IMHO, this is the worse UI problem with GNOME. They should move all options into the app, and work on making less options all together or those options easier to use. Instead they make poor options and UIs because, "hey, you can always use gconf-editor."
But I disagree with the Xine point. The problem with "scaling UIs" is that they're never correct. Their either to simple for the advance, or to complex for the novice. A better way to go about it is to make a UI that does not get in the way of any user, but is still powerful enough for them all as well.
I realize this isn't a popular position these days, but you do not have the right, nor can you have the expectation, to be entertained 24/7.
Turn on the radio. Get a CB radio. Pull over every hour or so and watch something, fine. But watch a DVD while driving a car?!? Hey, maybe just passively watching would be too boring, hook a PS2 up to that and play some Grand Theft Auto. I'm sure that'll keep the roads real safe.
Now, the industry hasn't changed. That's the problem.
We're on the 20000th iteration of the same FPS over again and again. It's boring because no one is making anything new. In the past, we had platforms, then shooters, then this, then that. Sure, the plot and goals stayed relatively fixed, but the "gimmick" changed. Now, the industry is stuck soully on 3d platformers and FPS, both of which have lost the "novelty" appear and thus feel old. Not just old, old and busted.
What the game industry needs is a new hotness. That's the reason people looked in the magazines, to see better-graphics, new game ideas, and something to look forward to. These days, other then titles there's nothing that changes, so there's nothing to look forward to.
Gaming's always been a factory. The problem now is they've been pushing the same models out too long. The very fact that you know the game market is predictable is the thing that is sapping the life from it.
You simply want a button that you're going to use once and never need to use again?
Well, ya. Of course he does. That's what a option IS. If you changed it daily/weekly/hourly, it should be in the app's main window.
It's not option bloat to want a button to change from "new UI which is untested and is completely different from what's been used for ~3 years now for this app" to "old mode which is still available."
If Lucas decides to explain the source of the Sith's hatred in this final film, I would definitely appreciate the effort (as long as he doesn't reduce it to something as stupid as mitochlorians).
Obi-Wan (standing over a fallen Sith): My God. I've never seen such high levels of "hater-aide" before...
That would also destroy the "any later version" escape hatch for fixing any found loopholes or similar. It would be unlikely that any future evil turned FSF would be writing licenses that GPL lovers would wish to use.
You say this as if the FSF are the only body that makes Free licenses. Hell, I could write one up myself in a few moments. The sole, lone benefit of the GPL is that it's had a handful of law-speakers pour over it. Nothing else. That can be done by any other org. that wants too, to any other license they wish to use as a Free license. And if we're assuming someone were to take over the FSF for evil, a large number of the pro bono legal beagals would soon be looking for somewhere else to help.
Actually, it would be both. As a active contributer and maintainer of the main tree, he holds copyright over the kernel codebase proper, in addition to holding the trademark "Linux."
Imagine this (mainly because I imagine it daily, and slowly, ever so slowly, am hoping to code something like it:P;) ):
You insert your camera, and drag it's folder (USB drive) to your image program. It opens up, and you have them all in a wee slide-show esqe format. Now you notice the little info box over the image. You double-click it, and it brings up something like:
"Location: People: Year:"
There can be more then 3 fields, but this is just a "I'm lazy and making a comment" thumbnail.;) So you look at it, and you check two boxes for the fields "Location" and "People," then fill in the people and location. All the photos then get those values. It was a vacation, so the contents are relatively set. (granted, it'd be a bit more laborous with just random photos, but that's a problem solved by a better UI. As I said, this is just a thumbnail) Later that year, you want to find all those photos. You open a filer window, and enter "vacation." It pulls up all your photos. Then you say "2004." It pulls up all the photos from 2004. Why? Because it snagged the date data for the EXIF tags. Most importantly, though, you didn't have to go looking by filename, which since it's from a digital camera is meaningless, and since it's on a computer is too short and nondescript to mean anything. Instead you searched with the metadata, and now you go looking for it by thumbnail from a population that's already pretty close to what you want.
As more apps did this (and more users want the camera to store more info the in EXIF tags so they don't have to;) ) grabbing this info with even less user input at the app becomes more and more possible. In the end entering any metadata would be a simple to do and wholely replace using filenames.
You can already (to a point) do this with audio, and RoxCD comes with defaults that enters the artist, album, track number, and track name into the OGG Vorbis file it generates. (Which is one of the reasons I started it - I was annoyed so many rippers never bothered with that) Eventually I hope to get something coded up so I never have to search through the file system for a track again.
