SEP
8
2004

KDElibs/win32 Introduced

Yes, first part of my QKW patches are merged with first few (most hardcore/lowlevel) KDElibs directories. Beside Kexi, that's what I am playing with since march 2003. Everyone is invited to contribute.

More information

Comments

This is so cool.
KDE on any UNIX, recently KDE also runs natively on Mac OS X, and now we are on the way to have KDE on Windows.
You rock ! :-)

Alex


By Alexander Neundorf at Wed, 09/08/2004 - 20:09

I tested Kexi beta on Windows 2000, and it was very stable and responsive. I have tested Kexi development versions from KDE CVS, and cvs versions have not been this stable. The stability looks _very_ promising.

I'd use some KDE apps on Windows sometimes, but using Mingw is complicated. Last time I tried, KDE desktop did not start on Mingw, a year ago it did not support finnis keyboard layout etc.

This is one great example of KDE flexibility. This might kick some Microsoft ass in it's own ground :)

I can't wait to be able to run Umbrello on Windows, I _really_ need a nice UML tool at classrooms where there's no Linux workstations. I'd also use Kate and Quanta on Win there, if possible.

Hopefully free apps on Windows will make transition from Windows to Linux / FLOSS easier and more tempting.

Eleknader


By Tapio Kautto at Fri, 09/10/2004 - 09:19

i am sorry you belive the lie.

Linux software on windows has absolutely no bearing on the adoption on linux. There has never been evidence that porting software from platform A to platform B will increace adoption of platform A. In fact we have quite a mountain of evidence of the above helping the death of platform A.

If what you say is true, shouldn't people be flocking in droves from win32 because of OOo? or Mozilla? Word Perfect?, Eclipse? or Apache? The sad truth is every KDE application ported to windows will keep yet another user from ever using KDE. If you are cool with that got nuts, I for one would rather just concentrait on making KDE a killer destination for developers and users.

Please do not confuse a linux user who wants their pet app on some platform where they are forced to use, with someone who is too lazy to move off of windows, and too cheap to buy software.


By Ian Reinhart Geiser at Fri, 09/10/2004 - 11:20

I have to completely agree with you.

The fact is that Windows environments are so wrapped up in what people call standard software, then if they are running Windows then they are running MS Office, Rational Rose, Photoshop etc. If I am moving away from these apps, then I am already moving away from Windows.

Free alternatives on Windows aren't going to make any difference whatsoever, and then there is the question of providing apps that aren't really integrated with the environment and an absolute pain to maintain. Yes, cross-platform development is important, but you have to look at where it is appropriate and on what scale you do it.

I think some major apps on Windows are important, like Open Office, Firefox and Thunderbird, but even then, medium to long-term you're looking at ditching Windows whether you realise it or not because you simply no longer need it. Beyond these major applications, there is absolutely no point spending time, effort, sweat and tears porting to Windows. It is pointless.


By segedunum at Fri, 09/10/2004 - 21:40

Why make life difficult for those who use multiple platforms? The point FOSS should not be to cause an operating system revolution, but to provide freedom to everyone, no matter what platform they use.


By jtjumper at Sat, 01/26/2008 - 19:32

1. I believe what you mentioned is not quite a point for me.
It's not enough to mention OOo, Mozilla, Firefox. These apps are considered as a replacement for existing and well recognized Microsoft's offerings.

Do you know any other way to enable migration for office users and individual users other than allowing them (possibly slow but continouos) painless migration to FOSS apps? What is needed to do this? People need to have removed most barriers that force them to stick with Windows-only apps. These barriers are: high-level file/protocols available only for selected OSes (currently many specialized not ported KDE apps have these formats).

2. You wrote: Linux software on windows has absolutely no bearing on the adoption on linux. There has never been evidence that porting software from platform A to platform B will increace adoption of platform A. In fact we have quite a mountain of evidence of the above helping the death of platform A.

Maybe just in the USA? I've read many reports that desktop Linux is gaining momentum worldwide, including governments (e.g. see linuxtoday.com archives), software developers are more frequently picks FOSS tools instead of MS tools. .NET and Longhorn is not doing well jsut because it inherently use the reules you mentioned are so good.

Which destroyed platform do you have in mind? OS/2 --stupid marketing and badly prepared Microsoft/IBM deal. Fall of some flavours of Unix --this is mostly dependent on overpriced hardware. CP/M? OS/400?

Or perhaps you have RDBMS in mind? Again: Sybase? -stupid deal with MS, so sublicensed engine (MSSQL) is now more mainstream than the original Sybase product.

PHP/Apache/Perl/Python is a king of web scripting, all also available on win32. So why there is defined LAMP term and L==Linux? Because more attention is paid for using all these tool on Linux, as they are considered as inherently Linux tools. ASPs have chosen to agree with this. (and e.g. with Kexi, KDevelop, Umbrello etc. I believe ISVs will chose to agree as well).

I think your way of forcing people to move to Linux could work only if you were large vendor and have influence people, governments, and so on.

