I first started out with Linux in the days of Windows 3.1 or maybe it was 3.11. As you can see it’s user interface were limited. It had nothing but a proprietry network stack. In general it was just not really advanced, it was the OS of the last decade waiting for the upgrade to take it into the next.
Enter Linux. I immediatly recognised a good thing when I saw it and started using fvwm on Linux as my main desktop system. Now lets be honest here, the competition was pretty much absent so it was not a big task to beat it. This worked brilliantly for me, it did everything I wanted to do. I could customise it to death, assign shortcuts and so forth, just what I wanted at the time.
There were many reasons why this worked for me.
- I had lots of time on my hands, so manually editing config files were no problem.
- I was eager to learn new things and so found it all to be a great adventure and learning experience. If anything it showed me the value of reading documentation.
- I was a deep techy, I never had to open a MS Word or Excel document and it was the days before HTML email, proprietry rich text formats and everything else bad about the net today.
So to put it in one line: Linux Desktops were radically different from Windows, it was an obvious improvement and it was innovative.
It is now 7 to 9 years later. The opensource movement has taken off in a big way, 100’s if not 1000’s of developers have been slaving away at the code. Things have been evolving with great leaps and we now have the most advanced, innovative desktop system ever developed and all this for free. Ok, maybe not.
What we have instead are window managers available in 3 flavours.
- Blatant Windows Rip-Offs, think KDE, Gnome, Bluecurve.
- Rather simple window managers thats themeable to death, think Fluxbox, Windowmaker etc.
- Radically different window managers such as 3dwm
The defining design principal today seems to be: Make the transition from Windows to Linux as simple as possible by simply makingeverything behave like Windows. It is true that these window managers all allow for themes to be applied extensivly. This allow you to make it look at least like it does not resemble a Windows Rip-Off, but other than that there are little innovation that sets them aside. They all try really hard to be as close to windows as possible, without ever getting quite there.
The few that does not pretend to mimic Windows as close as possible have a small amount of innovative features.
- Tabbed windows like those found in Fluxbox.
- Docks for various small applications usually called dockapps to run in, mp3 players, network monitors and the like.
- Extensive abilities to change the appearance of the window manager using themes.
The only feature above that actually sets these window managers apart from fvwm that I used in 1994 is really the tabs and code maturity. This group of window managers has not done much for innovation.
Applications come in 2 flavour.
- Blatant Microsoft Rip-Offs
- Industry specific systems that fill a niche market, such as Blender.
Name 1 major desktop application that you use that isn’t a windows rip-off. Some people who I asked this question answered by getting upset, refusing to partake in conversations or mention things like terminals. Others immediatly said Windows just stole from Xerox or Apple. This demonstrates tome a major problem with the OpenSource world, they have zero ability to look at themselves in a ojective manner – Microsoft stole everything they have, they are the enemy, we are right, just dont ask us to substantiate it with arguments. The only person who answered the question properly mentioned LyX. This is a good example of a non-ripoff application that is being used in certain communities. But for the biggest part everyone juse use applications that resembles their Microsoft counterparts.
It is true that for a desktop to be usefull today – especially in the Enterprise – it needs to “speak Microsoft”. It must understand MS Office documents, it must understand MS Enhancements to protocols. It must talk to MS Exchange servers and so forth, you get the idea. So I guess you cannot blame developers for ripping-off the Microsoft products since it seems those are the only ones that will get used. A good example would be KWord, it went the route of being “FrameMaker-like” and I could not find anyone who is using it. Since being different from MS Word means it does not render MS Word documents right, just like the real FrameMaker.
Now this is not a good thing, I know this, all about monopolies and illegal tie-ins. Bottomline is though you cannot partake in the enterprise world without being MS Compatible. There are a number of options here, Evolution at $70 per user Exchange Connector, Crossover Office at $55 but you still need a Office licence! and so forth. The opensource world is hitching its bets ofcource on OOo – as it stands now, this is the quickest way to corporate embarresment when you send out a document that looks like arse becuase your co-workers or clients with Office XP will be in the position you were, the “Word” documents you send them will look bad on their MS Words since you are not quite compatible with them.
The other option ofcource is VMWare. I dont see how this saves time, money or anything. It requires a Linux Desktop (a big one – read expensive – since VMWare is really bad on resources). It requires a Windows licence and MS Office licences. I dont know how this fixes any problem, all it gives the enterprise is many more complaints about things being slow and complicated.
Your usual Linux geek will tell you how Linux is easier on the desktop to maintain etc, but how many of them have experience making Office under crossover talk to your Windows PDC etc? One Unix geek may be able to get his desktop system “just right” by spending many nights at home tweaking themes, or many work hours doing the same, I dont want to imagine how much work it will take to get 100 desktops “just right”, and I don’t think you want to imagine the amount of waste man-hours due to employees tweeking their Linux desktops endlessly.
For now the best options in my mind are:
- A dual boot system using either a dedicated Linux install for your power users or a Knoppix CD with some dedicated drive space for user files – you could simple distribute new CDs when a upgrade comes out. This is suitable in a very specific environment where some Unix tools are needed for some tasks only. You may have a specific rendering tool or something.
- A dedicated unix “shell” server maintained by one, or a group of people. This big server will have all the typical unix shell tools on it, but could also have some X application on that can be made accessible through one of the many Windows based X servers. This gives the user a dedicated Enterprise Compatible Desktop while maintaining the power user utilities of Unix available to all.
- If you are lucky enough to have 2 machines, one with Windows one with Linux or FreeBSD. Share a keyboard and mouse between the two using synergy and create a shared document drive using Samba. In this case you would also possibly want to look at one of the non-windows type Window Managers so you can actually take advantage of the little bit of innovation that is on-going.
Personally I have a combination of the 2nd and 3rd options above for my work environment.