So while it doesn't take all the work out of making files, and most importantly doesn't stop those of us (like me) who do keep files in a deep directory heirchy;), it does take the work out of storing and finding the files. And that is where the user spends most of their 'filing clerk' time.:)
An even better story will be when folks realize that it is OK for the whole world not to agree with them on philosophy. Especially when those philosophies have economic ramifications.
Closed software has it's place. And that place is in a closed OS.
I'm sorry, but when you bring something that is against the very nature of why a set of software came into existance (GPLed software making up the majority or whole of a GNU/Linux system), and the situation people are up in arms about resembles the reason why one of the main originators of the movement that spawned it, you can't say "hey, this is totally against why this software exists. Why are you people so mean."
Complaining that "oh wow, these GNU/Linux people should stop bitching about Open Sourced drivers" is like saying "wow, these Alaskans should stop bitching about warm clothing and shelter from the winds." It misses the very context in which the discussion take place.
Since you like Cocoa, you might also want to look at GNUStep on the GNU/Linux and *N*X side of things. It attempts to provide the next-step-after-OpenSTEP UI toolkit, and even brings some of the Cocoa APIs as well.
Best of all, I hear taking code from it to OS X is just a quick recompile for a native experience, which is always nice. You can see screenshots of a app that does just that here.
Let's just stop calling it "Linux" in general. It's not.
If I use a GNU/Linux system these days, it's usually Debian. If I sit down at a RedHat box, it's not even halfway similar. I sit at a Mandrake box, and it's not even half way similar. And there's no telling what the hell will be on a Gentoo box, with what configurations. Different inits, different tools for configuring said inits, different userspace tools (in version, behavior, etc) different patches against the kernel/tools/desktop, different desktops.
They aren't "linux," they are Debian, Mandrake, Gentoo and RedHat. Just call them what they &*@^ing are.
So, people in fact love when the machine works in a way resembling behaviour of real-life objects, but it seems that only when the "spatial" application is a web browser: they accept the book metaphor with web pages, but reject the drawer metaphor with folders and files. Sometimes they even abuse the physical metaphor of tabbed browsing by opening multiple pages - not subpages of the same web site! - in multiple tabs of a browser window. I even know few people who never open more than one browser window, viewing all pages in tabs; I hope they do not try to glue a daily set of newspapers together before reading them...
If you take a child, and give it a controller where you've wired it to send the left button press as a right, and the right as a left, and have them play something like Super Mario Brothers or some other side-scrolling platformer, then they just happily learn that the left/right buttons are "scroll screen left" or "scroll screen right." They'd learn it like a airplane yoke, instead of as "make the player move left/right."
If you have a 20 year old who's played games for years and do the same, they will not be able to get used to it because they've become habituated on the "common" controller interactions. They may be able to use it, but not well, and will often make a mistake because the habits they had will take over in a situation where they must make a split-second decision.
It has nothing to do with visual perception, and everything to do with the habitual nature of using a computer. You take on habits for getting a response, such as entering the letters A, B, or C. Consider you're a typist and you change keyboard layouts, the old layout has become habitual and you will have a hard time getting used to the new layout. Even after you've learned it, you will fall back into the old habits from the other layout when your not paying active attention to the keys you are using. It's annoying, frustrating, and to be expected as there isn't something unique to allow you to create new habits and not fall into your old ones without be constantly concious of the enviroment, which is not a good interface.
A UI that is "close" to the old UI is bad, as the user will have retained too much muscle memory and habits from the old one. Presumably they for the most part might work, but when they don't they will lead the user into annoyance and confusion. A better UI would be one that is instantly and obviously dissimilar but also trivial to learn and easy (if not easier) to use once you have learned how.
Bullox. They are bloated because the coders bloat things, not because making a UI that does not use the terminal needs to be excessively bloated.
Saying "GNOME/KDE are bloated because they can't not be bloated" is not a answer. ROX's desktop uses mostly python apps, and is STILL snappier and quicker to load then the latest GNOME release. And other then a UI for positioning the panels, it has all the features of the base GNOME DE now. As it adds features it doesn't add load time because it does things right, pushing the code into external apps for configuring for example. The user doesn't notice because they just find the AppDir using ROX-Filer, and run the AppDir like a option in KDE's ControlCenter. The user doesn't know or care that it's not all one app, but they do know that it loads faster at boot and at click then the counterpart in the other DE.
XFCE does similar with it's configuration. It's all plugins for a seperate app that isn't run unless the user wants to configure. The plugins are small, and app is small, it's dependancies are small, it loads fast and snappy, and the user can actually use their computer, instead of the other way around.
Two answers, both work well, that the user doesn't even know happen because they are hidden with a correct UI. It's simple things like this that make the big DEs bloated.
