If you do a distro upgrade from one finished version to another (skipping all betas and alphas), generally things work, but you are right in that this is something that needs more attention: drivers. DKMS is helping with that, dragging along any drivers you've installed when you switch kernels, but it's new and needs more attention.
Software portability in general needs a lot more love in Linux. You shouldn't have to be dependent on your repository to install the Linux software you want. An elegant solution for this still needs to be found, though Zero Install is doing fairly well at overcoming a lot of the challenges, but it's sad that such a flexible package management system isn't the root system. Who knows, maybe eventually they'll make a Zero Install distro. ^^
Xubuntu 9.04 would be a bit faster at booting than Ubuntu 9.04, but there are several lightweight GUIs available. You can use a normal Ubuntu install and install the xubuntu-desktop metapackage to get it. Under System > Admin > Login Window, you can set automatic logins and other things (it also asks you if you want auto login during the install).
Yeah, try telling that to a normal computer user. I'm sure they will be in awe at their freedom to compile. This is one of the things that makes Linux so annoying to those who don't know how or want to deal with the hassle of compiling, and oh don't forget even to those who DO know how. Sure, the kernel is "advanced" and most users don't need to care about it, but loading drivers IS something that is important to normal computer users, as is general program portability. The "oh just compile it" attitude has hurt the push for "portable apps" (cross-distro) on Linux, and as such means that if you want something outside of your repos, it's often spans from being difficult to requiring a wizard, from the point of view of, again, normal users. Until Linux does a better job at appealing to these users and providing this amazing feature called portable apps in a much better way than it does now, Linux will be held back.
Sure, Linux has other problems too, but IMO this is the biggest and most widely ignored. At least there is Zero Install, but it's more of a patch than solution.
A free(dom) OS isn't very free to users if they don't have the freedom to easily install apps. An OS that stands for freedom should work heavily on solving Linux's packaging mess in order to provide this freedom.
That's sad to think about, since GPCPUs are coming out this year I think. So you're putting everything inside one chip, but still segregating each core to do different tasks. The end result is a computer which, while being smaller possibly, means it's not modular. Consumers of course have more control when they can replace/remove/upgrade different dedicated parts.
Or, are these new fangled GPCPUs going to be dynamic enough to get over this problem?:/
I think you have it right to some degree, I think you've identified a major hole, and that's a good idea. Standards greatly effects your freedom. What I see happening is lots of Linux companies making basically "proprietary Linux" software stacks. Yeah, it's not completely locked down, yes you can still run Linux binaries on any Linux distros thankfully. The problem is they each force their own packages to be required if you want to do automatic software updates and such through the package system, or if you want to even THINK about replacing something that you have installed. Simple binary packs are indeed just annoying for so many reasons. Want to try out a new version of X? You can't, good fucking luck. Even if you use binary installers for things that you *can* install that don't conflict with other things on your system, it's annoying because they'd need to put an uninstall icon in the menu.
The problems are all very solvable, but distro companies aren't interested in playing nicely with those solutions, they'd rather have the large sizes of their repos and such be a "feature" to attract users from other distros. Sad sad. Until Linux makes software installation as easy as on Windows, Linux will be held back and users won't be truly free.
The only thing that sucks for me is Flash is slower than the Windows version, otherwise Firefox 3 in Linux works faster for me than it does on a faster Windows machine. I love having the freedom to keep using an older comp with an updated OS.:)
Ya, that was my point, that society is dumb and cares about being "cool" and the good feeling they get by knowing that what you have has been labeled as being approved and "cool", and that's all that matters, or so they want you to think.
This mount option should have no real impact on performance, but will ensure the atomic update idiom that works on Ext3 will also work on Ext4.
Performance not being harmed pretty much tells us that it was a bug, or that this new "option" is a vast improvement. In any case, sane defaults should be the standard, and even if there was some small performance hit, should be the default for normal users.
Data centers have redundancy in data and power, normal users don't.
ReiserFS is pretty dead, while btrfs may actually be better though I can't believe it, so perhaps when Tux himself, in his 3rd incarnation, finally is my secretary and is managing my files will my computer finally have "sentience".
