Any X application on OS X stands out like dogs balls and looks like shit. Sure, OS X has a built in X server, but you really don't want to be running X applications unless there's no real alternative.
I say that as an OS X and BSD / Linux user. Imagine running motif apps on your shiny new gnome install and you'll understand what i'm talking about.
The most basic change sometimes involved rewriting major subsystems, and tons of projects ended up being abandoned because they were practically impossible to incorporate into KDE 3.5.
end user = don't care
harsh? yes, at the end of the day, it if works and i can get my job done using it, the end user doesn't care how it works under the bonnet.
If it is KNOWN TO BREAK on certain combinations of hardware/driver then the sensible thing to do would be to fucking DISABLE THE FEATURE (or work around it, in a sub-optimal manner) on such hardware, no?
Not simply wait for bug report, pass the buck and piss off your userbase.
As the GP said, the end user doesn't give a shit WHY it broke. If it broke while he was running X piece of software and is stable running something else, he's going to run something else. Especially if the response he gets is a "not our problem"
I've used KDE between 1.0 (compiled on slackware back in the day) and 3.5.1. With brief, painful forays into 4.x during the past couple of years.
KDE4 is a turd. There's no other word for it. I am what one would consider a power user, and the breakage of my workflow between 3.5.1 (which i picked up without needing any degree of reading of documentation) and 4.x, combined with the truly unstable/glitchy nature of every 4.x release I have tried is totally unacceptable to me.
I am not alone.
Earlier versions of KDE were revolutionary and truly pleasant to use in the way I could simply get shit done.
KDE4 is/was painful and unstable to boot.
I don't give a fuck about plasmoids and shiny desktops. They broke core functionality.
The more abstraction, the less you NEED to know to build a functional app. The smaller the individual components can be, and as a result, the easier they are to debug in isolation.
Badly coded abstraction with no logging facility or spaghetti code may be bad to debug, but if you have an appropriate level of sanity in your code with a properly documented API there is no reason it needs to be difficult to debug.
Further, if a bug IS found, and the API is stable, then there a shitload less code that needs to be re-written, lessening the likelihood of NEW bugs being introduced in the process.
OS X exists because it's pretty bloody good at what it does. Unfortunately, gnome does very little of the important aspects of this, despite copying vast amounts of the GUI window dressing.
Attention gnome developers: The OS X GUI is pretty shit in a lot of places! The GUI is NOT necessarily why a lot of power users actually like it (I happen to like it DESPITE a lot of the GUI choices). We like it because of things like filesystem metadata, apps that automatically become aware of things that other apps do, or changes to the filesystem. Things like open standards support for zero configuration networking, excellent hardware support (given that you're restricted to certain hardware), sensible defaults, easy to use tools to configure things, etc.
Copying the global menu, the window frame style and icons is merely scratching the surface of what makes OS X - and its the more important stuff that has nothing to do with the GUI that makes it awesome.
Replicating that superficial crap without bothering to make things work anything like OS X under the covers just makes you look like a cheap imitation that will fool the uninitiated.
That said. I wasn't intending to have a rant at the state of the unix desktop, but there we are. The world is crying out for something that actually works (and etoile shows promise) but so many people are caught up with the latest eye-candy that we're progressing towards that at a glacial rate.
And before someone tells me to write the code myself - I'm happy to just buy it if it works. Its simply a shame that the direciton of free software is caught up with so much of the "shiny" superficial bullshit that doesn't actually matter.
I agree with you on windowmaker config, but thats basically because development has been pretty stagnant since the early 2000s. I used it as my primary desktop in 1996-2003, and installed it again the other day to see what had changed. Very little.
Its a good product, but its only a Window manager, not a desktop platform. I'm keen to see the actual important stuff from Next like the openstep framework actually get implemented properly.
Infinitely customizable flashy desktops are nice and all, but back in the real world people just want to get shit done in the most efficient manner possible.
There needs to be more focus on this sort of thing than the newest imitation of some feature Microsoft implemented 5 years ago, or NEXT implemented 18 years ago.
Give me a usable scripting interface that I can point/click on (like say, automator). Build in support into every app so that they can communicate with the system-wide scripting engine.
Fix bugs. Make hardware (and software) less painful to configure. Design a good, extensible API and for fucks sake, stick to it - rather than changing things and breaking compatability every 3-5 years "just because".
Its all about the APPS and what you can do with them - the desktop environment is merely window dressing to enable you to interact with the tool you are using.
I'm quite sure there are plenty of older Linux peeps out there who are of a similar mindset.
Again, creating a flashy desktop is fine, if thats what you want to do. But you're never going to win substantial market share with it.