And I'll say it again. KDE and GNOME not only want the broken UI behaviors of Windows, but the bloat as well.
If you look at the Amiga, RISCOS, and (for the most part) the Macintosh, they achieved unbelievable usablity not because they had flashy libraries or k3wl tr4nsp4r3nt t3rm1n4lz, it's because they use the hardware well and tried to give a small, innovative UI to the user. Just because their not mainstream you might not see them, but look at XFCE and ROX. Both are desktops (hell, ROX is trying to write our own desktop apps) that take the GTK libraries and make fast, powerful, and small desktops that do their jobs and do them well. Not only are their easier to debug, they are easier to use and faster even on dated hardware because they are not trying to be the massive beasts KDE and GNOME are pushing them to be.
You will not see ROX push a web browser into ROX-Filer. You won't see XFCE do alpha-blending in the dock. You'll just see programs that do a job, and do it well. With GNOME's main library.
They follow the UNIX way which has been lost on the big desktops and it shows.
For a example of how bad the big desktops are at deciding where to put things, look at the fake SSH/FTP/HTTP/HTTPS things in KDE via FISH, and GNOME via GNOME-VFS. THESE SHOULD NOT BE DONE BY THE DESKTOP AT THE LEVEL THEY ARE! They should be pushed to a small userland daemon at best, or a small combination of userland and kernelspace at worst. This way all apps could use them seemlessly and without the massive overhead of bringing in Yet Another Library, including current commandline tools, without changing a line of code. It's simple things like that that bloat them, and as they act as a point of code "contention," hurt them both because it forces KDE developers to work on one implementation, and GNOME developers to work on another.
And for the last time, UI and library bloat != UI ease of use. Just look at the old Macs. If a interface is correctly done, even if it bares no resemblence to a existing one, the time to relearn it should be trivial.
Actually, the core (drivers, servers, etc) was put under the new license, meaning they could no longer use GPLed patches. Assuming you'd want to.
The XLib was still MITed, so linking GPLed software against it was fine. Stallman even "vetted" it, so it was kosher there. IMHO, it was more a case of coders saying "we can no longer trust this group of project leaders," and deciding to just fork with [XOrg/Freedesktop].org under more open leadership (of which one can not discount Keith P.'s place) then anything else.
Slashdot Mirror
IMHO, this is the worse UI problem with GNOME. They should move all options into the app, and work on making less options all together or those options easier to use. Instead they make poor options and UIs because, "hey, you can always use gconf-editor."
But I disagree with the Xine point. The problem with "scaling UIs" is that they're never correct. Their either to simple for the advance, or to complex for the novice. A better way to go about it is to make a UI that does not get in the way of any user, but is still powerful enough for them all as well.
I realize this isn't a popular position these days, but you do not have the right, nor can you have the expectation, to be entertained 24/7.
Turn on the radio. Get a CB radio. Pull over every hour or so and watch something, fine. But watch a DVD while driving a car?!? Hey, maybe just passively watching would be too boring, hook a PS2 up to that and play some Grand Theft Auto. I'm sure that'll keep the roads real safe.
from xorg-x11-6.7.0-r1.ebuild:
Now, the industry hasn't changed. That's the problem.
We're on the 20000th iteration of the same FPS over again and again. It's boring because no one is making anything new. In the past, we had platforms, then shooters, then this, then that. Sure, the plot and goals stayed relatively fixed, but the "gimmick" changed. Now, the industry is stuck soully on 3d platformers and FPS, both of which have lost the "novelty" appear and thus feel old. Not just old, old and busted.
What the game industry needs is a new hotness. That's the reason people looked in the magazines, to see better-graphics, new game ideas, and something to look forward to. These days, other then titles there's nothing that changes, so there's nothing to look forward to.
Gaming's always been a factory. The problem now is they've been pushing the same models out too long. The very fact that you know the game market is predictable is the thing that is sapping the life from it.
Well, ya. Of course he does. That's what a option IS. If you changed it daily/weekly/hourly, it should be in the app's main window.
It's not option bloat to want a button to change from "new UI which is untested and is completely different from what's been used for ~3 years now for this app" to "old mode which is still available."
Obi-Wan (standing over a fallen Sith): My God. I've never seen such high levels of "hater-aide" before...
(music dramatically swells)
And even if you do, it's still a Windows sale as they usually are just "pre-wiped" for you.
They why would they put it on iTunes where you can sample it before you buy?
You could have saved a lot of typing by just saying six words:
Team Fortress Two
Dreamcast Half Lift
While I don't doubt HL2'll ship, AFAI am concerned they lost all cred IRT ship dates after that.