I know they're screwed in the bad way (pineapple...or spiked mace...you get the idea) and of course that's unfortunate, and this isn't any consolation, but that's what they get for buying an Apple product, don't expect it to work 100% with non-Apple products. I wouldn't go out and buy a Windows Mobile cell phone and 100% expect it to work on a Mac. Not that I would touch either of those things any way, just making a point.
I just find it really sad that, at the very least even if you do have a reason for some other system that does a similar thing, that's still no excuse to not make it play nicely with some standard interface. In other words, modularity. Stop making stacks of software. Software stack = lack of choice and freedom, hard to use, wasted effort, and I'm sure you could mention several other things. In other words, fragmentation. Competition is fine and good, but make it all compatible with standardized APIs so that switching is easy, and your development effort can be, you know, actually utilized by others instead of everything on top of your "stack" being wasted if your program dies off.
This isn't a jab at you or anything, but for the longest time I always thought ads were lame for just this reason. Ads a few decades ago were somewhat more intelligent, they would actually talk about a product and list some specific reasons why it was better than another product. While of course the points that were made were sometimes silly and skewed and not at all true, at least these were more competitive and intelligent.
Then came the newer ads, the ones that thought it was cooler to completely ignore the competing product and instead merely state that their product was "cool". Of course "cool" is a ridiculous concept to the intelligent, but to the general public, they care, you're completely right. Linux users, many of which are geeks and don't care about "being cool", will have a hard time marketing to the "cool crowd", indeed. Linux, fortunately, has grown enough to grab attention from many who do care about being cool, and there are a lot of cool things about Linux.
Just so long as Linux development remembers to care about the really important things, like gaining developers and making development easier, making fundamental aspects of the OS easier for everyone (*ahem*installing drivers without compiling*cough*easy cross-distro program installation aka basic package portability), and in general making the average computer user's life easier (and the non-average user's life too), as well as being "cool", with your 3D desktop and your sleek black PS3s (which can barely do anything anyway due to hypervisor poop).
99% of the time however, taking the anti-Microsoft side happens to also be the intelligent/reasonable/correct side, and already comes with many reasons to back it up. Most people don't just hate Microsoft just because, they hate them for real reasons. But yeah, I got your point and agree. It's not just MS, it's Adobe and other pedlars of proprietary software that use tactics to try to push their software on students instead of being fair and using standards where possible.
Wow, maybe Firefox 1 and 2, but 3 seems to have done away with that problem. I leave Firefox 3 open all the time and never have an issue, and I'm running an oldish computer by today's standards (AMD Athlon 64, AGP GFX, etc).
The open source does-it-all collaborative suite programs like Exchange certainly need to be more competitive, but so far they are promising. So, maybe your Exchange-using organization should switch. Two "larger" looking ones I've found are:
As for an Exchange client solution, well...difficult, since Microsoft controls the communication interface for that and I doubt at this stage they are interested in freely playing nice with non-Microsoft clients. In that case, your best bet is to change the server.
I take that back, they may be getting enough pressure for playing nice with clients due to the whole.NET + Evolution thing, seems like they might be pushing for it a bit, since Evolution is somewhat working with Exchange. I can at least get email as well as see calendar events and get warnings about them, but sending emails I haven't tried very hard to figure out how to do. Hopefully the push for that isn't some sort of patent backstabbing deal though. Regardless, I'd rather not deal with that patent troll if I can help it and would instead ask for something that at least used standards so that you could use what client you wanted to. Microsoft doesn't know the meaning of the term "standard", plus, you know, there's the whole hatred for Linux and just trying to pull everyone onto their platforms thing so they can "collect at a later date" - Bill Gates.
That would be a very silly limitation for OOo to have, agreed, and one would think that would be an easy thing to fix, though that's definitely a big sheet...maybe someone should break things apart a little more.