Although long-winded, the post has a point. Anyone who has ever tried to set up any sort of non-basic authentication on a unix box doing anything other than really basic desktop shit will have run into similar problems. Getting the OS to log you in using LDAP is the easy part. Getting all the related services to use LDAP (and doing a full audit before hand without missing anything to ensure that everything can be configured for LDAP) is not.
Maybe its simpler if you're starting out from scratch and have no network to manage a transition for, but if you have existing services to migrate - good luck!
VOIP is a market in which microsoft does NOT have a stranglehold. They are competing with Google on this, and to completely fuck Skype up in the face of cross-platform Google talk would be a rather foolish idea.
Microsoft don't clamp down on a market until they dominate it.
I mean, they can't make it much worse - the Skype client as it currently stands, is malware.
I'm not saying apps *should* be doing that. However, like it not, there are plenty out there that do. And no doubt plenty of people running such apps, without knowing that they will in fact break on version number update. This change will break them, for no real good reason (such as major new functionality).
Its free. Not open source, but its as free as skype was, and not "borged". A proper FOSS alternative (at least with any form of market penetration, so you know... you can actually talk to other users) doesn't really exist.
Seriously, some people need to realise that microsoft is a BUSINESS. Asterisk = compete with Lync. Skype = now microsoft owned. Why the hell would they continue development of one of their products to help kill another one of their products?
Is this crap for asterisk / asterisk users? Yes. However Microsoft would have a hell of a lot of explaining to do to their shareholders if they were to continue killing the market for their own product(s) by enabling/maintaining it.
This may be the case. However exposing their broken-ness for the sake of it, rather than some actual new feature is just plain idiotic. Business doesn't care why things suddenly don't work - just that they were working just fine, and suddenly they aren't. This shit would never fly in BSD land.
All that will happen is that the new kernel release will break a heap of shit (again, for no good reason), people will back it out and it will gain a similar rep to Windows vista for being bad news and a risk to deploy.
Just look what happened when oracle rebadged java by removing all references to sun - again for no real good reason...
So because GIT is broken, we're going to change the major version number of the kernel, and possibly break an untold number of apps that check the version number for compatibility validation purposes?
Slashdot Mirror
Any X application on OS X stands out like dogs balls and looks like shit. Sure, OS X has a built in X server, but you really don't want to be running X applications unless there's no real alternative.
I say that as an OS X and BSD / Linux user. Imagine running motif apps on your shiny new gnome install and you'll understand what i'm talking about.
I really can't see why there isn't more BSD support behind gnustep/etoile.
Evidently, the testing was "w00t, i can get a desktop login and plasmoids work!"
end user = don't care
harsh? yes, at the end of the day, it if works and i can get my job done using it, the end user doesn't care how it works under the bonnet.
If it is KNOWN TO BREAK on certain combinations of hardware/driver then the sensible thing to do would be to fucking DISABLE THE FEATURE (or work around it, in a sub-optimal manner) on such hardware, no?
Not simply wait for bug report, pass the buck and piss off your userbase.
As the GP said, the end user doesn't give a shit WHY it broke. If it broke while he was running X piece of software and is stable running something else, he's going to run something else. Especially if the response he gets is a "not our problem"
I've used KDE between 1.0 (compiled on slackware back in the day) and 3.5.1. With brief, painful forays into 4.x during the past couple of years.
KDE4 is a turd. There's no other word for it. I am what one would consider a power user, and the breakage of my workflow between 3.5.1 (which i picked up without needing any degree of reading of documentation) and 4.x, combined with the truly unstable/glitchy nature of every 4.x release I have tried is totally unacceptable to me.
I am not alone.
Earlier versions of KDE were revolutionary and truly pleasant to use in the way I could simply get shit done.
KDE4 is/was painful and unstable to boot.
I don't give a fuck about plasmoids and shiny desktops. They broke core functionality.
Not necessarily.
The more abstraction, the less you NEED to know to build a functional app. The smaller the individual components can be, and as a result, the easier they are to debug in isolation.
Badly coded abstraction with no logging facility or spaghetti code may be bad to debug, but if you have an appropriate level of sanity in your code with a properly documented API there is no reason it needs to be difficult to debug.
Further, if a bug IS found, and the API is stable, then there a shitload less code that needs to be re-written, lessening the likelihood of NEW bugs being introduced in the process.
CPU/ram is cheap, getting more powerful per $ at an exponential rate.
or just move onto something better and contribute to etoile
OS X exists because it's pretty bloody good at what it does. Unfortunately, gnome does very little of the important aspects of this, despite copying vast amounts of the GUI window dressing.