3. Could we really call this "Freedom", what you proposed? Authors and users of partcular great app (eg. KOrganizer, Umbrello) asked me when they will be able to use these apps everwhere they want (even on proprietary MacOSX) so to enable switching to FOSS world. Don't call them ignorants. Should we tell them not to do so because they are bad folks that didn't remove theirs win32 partition? So thus do you agree more with GNOME people saying that KDE 'is less free' because of LGPL?

4. What can be done to limit a danger of what you are afraid? Wy don't make win32 platform more expensive than Linux? How? E.g. within my company, I've started selling Qt/KDE apps with support, addons, etc. for win32/linux (and sometimes macosx). Why not to work together in this field as well and make Linux offerings at least a bit cheaper in terms of support costs (or attractive in other way). Why don't tell win32-customers everytime that they will be paying less for Linux version of the app/service next time when they come back to you?
You are free to do just that. And this is a strategy, after all.

We're using LGPL! We're free to add an exception to and newly added software so people won't be able to use it on win32 in so compfortable way as on FOSS oses. Say to Trolltech they are not doing good work this way.

We all have different experience, but let's don't send a signals to the world we're figting within the community.


By Jarosław Staniek at Sun, 09/12/2004 - 11:40

I really can't show you hard evidence showing, that FLOSS applications would lead or drive Windows users to Linux. That would not be freedom, now, would it? It works another way. If I was an IT Manager of a traditional Windows shop concidering move to Linux, I'd use steps like this:

1.Start using applications that can be run on both platforms (too bad those KDE apps can't be run here, luckily OO.org and Mozilla can so let's try them out). OK, we found out that those hacker-made applications actually work and they're usable.
2.Start tests of different Linux distributions to find out how well they support our hardware, how our IT support handles it etc. OK, this Distro X works and they support us if we need support.
3.Test those Windows apps we really need to run on Linux. Ok, our CRM tool runs on Wine/Win4Lin/or something.
4.Start a test group which uses Linux everyday. Ok, 8 weeks and they get their work done.
5.Gather experiences and make decisions of full OS change. Ok, we need to educate end users a bit, but we think it's worth while.

The way I see it: IT people want to test things carefully and slowly. IT Managers do not want to change everything overnight, it would be too risky. After all, they are providing software and hardware that supports the business, if they fail that they're out of job. If you provide no small steps, you'll prevent some Linux adoption. Sure, there might be some people that will be happy running Mozilla and Apache on Windows, that's their freedom. Some of them realize that Apache/MySQL/PHP might run easier/better on Linux, they'll move to Linux if they can. I do not see this as a reason not to support Windows at all, I'd say it would be nice to support Windows, but not fully: for example release new or full versions for Linux only. Or do it by prizing, sell/support Linux versions with smaller cost.

For developers I'd imagine following advatages for porting applications to different platforms:
1.You'll get wider audience
2.You'll find and fix some bugs
3.You'll chance your code to be more portable and hopefully better encapsulated
4.It's fun :)

Anyway, thanks for your comment and I hope you'll find some food for thought from my answer :)

Tapio


By Tapio Kautto at Mon, 09/13/2004 - 11:44

...you completely misunderstand how business works. I know we have ideals that as programmers we think make logical sense, but they are wrong.

Fact: its easier and cheaper to do nothing.
This means unless a company is in unbearable pain they will change nothing. It won't matter if something is better, free etc. Unless they are in absolutely unbearable pain they will do nothing, or at least as little as possible. IE the argument of partial migrations are bunk. There is no proof, NADA NONE. Sorry you lost on this one, case closed. (Honest its closed or I will bury in examples of how this has failed about 15 noted time in the last 20 years)

Fact: introduction of a product into a already crowded market is nearly impossible.
This has been documented in almost every modern marketing textbook out there. If you wish to introduce anything into a market, you must first evaluate the competition. If you are one of many alternitaves then you have to be so remarkably different and better than the competition that you redefine the market. If we pair Kexi if Access its obvious how that will work out. Likewise OOo and Mozilla have had a marginal impact, because they are so much more notable than the competition. Basicly for a KDE application to compete on windows, it either must be so amazingly much better than anything else in the market, or a one of the kind application.

I hope you can see the lie. It's neat for js, cause ideally he'll make some money out of the Kexi port, but for KDE in general a windows port is infinitely more damaging to KDE/Linux than it will do good.


By Ian Reinhart Geiser at Tue, 09/14/2004 - 01:21

At this division were running MDK Linux & HPunix 11.0 with over a 150+ programs written in house and compiled for them all with a single source. They now want to use those program on WinXp and still maintain them under Linux/Unix. This would allow me to do just that with out having to maintain 2 version of source code. It is coming down to moving this division to WinXp if we can't maintain a single source code version. So yes being able to run Linux/Unix programs under windows does have an advantage to help Linux grow. I have tried to install MinGw but I can get past its installation bugs.


By Robert Courtright at Tue, 03/22/2005 - 15:30

Qt runs on Linux and WindowsXP. Why would running unix apps on windows do anything more than keep you on windows? Port your windows apps to Qt and call it a day.


By Ian Reinhart Geiser at Wed, 03/23/2005 - 14:19

Pages