Damn that sucks, I'd organise an appeal for the student body to have teachers adopt non-proprietary software. To be teaching proprietary software especially with things like this, a stupid office suite, and demanding it's use, is just all kinds of dumb, and I believe wrong for a college to do. And some people wonder why "student software" is so cheap, like a drug sample. No, students should have the freedom to choose which office software they want to use to complete an assignment.
Please point out areas in which GNU/Linux could use improvements, that will be a much more constructive post. It will help GNU/Linux to improve.
Most readers here find GNU/Linux to be a superior OS in most every way. It has some shortcomings, and they are great to address and talk about, since it's a community OS.
The licensing of GNU/Linux is very important and is what makes it special. The freedom to tinker and improve is a feature that outweighs most shortcomings in the minds of many readers here, but all the usefulness and benefits and perks of GNU/Linux of course really help too. Not to mention, you know, it's free. If you want to stay on the Microsoft upgrade treadmill and in it's OS prison though, be my guest.
So yeah, of course mocking many here with things like GNU/Linux leg humpers, etc, is going to get you flamed in return. Not a great way to start an intelligent debate.
And even stupider, who'd want to strengthen software patents and give Microsoft leverage by caving into their FUD and paying for this "protection" racket? All efforts to invalidate all software patents should be taken by all companies and citizens.
Cool, signed up. I don't know a whole lot about development, but certainly wouldn't mind helping out if it's something I could do. I would just like to see GNU/Linux get an upgrade if a more elegant solution can be adopted.
I think moving to a classification system is the best answer to avoid having to have the gamer completely define everything themselves, so that the game developer can simply say "when the controller layout is like type Z, assign this to this and that to that" but of course allow the gamer to change those assignments if they want to. Or better yet, the dev says "I need a thumbstick on the left of the controller to be assigned to this, and the dpad to my dpad here, and the primary buttons on the right side defined to this, with any leftovers left undefined". That way they wouldn't even have to define what to do for every classification situation, if you could make the communication system intelligent enough so it could figure out what to do in most cases automatically, since that is the goal, to make it automatic for both the developers and the gamers with options for adjustments later if needed.
Yep, and devs need to start listening to users like you more and waking up to the bits that are difficult about Linux, and redo things or patch things or do whatever needs to be done to make Linux easier. Just seems like us users who mention things that aren't functional in Linux are ignored, sometimes because "that's the way things are" and "it can't be changed (easily)". It amazes me when anyone has the mentality that there are some huge barriers in software that prevent certain things from being accomplished. That's a cop-out, it's software, if you can think it you can do it. It's like, sometimes Linux users who learn Linux well enough to understand what the issues are and how things work at the same time become indoctrinated with the status quo, thereafter unwilling to think outside the (Linux) box to how things could be, and could be better, and how things could work.
OK, well you're a good teacher then, but we'll see if that knowledge becomes mainstream or not, maybe it will dunno, right now the terms "Linux" and "Ubuntu" are pretty big though...
Since X and Y are not equal, you have to find a way to map the events of X to Y. And thats where the trouble starts, since you don't know which combination of events in X are good to produce the events in Y
Right, so you'd need to break a controller down into it's component parts, and then register those parts with the API, so that a) a gamer can map any of those parts that they want to any of the buttons defined in any game easily, and b) provide a calibration program so that if the default settings are off for that controller, or if the controller isn't recognised by the "system" at all, you can still teach it.
But of course thats not all, input devices aren't necessarily clear about their capabilities, so if a device says it has 8 buttons, it might in reality have 4 buttons and a dpad or worse you have something like the WIZ which reports the dpad has 8 buttons, since the diagonals are handled separately. You need a big fat table of information to clean that up. But even if you have it cleared up you are still left with a lot of ambiguity, shall a emulator use the dpad or the analogstick, shall button 5 and 6 get mapped to L/R or to C/Z (gamepad with 6 face buttons), which gamepad shall the emulator use when there are multiple available and so on.
