Linux Desktop Deployment Postmortems?
duffbeer703 asks: "My employer runs alot of desktop and laptop computers -- something in the neighborhood of 40,000 PCs. Currently they are all Windows 2000 & XP managed by Active Directory and other big, complicated enterprise management tools, all of which can support Linux in one form or another. I'm looking for ways of making Linux (and maybe Unix or even Apple desktops) an option as we replace or add PCs. The problem is, most of the resources that you find online about deploying Linux focuses on server environment, and the articles that I do find about desktop Linux focus on standalone developer workstations, the IBM conversion to Linux (which doesn't seem to have happened) or things like LTSP, that won't integrate well with our infrastructure. Is anyone out there successfully using Linux for regular users? How did it go, and how did your IT and user communities adapt to the new kid on the block?"
They converted to Linux in 1999.
no answers yet, not a good sign!
As a former contributor and developer on the GNOME architecture for many years I can say that GNOME is in no way ready to serve as the corporate desktop. There are simply to many issues inside and around the entire GNOME movement that should be mentioned here.
First of all GNOME has a very broken development framework with a lot of fragmentation. A lot of libraries are not working properly enough even in stable releases to give users a full working desktop environment. A lot of stuff are simply not working properly and a lot of stuff simply look too far disharmonic to be usefull. Not to speak about the poorly written third party applications that exists that don't serve any corporate needs.
From a developers view I believe that GNOME has reached a dead end where scalability isn't possible anymore. People have realized that with the C languge there is no progress and thus decided to code under Python, C++, Java, Ruby or MONO. But personally I believe that having a mature GNOME desktop these days require you to have Python, MONO, Java running next to your regular application, which makes it hard to have all of them incooperate correctly (to work correctly). This is not the problem of having different languages laying around or running in the background but more architectual nature as soon as it comes to bugtracking, feedback, expandability etc. Many bindings are not well implemented and have a lot of attributes not correctly defined which makes applications look and behave differently.
As example I always get back to the legendary Toolbar issues that I like to explain. I do explain it because it's the by far easiest thing people can test on their own system.
When looking at this legendary example picture:
http://img234.imageshack.us/my.php?image=screensho t34ji.jpg [imageshack.us]
You see a bunch of GNOME applications showing different types of Toolbars. I don't want to speak about the images inside the Toolbars but rather how they look. They all look differently, behave differently, react differently, some toolbars are higher than others (a few pixel) others have a drag handle, others show icons only, then others again show text below icons. There is no common approach of doing this correctly. Sure some people say these things are not important. But from a developers point of view - they are. It only shows in what bad shape GNOME really is even today with latest CVS you see the same issues still present. It should give the beginner and advanced users an impression what's wrong. A Desktop Environment should provide a consistent API and framework to do these things correctly. Please load up GNUMERIC, Abiword, Evolution, Evince and a few others and go through your "Menus & Toolbars" capplet (control center) and change around the values and you see that the majority of applications bundled in the corporate GNOME desktop do not react on these changes. Personally I consider these things to be a bug. I already reported many of these issues and recently my toolbar bugreport to gnumeric got closed as NOT A BUG with some random intransparent excuses why the HIG cant be applied to gnumeric. This is quite frustrating since the applications look bad that way (only the aesthetic view that GNOME always wanted to lay big values on). There are so many other areas like button padding, button padding between other buttons and and and.
It's a never ending story. Also I ask myself why tools like Evince or Epiphany (both part of the GNOME desktop) come with an own Toolbar editor while other applications don't support that. From a developers point of view this should be part of the GTK+ Toolkit and made available default to all apps or everything that uses the Toolbar.
Thats the big disadvantage of writing apps in C without proper object orientation (yes I know GNOME has some sort of object orientation). If we look over to KDE for example then we see that every application that uses a Toolbar (not all a
Apple is missing a huge opportunity here since they have done all the legwork to run OS X on X86 but refuse to licence it for non-Apple hardware. Windows is crap; period; always has been. No propraganda from Microsoft changes that. I am a HUGE Linux fan and use Linux for everything I can. However, I realize that for the average user, OSX would be a logical choice. However, I doubt many organizations will run out and replace all the PC's with Apple hardware. Leave it to Steve Jobs to always choose the most bone-headed choice.
I guess they shouldn't have been illegally using software they didn't own then. Out of compliance means you have more installations than you have licenses for. I find it interesting that criminals are touted as a Linux success story.
But Window IS broke. (Writing this on a XP system while my Linux laptop sits next to me acting as a jukebox...)
The Kai's Semi-Updated Website Thingy
That is the most retarded sig ever. What if a court case comes before an election?