As others already have mentioned Xara has a superb GUI and is easy usable. This is due the fact that Xara tries to be wyoGuide conformant (see http://wyoguide.sf.net/projectlist.php).
It's amazing what's written on the net. I always used LowEndMac for information about low end Macs and liked it very much. Yet does this story mean LowEndMac wants to switch to provide information about low end PCs after Apple switched to Intel processors?
Yet this story raises many other questions. How does IBM feel being famous for the most used kind of desktop processors but not being able to participate in that business anymore? How does Sony feel now its long time partner in several technologies (Apple) doesn't use the same processor (PowerPC Cell processor) anymore? How does Intel feel now Microsoft switched to PowerPC processors for their Xbox? Curious world these days, what can we expect next?
But what interests me most is how evangelists feel after so many changes.
... but I haven't been impressed with the Windows GTK ports I've seen.
Neither have I but there's where wxWidgets shines. It might not look perfect on each platform but on any I've seen it looks acceptable. wxWidgets uses native win32 on Windows, Carbon/Cocoa on the Mac and GTK/X11/Lesstif on Linux/Unix. QT might be better but there's no OpenSource outside KDE. IMO there's no reasonable alternative to cross-platform development.
I have been writing software for 30 years, and I don't think writing good software is simple...
So do I, having written my first code in 1972. Yet I still believe writing good SW, better said usable SW, isn't that difficult, all it needs is some sort of common sense and common understanding. I've seen quite a lot of code during this time and while most of these codes were good structured and enough documented, the resulting applications were seldom usable. The problem isn't that the developers did a bad job, no they mostly didn't care for the same issues as the users.
I'm sure the article writer talks only about GUI software since that's what ordinary users see. There's no hint about command-line or kernel or e.g. apache modules. Besides I don't question the quality of these codes while much OSS GUI code is rather unusable. This is a speciality of OSS because with CSS your boss will force you to make your code not only good but also usable. Unfortunately there's no autority which forces OSS developers to make their code usable.
...I have mod points (same here)... still. Part of the problem is a lot of software, but very little *good* software
I'll one-up you. The major problem isn't there little good software, but very few good software engineers.
I wouldn't say it's the fault of the developers, it's just many don't know how to write good SW. Yet even that isn't their fault because they never were taught how to write good SW since even teachers and professors mostly don't know.
Writing good SW is quite simple all you have to do is follow some good guidelines, the tricky part is what or where are these guidelines. A few years ago when I started writing OSS I was faced with exactly this question and couldn't find any. So I decided to create my own guidelines : wyoGuide (http://wyoguide.sf.net/). Sure they might not be as good as they should be but so far they have suited me quite well.
but he's not really talking about software, is he? He's talking about Microsoftcompatibleware and Buzzware.
Well there are alway several oposite point of views about what's exciting SW. But when you look what the majority of the Linux users consider useful SW the article is just right. Read the OSDL survey about what users think of the Linux applications (http://www.osdl.org/dtl/DTL_Survey_Report_Nov2005 .pdf), especially about the first top inhibitor of the Linux desktop adoption.
However, that's going to require OSS to start thinking about polish - making applications that Grandma can use. It's not impossible, but a lot of OSS projects need to concentrate on making applications that work well and look decent on Windows - even if we don't particularly care for the platform or the company that makes it.
I've said it numerous times the way to develop OSS is to do it cross-platform. Audacity is a superb sample how it can be done. Yet many other OSS don't follow its path most probably because they don't know how. Especially for helping developing cross-platform OSS I've created wyoGuide (http://wyoguide.sf.net/).
There are two ways to make OSS more useful, one way is to look at the guidelines in text form and implement them as much as possible. This can be done with any framework and in any language, there's no restriction albeit the improvements will be limited to Linux. When true cross-platform functionality is needed there's IMO no way than to switch to wxWidgets and use the code samples of the wyoGuide guidelines as well.
IMO it's very important for the Linux desktop that free applications also can be tried on Windows. Only then users can get familiar with OSS and possibly switch to Linux later on.
