It went up perfectly. But the 'chute never opened, and it came down just as perfect. *Straight* down. It accordianed to only one foot in height.
But have the iTech authors even suggested this yet? I know a lot of GPL authors who would rather shoot their mothers than do allow an exception in their license.
The best thing CocoaTech should do is to strip out the iTerm functionality. The community obviously doesn't want it, considering the conniption fits they're having, so don't give it to them.
The next version will either not have iTerm-like functionality, or they will get similar code from some other source.
That's what I like about the GPL, it guarantees me freedom. Instead of having a product with GPL iTerm functionality, I now get a product without it. My freedom is increased by not having Free Software. I feel so much better now...
They might have just made a whole new release and bumped the version number. As it stands now, distributing diffs between 9.2 and a future 9.3 would be extremely difficult. At least bump the revision to 9.2.1 or something. You would think they would know how to do releases by now.
Slashdot Mirror
Having written a few Qt/KDE themes, and perused in depth dozens more, I have to admit that Qt is mostly the same way.
But to be honest, the reason is to prevent bloat. Caching pixmaps uses a lot of memory. For a checkbox, there would be six pixmaps cached (foreground and background palettes for active, inactive and disabled states). This lack of caching doesn't cause a problem for simple controls like checkboxes, because redraws are uncommon.
On some themes, I can definitely see a flicker on slower machines, while on others that appear to be more complex, I see none. For example, Liquid is a pretty complex theme with blended gradients everywhere, but it makes extensive use of caching, so it has a smoother redraw than some simpler themes.
Of course, if this becomes a problem for non-cached themes, it's simple to fix, because of the way Qt is architected.
one can have objective usability tests ... the X-window model has been known to be a weakness for quite a while now.
Show us those objective usability tests. The X way of "cut-and-paste" takes far fewer actions to complete.
Windows: mouse down, mouse move, mouse up, ctrl, C, mouse move, mouse down, ctrl, V
X: mouse down, mouse move, mouse up, mouse move, mouse down, mouse button
That's three less actions with X than with Windows. It's also accomplish by one hand instead of two.
In fact, X windows itself is full of weaknesses
Unlike Windows, which everyone knows is perfect. There's no need for extensions like DirectX.
Okay, seriously. If you haven't run into any serious weaknesses in the Windows GUI model, then you haven't done any Windows GUI programming.
But all the kiddies are screaming for fully alpha blended vector graphics rendered on the fly by the GPU. Since they scream louder than the rest of us, they're they one's that get heard.
Acording to the dude's website, he is a Scientologist.
Holy shit! That explains everything. Someone get the man a deprogrammer!
politics: Annoying tick like ideas which infest sensible discussions and are the scourge of intelligent people world-wide.
politics
Etymology: Greek poli, "many" + Latin ticus, "small bloodsucking invertebrate".
I've come to the conclusion that GNU-Darwin is a hoax! Oh, there may be some software there you can download, but it's all part of the trappings for an incredibly elaborate chicanery. I am here now to unmask the perpetrator of this hornswoggle. This has gone on long enough!
Proclus (otherwise known as Michael L. Love, of no relation to Ransom that I know of) started this scheme in an effort to illustrate absurdity by being absurd. His aim is to ridicule Free Software by taking the ideas of the FSF to a absurd extreme. His ultimate goal, I fear, is to completely discredit GNU. Look at the evidence! All of it points to either a madman or a cunning conniver.
GNU-Darwin has been an ardent defender of digital liberties, and it is a platform for digital activism.
Here Proclus is deliberately asserting that GNU-Darwin is a propaganda tool. This isn't a platform to write documents, edit videos or administrate servers, "it is a platform for digital activism." The sole purpose of GNU-Darwin is to spread an ideology. An ideology that Proclus has subverted towards his own diabolical ends.
The tools of such action include, but are not limited to; boycotts, blockades, community response, de-branding, labor solidarity, whistle-blowing, etc.
Here we see how he out-stallmans Stallman. While RMS may have some firm and inflexible opinions on everything from Bush to the local dogcatcher, he keeps all of his non-software ideology out of GNU, and on a separate personal website. The point of GNU is Free Software and Free Software only. The point of GNU-Darwin is everything imaginable. It's not about the software, it's about "labor solitarity." Huh? Is Proclus going to take down his site as a sympathetic protest the next time the AFL/CIO goes on strike?
Once your credibility to engage is established, then the mere threat of such action is often sufficient to produce results.
Fortunately, GNU-Darwin has no credibility. They can boycott, blockade and de-brand all they want, and no one will care. Remember, this is all an elaborate and intricate hoax! The aim was never to get credibility, but to discredit GNU.
It is important to realize that the software freedom status of GNU-Darwin was tenuous before the change to the APSL, so we were ready to cut our losses, and remove GNU-Darwin from the internet, if necessary.