You say this as if the FSF are the only body that makes Free licenses. Hell, I could write one up myself in a few moments. The sole, lone benefit of the GPL is that it's had a handful of law-speakers pour over it. Nothing else. That can be done by any other org. that wants too, to any other license they wish to use as a Free license. And if we're assuming someone were to take over the FSF for evil, a large number of the pro bono legal beagals would soon be looking for somewhere else to help.
Actually, it would be both. As a active contributer and maintainer of the main tree, he holds copyright over the kernel codebase proper, in addition to holding the trademark "Linux."
Imagine this (mainly because I imagine it daily, and slowly, ever so slowly, am hoping to code something like it :P ;) ):
;) So you look at it, and you check two boxes for the fields "Location" and "People," then fill in the people and location. All the photos then get those values. It was a vacation, so the contents are relatively set. (granted, it'd be a bit more laborous with just random photos, but that's a problem solved by a better UI. As I said, this is just a thumbnail) Later that year, you want to find all those photos. You open a filer window, and enter "vacation." It pulls up all your photos. Then you say "2004." It pulls up all the photos from 2004. Why? Because it snagged the date data for the EXIF tags. Most importantly, though, you didn't have to go looking by filename, which since it's from a digital camera is meaningless, and since it's on a computer is too short and nondescript to mean anything. Instead you searched with the metadata, and now you go looking for it by thumbnail from a population that's already pretty close to what you want.
;) ) grabbing this info with even less user input at the app becomes more and more possible. In the end entering any metadata would be a simple to do and wholely replace using filenames.
;), it does take the work out of storing and finding the files. And that is where the user spends most of their 'filing clerk' time. :)
You insert your camera, and drag it's folder (USB drive) to your image program. It opens up, and you have them all in a wee slide-show esqe format. Now you notice the little info box over the image. You double-click it, and it brings up something like:
"Location:
People:
Year:"
There can be more then 3 fields, but this is just a "I'm lazy and making a comment" thumbnail.
As more apps did this (and more users want the camera to store more info the in EXIF tags so they don't have to
You can already (to a point) do this with audio, and RoxCD comes with defaults that enters the artist, album, track number, and track name into the OGG Vorbis file it generates. (Which is one of the reasons I started it - I was annoyed so many rippers never bothered with that) Eventually I hope to get something coded up so I never have to search through the file system for a track again.
So while it doesn't take all the work out of making files, and most importantly doesn't stop those of us (like me) who do keep files in a deep directory heirchy
Closed software has it's place. And that place is in a closed OS.
I'm sorry, but when you bring something that is against the very nature of why a set of software came into existance (GPLed software making up the majority or whole of a GNU/Linux system), and the situation people are up in arms about resembles the reason why one of the main originators of the movement that spawned it, you can't say "hey, this is totally against why this software exists. Why are you people so mean."
Complaining that "oh wow, these GNU/Linux people should stop bitching about Open Sourced drivers" is like saying "wow, these Alaskans should stop bitching about warm clothing and shelter from the winds." It misses the very context in which the discussion take place.
Since you like Cocoa, you might also want to look at GNUStep on the GNU/Linux and *N*X side of things. It attempts to provide the next-step-after-OpenSTEP UI toolkit, and even brings some of the Cocoa APIs as well.
Best of all, I hear taking code from it to OS X is just a quick recompile for a native experience, which is always nice. You can see screenshots of a app that does just that here.
Not send them a free copy. ;)
Let's just stop calling it "Linux" in general. It's not.
If I use a GNU/Linux system these days, it's usually Debian. If I sit down at a RedHat box, it's not even halfway similar. I sit at a Mandrake box, and it's not even half way similar. And there's no telling what the hell will be on a Gentoo box, with what configurations. Different inits, different tools for configuring said inits, different userspace tools (in version, behavior, etc) different patches against the kernel/tools/desktop, different desktops.
They aren't "linux," they are Debian, Mandrake, Gentoo and RedHat. Just call them what they &*@^ing are.
Yes, and then those freaks will do something really crazy, like make a text editor with tabs and use it as their DE's default text editor.
Madness! Utter madness!
Actually, no. Consider it like this:
If you take a child, and give it a controller where you've wired it to send the left button press as a right, and the right as a left, and have them play something like Super Mario Brothers or some other side-scrolling platformer, then they just happily learn that the left/right buttons are "scroll screen left" or "scroll screen right." They'd learn it like a airplane yoke, instead of as "make the player move left/right."
If you have a 20 year old who's played games for years and do the same, they will not be able to get used to it because they've become habituated on the "common" controller interactions. They may be able to use it, but not well, and will often make a mistake because the habits they had will take over in a situation where they must make a split-second decision.