There's IMO no question, Ubuntu will become the major distribution yet this will take some years. It's a slow process but IMO not reversible anymore. Ubuntu currently draws users from other distributions since as the article mentioned it, this distribution is done right!
I guess the first distribution which sees this loosing users by large numbers will be Debian. Ubuntu is simply far more convenient for the average user. Yet Debian might still stay alive through its vast developer community albeit its outcome isn't clear. RedHat and also Novell/Suse or similar distributions will notice a drop in ordinary users but the drop in corporate users will come far later. It takes some time until Ubuntu gets established in the corporate market. On the other side Ubuntu won't affect Gentoo or Slackware since their users aren't interested in convenient usable. In the end Ubuntu won't demise all others but it will finally become the major distribution.
Yet since Ubuntu has the power to draw users from other distributions the question arises if it also has the power to draw users from Microsoft Windows. I think not, at least not in large numbers. The problem is Ubuntu as a distribution is just one part in a desktop system. Ubuntu can only be as good as the Linux kernel is, as the current desktops and all the applications are. Yet I'm rather pessimistic about the quality especially of the application but also about the desktops.
First wxPython is just the Python interpretation of wxWidgets. There are many more bindings for almost any practical language. There's no need to use Python, you can use whatever you like.
Second all the nice features (e.g. cross-platform, native look) are because of wxWidgets which main language is C++. I don't think C++ is the nicest language yet when it comes to design and write a rather full featured and complex application, there's currently no other choice.
Third when you want write not only ordinary but excellent applications you should follow the wyoGuide guidelines (which fit perfectly with wxWidgets). I haven't read the book so I don't know if references and uses wyoGuide but if it doesn't I wouldn't recommend buying it.
[Sarcasm on] 1. Quick kill the managers who set the impossible schedules 2. Quick kill the developers who can't stay within a reasonable schedule [Sarcasm off]
It's impossible to change the development time outside of the -10%/+10% margins. If your schedule is wrong you either have incompetent managers, incompetent developers or both. What ever you do, work overtime, drop in more developers or else, all you gain is a few percents.
Development time is IMO a rather static variable, the only thing which can improve this time is education. Yet education has to be done in advance before the project is even started. During the project there's no time for education while the knowledge should be available right from the start. The easiest solution is to always plan for a 25 - 30% education phase at the beginning of your project.
A geek has the eagerness and willingness to spend the time learning something new, something different. A user is reluctant to learn anything new and avoids anything different. While something looks rather useful for a geek it's many times a complete mystery to the average user. Yet while geeks thinks something usable, they most of the time don't even understand the problem of a user.
Since usability is always subject to the personal knowledge and taste and since users are the vast majority, usability is mostly defined as usable for ordinary users and not for geeks.
because they're started by developers for developers -- people scratching their own itch -- tend to end up with very geeky products. They don't believe in marketing, they don't believe in the mainstream. They're supposed to be the anti-mainstream, right, so it's very hard for most open-source projects to break out of that mentality
Geeks want geeky products, users want usable products. Why can't OpenSource projects break out and make usable products. Live would be much better if at least some could overcome this barrier.
O. Wyss
Will Ext4 be state of the art?
on
EXT4 Is Coming
·
· Score: 1
Is Ext4 able to do integrity checks during ordinary use on the fly, allowing to get rid of the startup/access limit checks?
Is Ext4 able to correct minor discrepancies on the fly, as long as the involved blocks/nodes aren't accessed?
Does Ext4 have a log of major discrepancies which may be corrected in an unmounted state without performing full checks first?
Is Ext4 fail save (power loss) after a certain amount of time (less than 30 sec) of no access? In other words does a power failure have no effect on any block/node after the last access is older than this time limit?
Can Ext4 be used cross-platform, e.g. in a multi boot environment or virtual server with different systems?
IMO these are the requirements which a state-of-the-art file system should have these days. Creating and naming a few file system makes only sense if these requirements are full filled.
Look, I understand you want to promote wxWidgets...
I don't want to promote wxWidgets, I want to promote wyoGuide with the right tool. I don't care if you write a wyoGuide conform application with Java/Swing albeit I've no idea if it's possible and how much it costs. But when you start with an MFC source you're at least 10 times better off with wxWidgets than GTK.
there is no wxCDBurner or wxPicasaBlogIntegration or wxInternetExplorerOnTopOfMozilla class...