So you have specific default settings being mapped if the controller is being recognized, but if the "system" recognizes them incorrectly or doesn't at all or whatever, the joystick program that should come with KDE and Gnome should not only be able to calibrate, but change the input type. Say, change what came up as a "button" to a "trigger", if that is necessary. Once the inputs are all defined and calibrated, those standardized input classes that the system is now configured for can be easily passed to any game that uses the standardized API/system. In the game, you can then adjust those further as far as what buttons on X controller should be used for the Y controller for the gaming system.
So, what I'm saying then is desktop GNU/Linux needs to do as good of a job as it can to get things right, but then it should offer the ability to easily allow users to tweak and define what it detects, and to even make big adjustments if necessary like changing a sensed button to a trigger or whatnot. If it takes care of all of that part of it, developers of Linux games and other software won't have to deal with it themselves. You offload lots of work and headaches to the system developers, so that Linux development is easier for them and programs are easier to make.
Microsoft has invested a lot of interest in doing this, so that their OS is easier to program for. I'm not saying they were successful or anything about them, believe me I hate many many things Microsoft has done as much as the next informed Linux user, but the goal to make it easier for developers is a very good goal. It will help increase the adoption of Linux, not to mention make it easier for developers to flex their brains and creativity, which is exactly what an open OS should be focused on. I want to help make Linux more user friendly than Windows in every way possible, that will be how Linux succeeds on the desktop.
Some of this is fixable, but most is not, unless you actually hack the game. The problem is most obviously demonstrated when you look at PC to console conversions or the other way around, those that end up with a good interface need to basically redesign large part of it, since different input devices have just very different capabilities.
Yes, there has to be translation there I understand that and agree, but if you centralize the translating work to the OS, you simplify it for game developers. They can even have the capability of using the input device directly like the old ways, or to interface with the new standard. So here's a concept mock-up of what I'm saying:
Controller is plugged in >>>> Controller is either recog
Trying to explain to a normal computer user what a kernel is, why they should care, and what all the different "Linuxes" actually are, will confuse, not to mention be a very boring conversation for them. GNU/Linux needs some names to get behind to make it simple, Linux and Ubuntu are two that are becoming somewhat common.
Slashdot Mirror
If you do a distro upgrade from one finished version to another (skipping all betas and alphas), generally things work, but you are right in that this is something that needs more attention: drivers. DKMS is helping with that, dragging along any drivers you've installed when you switch kernels, but it's new and needs more attention.
Software portability in general needs a lot more love in Linux. You shouldn't have to be dependent on your repository to install the Linux software you want. An elegant solution for this still needs to be found, though Zero Install is doing fairly well at overcoming a lot of the challenges, but it's sad that such a flexible package management system isn't the root system. Who knows, maybe eventually they'll make a Zero Install distro. ^^
Don't use it, don't pay for it, don't support it.
Xubuntu 9.04 would be a bit faster at booting than Ubuntu 9.04, but there are several lightweight GUIs available. You can use a normal Ubuntu install and install the xubuntu-desktop metapackage to get it. Under System > Admin > Login Window, you can set automatic logins and other things (it also asks you if you want auto login during the install).
Yeah, try telling that to a normal computer user. I'm sure they will be in awe at their freedom to compile. This is one of the things that makes Linux so annoying to those who don't know how or want to deal with the hassle of compiling, and oh don't forget even to those who DO know how. Sure, the kernel is "advanced" and most users don't need to care about it, but loading drivers IS something that is important to normal computer users, as is general program portability. The "oh just compile it" attitude has hurt the push for "portable apps" (cross-distro) on Linux, and as such means that if you want something outside of your repos, it's often spans from being difficult to requiring a wizard, from the point of view of, again, normal users. Until Linux does a better job at appealing to these users and providing this amazing feature called portable apps in a much better way than it does now, Linux will be held back.
Sure, Linux has other problems too, but IMO this is the biggest and most widely ignored. At least there is Zero Install, but it's more of a patch than solution.
A free(dom) OS isn't very free to users if they don't have the freedom to easily install apps. An OS that stands for freedom should work heavily on solving Linux's packaging mess in order to provide this freedom.