This is a prime example of "kindergarten activism". Proclus threatens to hold his breath until he gets his cookie. But since Apple wasn't paying attention, he has to call attention to his self-serving breath-holding, hoping that we'll feel compassion for his almost-sacrifice. How noble!
In general the effects of such action are ambiguous by nature, because the rulers certainly have an interest in denying the influence of democratic power.
This is a classic example of radical pseudo-think. He's telling you in advance that he can't point to any concrete results of his political activism, and the reason he can't is because he's oppressed by the ruling class. This is the crux of his ideology. Upon this all his hopes ride. You either laugh at this point and go on with life, or you stretch your credulity too far and you fall victim to his rantings. Don't go there!
Our third year commenced with the initiation of the The Free Darwin campaign which was followed by the antiwar blackout in March.
With this antiwar blackout, he's obviously playing along with the Free Software protests against European patents. But he went too far this time. Protests against "europatents" were justified because they were inextricably linked with software. But the war in Iraq, no matter how vile, has nothing whatsoever to do with software.
This is part of his insidious scheming. But engaging in this blackout, and widely announcing it on each and every forum he can possibly find (kuro5hin, slashdot, osnews, etc), he's discrediting the entire Free Software Movement. Real Free Software developers,
other than that he put together a version of Darwin that has a bunch of GNU software.
If I wanted GNU software, I could get it from Fink or OpenDarwin instead. Frankly, the only point of GNU-Darwin, as far as I can tell, was to put "GNU" in front of the name.
It's almost as though people feel threatened or something
It means that KDE is winning. What it's winning has yet to be determined...
A prime motivation to develope SVG-based UI imagery is to leverge the vector-processing abilities of current videocards, which are currently utilized only by videogames.
This worked for Apple with Quartz, since Apple is in control of the hardware. (actually, Quartz does a lot of prerendering, IIRC). But KDE is not in control of the user's hardware. This is a huge obstacle to overcome.
In case you do not realize it, KDE is not limited to Linux on x86 platforms. Don't forget the BSDs and commercial Unices. And don't forget the PPCs, Sparcs and Alphas. Even on the "Lintel" platform, you still have a wide variety of graphics cards, some of which do not have decent Linux drivers. Are there even any 3D accelerated NVidia or ATI drivers for LinuxPPC?
If I'm not playing video games or rendering 3D animations, I don't need a $500 video card with attached HooverMatic fan. That's because everything else is 2D. I don't need my 2D desktop any faster, because it's so fast now I wouldn't notice the difference.
Apple does this very well. Both KDE and GNOME struggle.
Probably because Apple has paid usability professionals with some experience behind them, rather than a bunch of people spouting their mouths off.
Why do you think that good GUI development consists of doing screen mockups before actual code is written?
Good GUI development starts with a GUI mockup. This is not the same thing as a screen mockup. It's far different. And this mockup is just the *start*, not the middle or the end. Good usability designers do NOT use photoshop.
In the same way that a photo of a sports car tells me nothing about the way it handles, a mere screenshot tells me little about the usability of an application. It might reveal some gross mistakes that were made, but not much else.
How much longer until we get a completely SVG user interface? Thats the main feature I want
Why? What does SVG give you, as a user, that the alternative doesn't? Positing a decent API, I can think of some benefits to the developer, but I can't think of any to the user.
First, it will be slower. Period. Your video display is bitmap based, while SVGs are vector based. It will always be faster to use a png than a svg with the current hardware. A vector based display would be different, but I just don't see any truecolor vector displays anywhere on the horizon.
Second, you won't notice any differences. The Crystal icon theme is touted as being SVG, but if you actually look at the icons, they're all pngs. A 32x32 SVG icon will look identical to the 32x32 png icon generated from it.
Okay, the scaling issue might make a difference. What if you wanted your icons to be 33x33 instead of 32x32. In that case SVG would be the way to go. But are you REALLY going to want this fine of a control over the sizes of your icons, buttons and doohickeys? But it would still be faster to render pngs from the SVGs on the fly. Unless of course you wanted fully dynamic scaling of everything. That would require a vector format, but that level of eyecandy is going to seriously impact the performance of your machine.
I also would like them to work on making it faster.
Well which is it! A dog slow SVG interface or a quick and snappy UI without SVG? You can't get both.
Why arent they trying to optimize the current code before adding new features?
You do realize that SVG is a new feature, don't you? Don't you?
Geez...
Only stuff I use all the time should be visible.
How does the developer know what you use all the time? The obvious answer is to run some usability studies. The the unobvious results are that different people will use different things. What you use all the time is not going to be what I use all the time, or what Joe down the street uses all the time.
One quick example from GNOME. I just changed my wallpaper. Now the theme's color scheme doesn't match my wallpaper. How to change the color scheme without changing the widget or wm theme? In KDE this is simple. In GNOME this is an advanced option to themes. Huh?
Come on, look at these JuK screen shots and compare to iTunes screen shots.