I'm curious why a photo imaging SW should burn CD's, all it may have to do is create ISO format through a free library. Yet I can't believe Picasa has blog SW integrate or ever need to know on which browser it work. Does this mean Picasa can't work with Safari or Opera?
Virtually no "native" apps on Linux are based on wxWidgets so there's no gain to be had from using that over Wines own toolkit implementation. The benefit of using native UI and guidelines is that it fits in with a particular desktop. WX doesn't fit in anywhere. And yes I know it sits on top of GTK but sorry, I've used WX apps and they didn't feel native at all.
I agree there aren't that many wxWidgets applications on Linux. Yet Linux applications are mostly written by OpenSource developers which are completely free to use whatever they like. They aren't pressed to deliver in a tight time frame nor are responsible if it doesn't work.
That more GTK applications are available is no surprise since GTK is far longer around. Yet there isn't a single GTK application which achieved a significant market share even because of the long time. Both Mozilla and OpenOffice use their own framework and the next possible candidate is Xara, a wxWidgets application.
I agree that neither wxWidgets application feels very native but that's because these aren't written wyoGuide conformant either. I can't and won't force any developer to use guidelines, the can and should do it for fun. Yet I want to give them a tool so they can write good application while still having fun.
I'd be completely surprised if there's a single third party library in Google Earth on Windows. It's impossible to give a way such an application unless Google paid for a world wide distribution of such a library. Yet it might be a free library but then this library is most probably also free on Linux.
Besides Audacity, which is a true cross-platform application, uses a multitude of free libraries without the time lag and Audacity is just a spare time development project without the resources Google could afford.
here's no fundamental reason why a Wine based app should look or feel different to a native app. If you want to port your app to Linux and have it look and feel like the most native open source program there is, it can be done.
Sure it can be done, yet the development cost aren't much lower than if a native application is built right away. So why not go all the way? I don't question the quality of Picasa yet a Wine solution is half hearted and sends the wrong signal out to the users.
This is clearly not true, many people have written positive reviews of Picasa for Linux.
I can only say, go to the Slashdot discussion about Picasa and read the comments yourself. I haven't made a complete survey but the overall impression isn't that good.
The sad truth is that Win32 is so deeply embedded in most apps that they will never be natively ported.
I can fully understand that it's a horrible task to change an MFC application into a GTK application. Yet it's rather easy to change MCF into wxWidgets which both are based on Win32 on Windows. So if the work on Picasa was too much for a fully native port is only because you might have chosen the wrong framework for such a task.
As it's possible to build a fully native Wine application it's possible to port any Windows application to Linux, it only depends on the involved costs. And the costs depends on if you choose the right solution for the task.
Because the Windows design and UI guidelines aren't actually that bad...
True. Also on Linux there's more or less all kind of UI design used, so when everything is different it doesn't feel that uncommon. Yet it's exactly this "not possible to become familiar" which is the top inhibitor of the Linux desktop adoptions (see http://www.osdl.org/dtl/DTL_Survey_Report_Nov2005. pdf).
Google doesn't release SW on Linux only to get users and customers, they also release them for their image as the white knight. Yet Picasa quite obviously hurts this image. Besides I bet Google Earth for Linux is a complete new development tree, meaning they have paid for it twice. Google may be able to do that but others have to earn money with their products.
How can anybody base his ordinary software on Wine. Wine is an emergency solution which should only be used to bring your desperately needed third party product to run on Linux for a limited time but never ever to sell your product. Any Wine application is still an ordinary Windows application following the Windows design and UI guidelines.
Just go back and look at the discussion about Google's Picasa here at Slashdot. No sensible person is satisfied with it, all it achieves is showing Google's incompetence to produce real Linux applications. Releasing a Wine solution just shows that Google capitulated from being able to build ordinary Linux applications.