That's sad to think about, since GPCPUs are coming out this year I think. So you're putting everything inside one chip, but still segregating each core to do different tasks. The end result is a computer which, while being smaller possibly, means it's not modular. Consumers of course have more control when they can replace/remove/upgrade different dedicated parts.
:/
Or, are these new fangled GPCPUs going to be dynamic enough to get over this problem?
I think you have it right to some degree, I think you've identified a major hole, and that's a good idea. Standards greatly effects your freedom. What I see happening is lots of Linux companies making basically "proprietary Linux" software stacks. Yeah, it's not completely locked down, yes you can still run Linux binaries on any Linux distros thankfully. The problem is they each force their own packages to be required if you want to do automatic software updates and such through the package system, or if you want to even THINK about replacing something that you have installed. Simple binary packs are indeed just annoying for so many reasons. Want to try out a new version of X? You can't, good fucking luck. Even if you use binary installers for things that you *can* install that don't conflict with other things on your system, it's annoying because they'd need to put an uninstall icon in the menu.
The problems are all very solvable, but distro companies aren't interested in playing nicely with those solutions, they'd rather have the large sizes of their repos and such be a "feature" to attract users from other distros. Sad sad. Until Linux makes software installation as easy as on Windows, Linux will be held back and users won't be truly free.
The only thing that sucks for me is Flash is slower than the Windows version, otherwise Firefox 3 in Linux works faster for me than it does on a faster Windows machine. I love having the freedom to keep using an older comp with an updated OS. :)
Ya, that was my point, that society is dumb and cares about being "cool" and the good feeling they get by knowing that what you have has been labeled as being approved and "cool", and that's all that matters, or so they want you to think.
Lame.
This mount option should have no real impact on performance, but will ensure the atomic update idiom that works on Ext3 will also work on Ext4.
Performance not being harmed pretty much tells us that it was a bug, or that this new "option" is a vast improvement. In any case, sane defaults should be the standard, and even if there was some small performance hit, should be the default for normal users.
Data centers have redundancy in data and power, normal users don't.
So at what point is "sentience" achieved then? :3
ReiserFS is pretty dead, while btrfs may actually be better though I can't believe it, so perhaps when Tux himself, in his 3rd incarnation, finally is my secretary and is managing my files will my computer finally have "sentience".
I know they're screwed in the bad way (pineapple...or spiked mace...you get the idea) and of course that's unfortunate, and this isn't any consolation, but that's what they get for buying an Apple product, don't expect it to work 100% with non-Apple products. I wouldn't go out and buy a Windows Mobile cell phone and 100% expect it to work on a Mac. Not that I would touch either of those things any way, just making a point.
I just find it really sad that, at the very least even if you do have a reason for some other system that does a similar thing, that's still no excuse to not make it play nicely with some standard interface. In other words, modularity. Stop making stacks of software. Software stack = lack of choice and freedom, hard to use, wasted effort, and I'm sure you could mention several other things. In other words, fragmentation. Competition is fine and good, but make it all compatible with standardized APIs so that switching is easy, and your development effort can be, you know, actually utilized by others instead of everything on top of your "stack" being wasted if your program dies off.
This isn't a jab at you or anything, but for the longest time I always thought ads were lame for just this reason. Ads a few decades ago were somewhat more intelligent, they would actually talk about a product and list some specific reasons why it was better than another product. While of course the points that were made were sometimes silly and skewed and not at all true, at least these were more competitive and intelligent.
Then came the newer ads, the ones that thought it was cooler to completely ignore the competing product and instead merely state that their product was "cool". Of course "cool" is a ridiculous concept to the intelligent, but to the general public, they care, you're completely right. Linux users, many of which are geeks and don't care about "being cool", will have a hard time marketing to the "cool crowd", indeed. Linux, fortunately, has grown enough to grab attention from many who do care about being cool, and there are a lot of cool things about Linux.
Just so long as Linux development remembers to care about the really important things, like gaining developers and making development easier, making fundamental aspects of the OS easier for everyone (*ahem*installing drivers without compiling*cough*easy cross-distro program installation aka basic package portability), and in general making the average computer user's life easier (and the non-average user's life too), as well as being "cool", with your 3D desktop and your sleek black PS3s (which can barely do anything anyway due to hypervisor poop).