How the hell can you determine "user experience" based on screenshots!?! I've never used iTunes, but I can guarantee you that it's outward visual appearance is completely irrelevant to its usability. Don't get me wrong, I'm sure it's very usable. But usability is not related to appearance. My Dodge Neon is every bit as usable as my neighbor's Plymouth PT Cruiser. Which should come as no surprise since the everything's the same between those vehicles except for the styling.
JuK is very usable, despite its lackluster appearance. It's probably the most usable multimedia application I've every used. It's just not pretty.
The only hitch to this is that Red Hat has a services EULA that forbids redistribution once you buy their services.
WTF!
I must have been asleep during the last year, because I definitely missed that huge legal action the FSF took against Redhat.
I think you're right. M$ is in the enterprise because M$ is what the decision makers use at home. Redhat is in a lot of enterprises because Redhat is what a lot of IT decision makers use at home.
At my work we had a lot of small group servers running Redhat because the guy in charge of setting them up ran Redhat elsewhere. He's gone now. These are slowly getting converted over to FreeBSD because the people who inherited them run FreeBSD. Now there's this new guy who is bitching that we should really be running Windows XP instead. Fortunately we won't, because these are 100-400MHz machines with no budget for replacement or licensing.
Redhat may be making all of its money with Redhat Enterprise, but all of its advertising comes from plain old free-beer Redhat Linux.
Ever seen a simple UML diagram for any real world solution? I haven't. UML wasn't designed to be simple. It was designed to that you could design complex software without worrying about the trivial details of implementation.
Unfortunately, even the UML advocates tend to forget this. I remember a Rational saleman giving a presentation at my work. "Look how easy this is," he said, as he brought up the UML diagram for Rose itself. Ick!
There's some unnecessary redundancy there.
Don't confuse "complexity" with "details". This is the mistake that the author is making. Take for example a web server. The http protocol is complex no matter what language you write one in. With C you have to worry about buffer overflows. With Ruby you don't, but you still have to worry about invalid URLs. With C you're stuck with the drudgery of for loops. With Ruby you don't have to wade through that, but you still have to manage your server threads or processes. The complexity is still there, even though you're no longer worrying over the trivial details.
Building a house is a complex task, whether you forge all your own nails or buy them in bulk.
"Software should be as easy to edit as a PowerPoint presentation," Simonyi asserts.
When's the last time you saw a quality PowerPoint presentation? I've seen them, but they're rare. Presentations from people who don't know how to communicate effectively are lame as Visual Basic programs from people who don't know how to program. The style takes precedence over the actual substance.
Complexity is not something that needs to be hidden away. Software is complex. Using software is a complex activity. Writing software is more complex still. You cannot manage that complexity by imagining that it is not there. The way to manage it is to recognize that it exists.
It doesn't matter if you use C, Java, VB or Ruby, the complexity is still there. The advantages of high level languages is not that they hide away the complexity, but rather that they enable you to manage the complexity by taking care of the details.
Take any book on software development. Not programming, but development. How much time is spent on implementation? Not much. For a good project, 90% or more of the time is spent analyzing, specifying, designing and testing. This is the HARD part of developing. Give me complete specs, a valid design, and a top-notch QA group, and I could code just about anything. All that other stuff is there to MANAGE the complexity.
I've seen what Microsoft offers to make things easy. They're solutions to complexity is to ignore it, which is the wrong approach. And thus we end up with crap presentations, crap documents, and crap VB programs. It's not because these tools are crap in-and-of-themselves, but simply because they lead the user to disregard the existing complexity.
FreeBSD uses Forth in the boot manager. We're the future, and Linus is following.
p.s. Yes, I know the parent post was a joke. So is this one. Get over it.
p.p.s. Don't you long for the days when you didn't have to put stupid disclaimers like the above in your Slashdot posts?
Good point, but I'll go one further. Maybe the government shouldn't be in charge of information about the government. How accurate do you think a report by the DoJ about the DoJ is going to be anyway?
A friend at school actually completed one. We took it out to the launch pad, and launched. It went up perfectly. But the 'chute never opened, and it came down just as perfect. *Straight* down. It accordianed to only one foot in height.
But have the iTech authors even suggested this yet? I know a lot of GPL authors who would rather shoot their mothers than do allow an exception in their license.
The best thing CocoaTech should do is to strip out the iTerm functionality. The community obviously doesn't want it, considering the conniption fits they're having, so don't give it to them.
The next version will either not have iTerm-like functionality, or they will get similar code from some other source.
That's what I like about the GPL, it guarantees me freedom. Instead of having a product with GPL iTerm functionality, I now get a product without it. My freedom is increased by not having Free Software. I feel so much better now...
They might have just made a whole new release and bumped the version number. As it stands now, distributing diffs between 9.2 and a future 9.3 would be extremely difficult. At least bump the revision to 9.2.1 or something. You would think they would know how to do releases by now.