Yet Corel doesn't do better than Google or any of the other vendors who don't sell Linux applications, they all don't know how to do cross-platform development efficiently. It's completely understandable that none want to pay for a second development line for a platform which hasn't more than a few percents market share. But this isn't needed if you do your development in true cross-platform development fashion (see wyoGuide).
But may we throw stones at the commercial vendors when we, the OpenSource community don't do better. Beside Mozilla and to some extend OpenOffice there isn't many true cross-platform application either. Please don't say an application is cross-platform when it builds or runs, it's only cross-platform when it's also used. That means when an application is sellable or is able to get above 10% market share.
Well then, why doesn't Novell sponsor the development of a financial application? They discovered the importance themselves (http://www.novell.com/coolsolutions/feature/16798 .html) while still selling SUSE Enterprise Linux. It's amazing how companies sometimes behave or don't take their business seriously.
There's no need to have everything in FOSS, yet FOSS is needed when commercial veldors don't release for a platform. Only good cross-platform FOSS can "encurage" vendors to release their products everywhere.
Why do you think doesn't Adobe release on Linux? Simply because its market share is just a few percents and GIMP is in their eyes no competition on their market segment Windows and MacOSX. Yet why did Microsoft finally rework IE6 into IE7? Just because Mozilla draws too much attention on Windows itself. This holds true for any commercial vendor, they simply have no reason to care for Linux as long as there isn't any competition.
Since there isn't a financial application on Linux (I don't know Moneydance so far) users who "need" or want one are pressed to buy a Windows system.
Yet the good news wyoGuide can be used by both FOSS and commercial Vendors, so it's quite likely the situation will change in a few years. The bad news it still takes a few years.
Slashdot Mirror
As others already have mentioned Xara has a superb GUI and is easy usable. This is due the fact that Xara tries to be wyoGuide conformant (see http://wyoguide.sf.net/projectlist.php).
O. Wyss
http://www.maa.org/editorial/mathgames/mathgames_0 8_01_05.html
Very good link. Unfortunatly most of the free solutions require either an additional GTK- or Java-Installation which is a killer argument for my use.
O. Wyss
Are there any screen shots of this GUI installation (links)?
O. Wyss
It's amazing what's written on the net. I always used LowEndMac for information about low end Macs and liked it very much. Yet does this story mean LowEndMac wants to switch to provide information about low end PCs after Apple switched to Intel processors?
Yet this story raises many other questions. How does IBM feel being famous for the most used kind of desktop processors but not being able to participate in that business anymore? How does Sony feel now its long time partner in several technologies (Apple) doesn't use the same processor (PowerPC Cell processor) anymore? How does Intel feel now Microsoft switched to PowerPC processors for their Xbox? Curious world these days, what can we expect next?
But what interests me most is how evangelists feel after so many changes.
O. Wyss
... but I haven't been impressed with the Windows GTK ports I've seen.
Neither have I but there's where wxWidgets shines. It might not look perfect on each platform but on any I've seen it looks acceptable. wxWidgets uses native win32 on Windows, Carbon/Cocoa on the Mac and GTK/X11/Lesstif on Linux/Unix. QT might be better but there's no OpenSource outside KDE. IMO there's no reasonable alternative to cross-platform development.
O. Wyss
I was never a KDE fan - I always liked Gnome stuff better, until I programmed in Qt ...
If you don't like coding with GTK for Gnome, why didn't you try wxWidgets?
O. Wyss
I have been writing software for 30 years, and I don't think writing good software is simple...
So do I, having written my first code in 1972. Yet I still believe writing good SW, better said usable SW, isn't that difficult, all it needs is some sort of common sense and common understanding. I've seen quite a lot of code during this time and while most of these codes were good structured and enough documented, the resulting applications were seldom usable. The problem isn't that the developers did a bad job, no they mostly didn't care for the same issues as the users.
I'm sure the article writer talks only about GUI software since that's what ordinary users see. There's no hint about command-line or kernel or e.g. apache modules. Besides I don't question the quality of these codes while much OSS GUI code is rather unusable. This is a speciality of OSS because with CSS your boss will force you to make your code not only good but also usable. Unfortunately there's no autority which forces OSS developers to make their code usable.