It has nothing to do with visual perception, and everything to do with the habitual nature of using a computer. You take on habits for getting a response, such as entering the letters A, B, or C. Consider you're a typist and you change keyboard layouts, the old layout has become habitual and you will have a hard time getting used to the new layout. Even after you've learned it, you will fall back into the old habits from the other layout when your not paying active attention to the keys you are using. It's annoying, frustrating, and to be expected as there isn't something unique to allow you to create new habits and not fall into your old ones without be constantly concious of the enviroment, which is not a good interface.
A UI that is "close" to the old UI is bad, as the user will have retained too much muscle memory and habits from the old one. Presumably they for the most part might work, but when they don't they will lead the user into annoyance and confusion. A better UI would be one that is instantly and obviously dissimilar but also trivial to learn and easy (if not easier) to use once you have learned how.
Bullox. They are bloated because the coders bloat things, not because making a UI that does not use the terminal needs to be excessively bloated.
Saying "GNOME/KDE are bloated because they can't not be bloated" is not a answer. ROX's desktop uses mostly python apps, and is STILL snappier and quicker to load then the latest GNOME release. And other then a UI for positioning the panels, it has all the features of the base GNOME DE now. As it adds features it doesn't add load time because it does things right, pushing the code into external apps for configuring for example. The user doesn't notice because they just find the AppDir using ROX-Filer, and run the AppDir like a option in KDE's ControlCenter. The user doesn't know or care that it's not all one app, but they do know that it loads faster at boot and at click then the counterpart in the other DE.
XFCE does similar with it's configuration. It's all plugins for a seperate app that isn't run unless the user wants to configure. The plugins are small, and app is small, it's dependancies are small, it loads fast and snappy, and the user can actually use their computer, instead of the other way around.
Two answers, both work well, that the user doesn't even know happen because they are hidden with a correct UI. It's simple things like this that make the big DEs bloated.
And I'll say it again. KDE and GNOME not only want the broken UI behaviors of Windows, but the bloat as well.
If you look at the Amiga, RISCOS, and (for the most part) the Macintosh, they achieved unbelievable usablity not because they had flashy libraries or k3wl tr4nsp4r3nt t3rm1n4lz, it's because they use the hardware well and tried to give a small, innovative UI to the user. Just because their not mainstream you might not see them, but look at XFCE and ROX. Both are desktops (hell, ROX is trying to write our own desktop apps) that take the GTK libraries and make fast, powerful, and small desktops that do their jobs and do them well. Not only are their easier to debug, they are easier to use and faster even on dated hardware because they are not trying to be the massive beasts KDE and GNOME are pushing them to be.
You will not see ROX push a web browser into ROX-Filer. You won't see XFCE do alpha-blending in the dock. You'll just see programs that do a job, and do it well. With GNOME's main library.
They follow the UNIX way which has been lost on the big desktops and it shows.
For a example of how bad the big desktops are at deciding where to put things, look at the fake SSH/FTP/HTTP/HTTPS things in KDE via FISH, and GNOME via GNOME-VFS. THESE SHOULD NOT BE DONE BY THE DESKTOP AT THE LEVEL THEY ARE! They should be pushed to a small userland daemon at best, or a small combination of userland and kernelspace at worst. This way all apps could use them seemlessly and without the massive overhead of bringing in Yet Another Library, including current commandline tools, without changing a line of code. It's simple things like that that bloat them, and as they act as a point of code "contention," hurt them both because it forces KDE developers to work on one implementation, and GNOME developers to work on another.
And for the last time, UI and library bloat != UI ease of use. Just look at the old Macs. If a interface is correctly done, even if it bares no resemblence to a existing one, the time to relearn it should be trivial.
I dunno, he's been kinda quiet lately.
Actually, the core (drivers, servers, etc) was put under the new license, meaning they could no longer use GPLed patches. Assuming you'd want to.
The XLib was still MITed, so linking GPLed software against it was fine. Stallman even "vetted" it, so it was kosher there. IMHO, it was more a case of coders saying "we can no longer trust this group of project leaders," and deciding to just fork with [XOrg/Freedesktop].org under more open leadership (of which one can not discount Keith P.'s place) then anything else.
GNU Classpath
GCJ/GIJ
Kaffe VM
Jikes Java compiler
SableVM
Java-GTK
Documents about how to compile and use QTJava and KDEJava
What else do we need? How does all this not play well with Free software? We've got the tools, why not use them?
Sounds like ROX AppDir as well. Personally, I don't know why everything isn't moving to something similar. It makes things so easy. :)