99% of the time however, taking the anti-Microsoft side happens to also be the intelligent/reasonable/correct side, and already comes with many reasons to back it up. Most people don't just hate Microsoft just because, they hate them for real reasons. But yeah, I got your point and agree. It's not just MS, it's Adobe and other pedlars of proprietary software that use tactics to try to push their software on students instead of being fair and using standards where possible.
Wow, maybe Firefox 1 and 2, but 3 seems to have done away with that problem. I leave Firefox 3 open all the time and never have an issue, and I'm running an oldish computer by today's standards (AMD Athlon 64, AGP GFX, etc).
The open source does-it-all collaborative suite programs like Exchange certainly need to be more competitive, but so far they are promising. So, maybe your Exchange-using organization should switch. Two "larger" looking ones I've found are:
:P
.NET + Evolution thing, seems like they might be pushing for it a bit, since Evolution is somewhat working with Exchange. I can at least get email as well as see calendar events and get warnings about them, but sending emails I haven't tried very hard to figure out how to do. Hopefully the push for that isn't some sort of patent backstabbing deal though. Regardless, I'd rather not deal with that patent troll if I can help it and would instead ask for something that at least used standards so that you could use what client you wanted to. Microsoft doesn't know the meaning of the term "standard", plus, you know, there's the whole hatred for Linux and just trying to pull everyone onto their platforms thing so they can "collect at a later date" - Bill Gates.
Zarafa
Zimbra
Both were just a simple google search away.
As for an Exchange client solution, well...difficult, since Microsoft controls the communication interface for that and I doubt at this stage they are interested in freely playing nice with non-Microsoft clients. In that case, your best bet is to change the server.
I take that back, they may be getting enough pressure for playing nice with clients due to the whole
That would be a very silly limitation for OOo to have, agreed, and one would think that would be an easy thing to fix, though that's definitely a big sheet...maybe someone should break things apart a little more.
Damn that sucks, I'd organise an appeal for the student body to have teachers adopt non-proprietary software. To be teaching proprietary software especially with things like this, a stupid office suite, and demanding it's use, is just all kinds of dumb, and I believe wrong for a college to do. And some people wonder why "student software" is so cheap, like a drug sample. No, students should have the freedom to choose which office software they want to use to complete an assignment.
Please point out areas in which GNU/Linux could use improvements, that will be a much more constructive post. It will help GNU/Linux to improve.
Most readers here find GNU/Linux to be a superior OS in most every way. It has some shortcomings, and they are great to address and talk about, since it's a community OS.
The licensing of GNU/Linux is very important and is what makes it special. The freedom to tinker and improve is a feature that outweighs most shortcomings in the minds of many readers here, but all the usefulness and benefits and perks of GNU/Linux of course really help too. Not to mention, you know, it's free. If you want to stay on the Microsoft upgrade treadmill and in it's OS prison though, be my guest.
So yeah, of course mocking many here with things like GNU/Linux leg humpers, etc, is going to get you flamed in return. Not a great way to start an intelligent debate.
And even stupider, who'd want to strengthen software patents and give Microsoft leverage by caving into their FUD and paying for this "protection" racket? All efforts to invalidate all software patents should be taken by all companies and citizens.
...as to why they don't pay royalties. Because they shouldn't. Fuck software patents.
I think moving to a classification system is the best answer to avoid having to have the gamer completely define everything themselves, so that the game developer can simply say "when the controller layout is like type Z, assign this to this and that to that" but of course allow the gamer to change those assignments if they want to. Or better yet, the dev says "I need a thumbstick on the left of the controller to be assigned to this, and the dpad to my dpad here, and the primary buttons on the right side defined to this, with any leftovers left undefined". That way they wouldn't even have to define what to do for every classification situation, if you could make the communication system intelligent enough so it could figure out what to do in most cases automatically, since that is the goal, to make it automatic for both the developers and the gamers with options for adjustments later if needed.