O. Wyss
...I have mod points (same here)... still. Part of the problem is a lot of software, but very little *good* software
I'll one-up you. The major problem isn't there little good software, but very few good software engineers.
I wouldn't say it's the fault of the developers, it's just many don't know how to write good SW. Yet even that isn't their fault because they never were taught how to write good SW since even teachers and professors mostly don't know.
Writing good SW is quite simple all you have to do is follow some good guidelines, the tricky part is what or where are these guidelines. A few years ago when I started writing OSS I was faced with exactly this question and couldn't find any. So I decided to create my own guidelines : wyoGuide (http://wyoguide.sf.net/). Sure they might not be as good as they should be but so far they have suited me quite well.
O. Wyss
but he's not really talking about software, is he? He's talking about Microsoftcompatibleware and Buzzware.
5 .pdf), especially about the first top inhibitor of the Linux desktop adoption.
Well there are alway several oposite point of views about what's exciting SW. But when you look what the majority of the Linux users consider useful SW the article is just right. Read the OSDL survey about what users think of the Linux applications (http://www.osdl.org/dtl/DTL_Survey_Report_Nov200
O. Wyss
However, that's going to require OSS to start thinking about polish - making applications that Grandma can use. It's not impossible, but a lot of OSS projects need to concentrate on making applications that work well and look decent on Windows - even if we don't particularly care for the platform or the company that makes it.
I've said it numerous times the way to develop OSS is to do it cross-platform. Audacity is a superb sample how it can be done. Yet many other OSS don't follow its path most probably because they don't know how. Especially for helping developing cross-platform OSS I've created wyoGuide (http://wyoguide.sf.net/).
There are two ways to make OSS more useful, one way is to look at the guidelines in text form and implement them as much as possible. This can be done with any framework and in any language, there's no restriction albeit the improvements will be limited to Linux. When true cross-platform functionality is needed there's IMO no way than to switch to wxWidgets and use the code samples of the wyoGuide guidelines as well.
IMO it's very important for the Linux desktop that free applications also can be tried on Windows. Only then users can get familiar with OSS and possibly switch to Linux later on.
O. Wyss
There's IMO no question, Ubuntu will become the major distribution yet this will take some years. It's a slow process but IMO not reversible anymore. Ubuntu currently draws users from other distributions since as the article mentioned it, this distribution is done right!
I guess the first distribution which sees this loosing users by large numbers will be Debian. Ubuntu is simply far more convenient for the average user. Yet Debian might still stay alive through its vast developer community albeit its outcome isn't clear. RedHat and also Novell/Suse or similar distributions will notice a drop in ordinary users but the drop in corporate users will come far later. It takes some time until Ubuntu gets established in the corporate market. On the other side Ubuntu won't affect Gentoo or Slackware since their users aren't interested in convenient usable. In the end Ubuntu won't demise all others but it will finally become the major distribution.
Yet since Ubuntu has the power to draw users from other distributions the question arises if it also has the power to draw users from Microsoft Windows. I think not, at least not in large numbers. The problem is Ubuntu as a distribution is just one part in a desktop system. Ubuntu can only be as good as the Linux kernel is, as the current desktops and all the applications are. Yet I'm rather pessimistic about the quality especially of the application but also about the desktops.
O. Wyss
First wxPython is just the Python interpretation of wxWidgets. There are many more bindings for almost any practical language. There's no need to use Python, you can use whatever you like.
Second all the nice features (e.g. cross-platform, native look) are because of wxWidgets which main language is C++. I don't think C++ is the nicest language yet when it comes to design and write a rather full featured and complex application, there's currently no other choice.
Third when you want write not only ordinary but excellent applications you should follow the wyoGuide guidelines (which fit perfectly with wxWidgets). I haven't read the book so I don't know if references and uses wyoGuide but if it doesn't I wouldn't recommend buying it.
O. Wyss
[Sarcasm on]
1. Quick kill the managers who set the impossible schedules
2. Quick kill the developers who can't stay within a reasonable schedule
[Sarcasm off]
It's impossible to change the development time outside of the -10%/+10% margins. If your schedule is wrong you either have incompetent managers, incompetent developers or both. What ever you do, work overtime, drop in more developers or else, all you gain is a few percents.