Attention gnome developers: The OS X GUI is pretty shit in a lot of places! The GUI is NOT necessarily why a lot of power users actually like it (I happen to like it DESPITE a lot of the GUI choices). We like it because of things like filesystem metadata, apps that automatically become aware of things that other apps do, or changes to the filesystem. Things like open standards support for zero configuration networking, excellent hardware support (given that you're restricted to certain hardware), sensible defaults, easy to use tools to configure things, etc.
Copying the global menu, the window frame style and icons is merely scratching the surface of what makes OS X - and its the more important stuff that has nothing to do with the GUI that makes it awesome.
Replicating that superficial crap without bothering to make things work anything like OS X under the covers just makes you look like a cheap imitation that will fool the uninitiated.
That said. I wasn't intending to have a rant at the state of the unix desktop, but there we are. The world is crying out for something that actually works (and etoile shows promise) but so many people are caught up with the latest eye-candy that we're progressing towards that at a glacial rate.
And before someone tells me to write the code myself - I'm happy to just buy it if it works. Its simply a shame that the direciton of free software is caught up with so much of the "shiny" superficial bullshit that doesn't actually matter.
*sigh*
I agree with you on windowmaker config, but thats basically because development has been pretty stagnant since the early 2000s. I used it as my primary desktop in 1996-2003, and installed it again the other day to see what had changed. Very little.
Its a good product, but its only a Window manager, not a desktop platform. I'm keen to see the actual important stuff from Next like the openstep framework actually get implemented properly.
And that's the point, kids.
Infinitely customizable flashy desktops are nice and all, but back in the real world people just want to get shit done in the most efficient manner possible.
There needs to be more focus on this sort of thing than the newest imitation of some feature Microsoft implemented 5 years ago, or NEXT implemented 18 years ago.
Give me a usable scripting interface that I can point/click on (like say, automator). Build in support into every app so that they can communicate with the system-wide scripting engine.
Fix bugs. Make hardware (and software) less painful to configure. Design a good, extensible API and for fucks sake, stick to it - rather than changing things and breaking compatability every 3-5 years "just because".
Its all about the APPS and what you can do with them - the desktop environment is merely window dressing to enable you to interact with the tool you are using.
I'm quite sure there are plenty of older Linux peeps out there who are of a similar mindset.
Again, creating a flashy desktop is fine, if thats what you want to do. But you're never going to win substantial market share with it.
Although long-winded, the post has a point. Anyone who has ever tried to set up any sort of non-basic authentication on a unix box doing anything other than really basic desktop shit will have run into similar problems. Getting the OS to log you in using LDAP is the easy part. Getting all the related services to use LDAP (and doing a full audit before hand without missing anything to ensure that everything can be configured for LDAP) is not.
Maybe its simpler if you're starting out from scratch and have no network to manage a transition for, but if you have existing services to migrate - good luck!
VOIP is a market in which microsoft does NOT have a stranglehold. They are competing with Google on this, and to completely fuck Skype up in the face of cross-platform Google talk would be a rather foolish idea.
Microsoft don't clamp down on a market until they dominate it.
I mean, they can't make it much worse - the Skype client as it currently stands, is malware.
I'm not saying apps *should* be doing that. However, like it not, there are plenty out there that do. And no doubt plenty of people running such apps, without knowing that they will in fact break on version number update. This change will break them, for no real good reason (such as major new functionality).
Like HURD?
well they can't really make it much worse. lack of platform availability for MALWARE is a feature.
Also... google talk is Jabber. Any open source jabber client will work. So actually, it kinda IS open-source.
Its free. Not open source, but its as free as skype was, and not "borged". A proper FOSS alternative (at least with any form of market penetration, so you know... you can actually talk to other users) doesn't really exist.
Google talk? Its a jabber server?
Seriously, some people need to realise that microsoft is a BUSINESS. Asterisk = compete with Lync. Skype = now microsoft owned. Why the hell would they continue development of one of their products to help kill another one of their products?
Is this crap for asterisk / asterisk users? Yes. However Microsoft would have a hell of a lot of explaining to do to their shareholders if they were to continue killing the market for their own product(s) by enabling/maintaining it.
Windows apps often can't handle version number changes either. So, good luck!
So you propose that all the legacy hardware support from prior to 1999 is dumped too?
This may be the case. However exposing their broken-ness for the sake of it, rather than some actual new feature is just plain idiotic. Business doesn't care why things suddenly don't work - just that they were working just fine, and suddenly they aren't. This shit would never fly in BSD land.
All that will happen is that the new kernel release will break a heap of shit (again, for no good reason), people will back it out and it will gain a similar rep to Windows vista for being bad news and a risk to deploy.
Just look what happened when oracle rebadged java by removing all references to sun - again for no real good reason...
So because GIT is broken, we're going to change the major version number of the kernel, and possibly break an untold number of apps that check the version number for compatibility validation purposes?