Yep, and devs need to start listening to users like you more and waking up to the bits that are difficult about Linux, and redo things or patch things or do whatever needs to be done to make Linux easier. Just seems like us users who mention things that aren't functional in Linux are ignored, sometimes because "that's the way things are" and "it can't be changed (easily)". It amazes me when anyone has the mentality that there are some huge barriers in software that prevent certain things from being accomplished. That's a cop-out, it's software, if you can think it you can do it. It's like, sometimes Linux users who learn Linux well enough to understand what the issues are and how things work at the same time become indoctrinated with the status quo, thereafter unwilling to think outside the (Linux) box to how things could be, and could be better, and how things could work.
OK, well you're a good teacher then, but we'll see if that knowledge becomes mainstream or not, maybe it will dunno, right now the terms "Linux" and "Ubuntu" are pretty big though...
Since X and Y are not equal, you have to find a way to map the events of X to Y. And thats where the trouble starts, since you don't know which combination of events in X are good to produce the events in Y
Right, so you'd need to break a controller down into it's component parts, and then register those parts with the API, so that a) a gamer can map any of those parts that they want to any of the buttons defined in any game easily, and b) provide a calibration program so that if the default settings are off for that controller, or if the controller isn't recognised by the "system" at all, you can still teach it.
But of course thats not all, input devices aren't necessarily clear about their capabilities, so if a device says it has 8 buttons, it might in reality have 4 buttons and a dpad or worse you have something like the WIZ which reports the dpad has 8 buttons, since the diagonals are handled separately. You need a big fat table of information to clean that up. But even if you have it cleared up you are still left with a lot of ambiguity, shall a emulator use the dpad or the analogstick, shall button 5 and 6 get mapped to L/R or to C/Z (gamepad with 6 face buttons), which gamepad shall the emulator use when there are multiple available and so on.
So you have specific default settings being mapped if the controller is being recognized, but if the "system" recognizes them incorrectly or doesn't at all or whatever, the joystick program that should come with KDE and Gnome should not only be able to calibrate, but change the input type. Say, change what came up as a "button" to a "trigger", if that is necessary. Once the inputs are all defined and calibrated, those standardized input classes that the system is now configured for can be easily passed to any game that uses the standardized API/system. In the game, you can then adjust those further as far as what buttons on X controller should be used for the Y controller for the gaming system.
So, what I'm saying then is desktop GNU/Linux needs to do as good of a job as it can to get things right, but then it should offer the ability to easily allow users to tweak and define what it detects, and to even make big adjustments if necessary like changing a sensed button to a trigger or whatnot. If it takes care of all of that part of it, developers of Linux games and other software won't have to deal with it themselves. You offload lots of work and headaches to the system developers, so that Linux development is easier for them and programs are easier to make.
Microsoft has invested a lot of interest in doing this, so that their OS is easier to program for. I'm not saying they were successful or anything about them, believe me I hate many many things Microsoft has done as much as the next informed Linux user, but the goal to make it easier for developers is a very good goal. It will help increase the adoption of Linux, not to mention make it easier for developers to flex their brains and creativity, which is exactly what an open OS should be focused on. I want to help make Linux more user friendly than Windows in every way possible, that will be how Linux succeeds on the desktop.
Some of this is fixable, but most is not, unless you actually hack the game. The problem is most obviously demonstrated when you look at PC to console conversions or the other way around, those that end up with a good interface need to basically redesign large part of it, since different input devices have just very different capabilities.
Yes, there has to be translation there I understand that and agree, but if you centralize the translating work to the OS, you simplify it for game developers. They can even have the capability of using the input device directly like the old ways, or to interface with the new standard. So here's a concept mock-up of what I'm saying:
Controller is plugged in >>>> Controller is either recog
Trying to explain to a normal computer user what a kernel is, why they should care, and what all the different "Linuxes" actually are, will confuse, not to mention be a very boring conversation for them. GNU/Linux needs some names to get behind to make it simple, Linux and Ubuntu are two that are becoming somewhat common.
Is all I'm saying. ^^