Development time is IMO a rather static variable, the only thing which can improve this time is education. Yet education has to be done in advance before the project is even started. During the project there's no time for education while the knowledge should be available right from the start. The easiest solution is to always plan for a 25 - 30% education phase at the beginning of your project.
O. Wyss
No. Read the last paragraphs of the original article and you know exactly what's meant.
O. Wyss
What is your definition of a 'geeky' product?
A geek has the eagerness and willingness to spend the time learning something new, something different. A user is reluctant to learn anything new and avoids anything different. While something looks rather useful for a geek it's many times a complete mystery to the average user. Yet while geeks thinks something usable, they most of the time don't even understand the problem of a user.
Since usability is always subject to the personal knowledge and taste and since users are the vast majority, usability is mostly defined as usable for ordinary users and not for geeks.
O. Wyss
because they're started by developers for developers -- people scratching their own itch -- tend to end up with very geeky products. They don't believe in marketing, they don't believe in the mainstream. They're supposed to be the anti-mainstream, right, so it's very hard for most open-source projects to break out of that mentality
Geeks want geeky products, users want usable products. Why can't OpenSource projects break out and make usable products. Live would be much better if at least some could overcome this barrier.
O. Wyss
Is Ext4 able to do integrity checks during ordinary use on the fly, allowing to get rid of the startup/access limit checks?
Is Ext4 able to correct minor discrepancies on the fly, as long as the involved blocks/nodes aren't accessed?
Does Ext4 have a log of major discrepancies which may be corrected in an unmounted state without performing full checks first?
Is Ext4 fail save (power loss) after a certain amount of time (less than 30 sec) of no access? In other words does a power failure have no effect on any block/node after the last access is older than this time limit?
Can Ext4 be used cross-platform, e.g. in a multi boot environment or virtual server with different systems?
IMO these are the requirements which a state-of-the-art file system should have these days. Creating and naming a few file system makes only sense if these requirements are full filled.
O. Wyss
Look, I understand you want to promote wxWidgets ...
...
I don't want to promote wxWidgets, I want to promote wyoGuide with the right tool. I don't care if you write a wyoGuide conform application with Java/Swing albeit I've no idea if it's possible and how much it costs. But when you start with an MFC source you're at least 10 times better off with wxWidgets than GTK.
there is no wxCDBurner or wxPicasaBlogIntegration or wxInternetExplorerOnTopOfMozilla class
I'm curious why a photo imaging SW should burn CD's, all it may have to do is create ISO format through a free library. Yet I can't believe Picasa has blog SW integrate or ever need to know on which browser it work. Does this mean Picasa can't work with Safari or Opera?
Virtually no "native" apps on Linux are based on wxWidgets so there's no gain to be had from using that over Wines own toolkit implementation. The benefit of using native UI and guidelines is that it fits in with a particular desktop. WX doesn't fit in anywhere. And yes I know it sits on top of GTK but sorry, I've used WX apps and they didn't feel native at all.
I agree there aren't that many wxWidgets applications on Linux. Yet Linux applications are mostly written by OpenSource developers which are completely free to use whatever they like. They aren't pressed to deliver in a tight time frame nor are responsible if it doesn't work.
That more GTK applications are available is no surprise since GTK is far longer around. Yet there isn't a single GTK application which achieved a significant market share even because of the long time. Both Mozilla and OpenOffice use their own framework and the next possible candidate is Xara, a wxWidgets application.
I agree that neither wxWidgets application feels very native but that's because these aren't written wyoGuide conformant either. I can't and won't force any developer to use guidelines, the can and should do it for fun. Yet I want to give them a tool so they can write good application while still having fun.
O. Wyss
Third party library dependencies can old you off.
I'd be completely surprised if there's a single third party library in Google Earth on Windows. It's impossible to give a way such an application unless Google paid for a world wide distribution of such a library. Yet it might be a free library but then this library is most probably also free on Linux.
Besides Audacity, which is a true cross-platform application, uses a multitude of free libraries without the time lag and Audacity is just a spare time development project without the resources Google could afford.
O. Wyss
here's no fundamental reason why a Wine based app should look or feel different to a native app. If you want to port your app to Linux and have it look and feel like the most native open source program there is, it can be done.
Sure it can be done, yet the development cost aren't much lower than if a native application is built right away. So why not go all the way? I don't question the quality of Picasa yet a Wine solution is half hearted and sends the wrong signal out to the users.
This is clearly not true, many people have written positive reviews of Picasa for Linux.
I can only say, go to the Slashdot discussion about Picasa and read the comments yourself. I haven't made a complete survey but the overall impression isn't that good.
The sad truth is that Win32 is so deeply embedded in most apps that they will never be natively ported.
I can fully understand that it's a horrible task to change an MFC application into a GTK application. Yet it's rather easy to change MCF into wxWidgets which both are based on Win32 on Windows. So if the work on Picasa was too much for a fully native port is only because you might have chosen the wrong framework for such a task.
As it's possible to build a fully native Wine application it's possible to port any Windows application to Linux, it only depends on the involved costs. And the costs depends on if you choose the right solution for the task.
O. Wyss
Because the Windows design and UI guidelines aren't actually that bad ...
. pdf).
True. Also on Linux there's more or less all kind of UI design used, so when everything is different it doesn't feel that uncommon. Yet it's exactly this "not possible to become familiar" which is the top inhibitor of the Linux desktop adoptions (see http://www.osdl.org/dtl/DTL_Survey_Report_Nov2005
O. Wyss
... released Google Earth 4 on Linux ...
Google doesn't release SW on Linux only to get users and customers, they also release them for their image as the white knight. Yet Picasa quite obviously hurts this image. Besides I bet Google Earth for Linux is a complete new development tree, meaning they have paid for it twice. Google may be able to do that but others have to earn money with their products.
O. Wyss
How can anybody base his ordinary software on Wine. Wine is an emergency solution which should only be used to bring your desperately needed third party product to run on Linux for a limited time but never ever to sell your product. Any Wine application is still an ordinary Windows application following the Windows design and UI guidelines.
Just go back and look at the discussion about Google's Picasa here at Slashdot. No sensible person is satisfied with it, all it achieves is showing Google's incompetence to produce real Linux applications. Releasing a Wine solution just shows that Google capitulated from being able to build ordinary Linux applications.
Yet Corel doesn't do better than Google or any of the other vendors who don't sell Linux applications, they all don't know how to do cross-platform development efficiently. It's completely understandable that none want to pay for a second development line for a platform which hasn't more than a few percents market share. But this isn't needed if you do your development in true cross-platform development fashion (see wyoGuide).
But may we throw stones at the commercial vendors when we, the OpenSource community don't do better. Beside Mozilla and to some extend OpenOffice there isn't many true cross-platform application either. Please don't say an application is cross-platform when it builds or runs, it's only cross-platform when it's also used. That means when an application is sellable or is able to get above 10% market share.
O. Wyss
Well then, why doesn't Novell sponsor the development of a financial application? They discovered the importance themselves (http://www.novell.com/coolsolutions/feature/16798 .html) while still selling SUSE Enterprise Linux. It's amazing how companies sometimes behave or don't take their business seriously.
O. Wyss
Why does everything need to be FOSS?
There's no need to have everything in FOSS, yet FOSS is needed when commercial veldors don't release for a platform. Only good cross-platform FOSS can "encurage" vendors to release their products everywhere.
Why do you think doesn't Adobe release on Linux? Simply because its market share is just a few percents and GIMP is in their eyes no competition on their market segment Windows and MacOSX. Yet why did Microsoft finally rework IE6 into IE7? Just because Mozilla draws too much attention on Windows itself. This holds true for any commercial vendor, they simply have no reason to care for Linux as long as there isn't any competition.
Since there isn't a financial application on Linux (I don't know Moneydance so far) users who "need" or want one are pressed to buy a Windows system.
Yet the good news wyoGuide can be used by both FOSS and commercial Vendors, so it's quite likely the situation will change in a few years. The bad news it still takes a few years.
O. Wyss