Recession economics and Linux

The biggest obstacle to FOSS (free and open source software) is the cost of retraining and support not to mention rewriting any specific applications used extensively in the company, ie - an Access database.

I agree with parados. It's quite possible that developers of cheap, new devices like the netbook will dump Vista or Windows CE for Linux. But the recession isn't changing the costs of switching over the benefits of switching, so I expect no dramatic change.

My old hometown of Munich, Germany has started changing its IT infastructure from Windows to Linux in 2000. They're happy with it, but it's a slow, gradual, and still ongoing process. We may well see more examples like that, but the time this takes is way longer than the expected duration of the recession.

The only things I can see taking any time or presenting any unusual problems switching to LINUX would be drivers for new hardware items or needs for unusual software. Should take about four or five weeks, and not five years.

You don't deal much with computers, do you gunga?

What about all their documents saved with proprietary markup? They'd need to start converting everything and in some cases writing improvements to the open source software to do some of the things they need it to do.

In addition to this there is a lot of software that just doesn't have viable options on linux, things like accounting software (some of the creators of Linux still keep a Windows machine around to do their finances), graphics and video editing, some publishing software....

Linux rocks, but for a large scale company or a government to completely switch to open source platforms is not as easy as just finding drivers for their hardware.

In the future, as more and more FOSS becomes legitimate competitors to commercial software it will be easier, but there are still some departments of some companies that can't get by on Linux right now.

You didn't even mention having to change over servers providing intranet and data to client computers at a company, Robert.

What gunga doesn't get is the biggest cost of computers is not the hardware or even the software itself. It is the hours of training and OS specific tweaks that have been done over the years. A single admin that has spent 5 years at a company will have spent many hours creating templates that utilize the existing system and help his/her productivity. Taking those templates away by changing the OS adds costs that have to be considered.

Something as simple as a mail sort may have taken 10 hours to set up at one time. Redoing it while learning a new OS will take at least that long again.

Yeah but that's not what has caused the delay for them. They had legal issues with patents and put the project on hold and they won't ever fully get off Microsoft, as they said:



One example I know they had running and can't find any alternative for was AutoCAD.



I think that depends on the software environment. There are certainly some places where the software and hardware costs completely eclipse the training costs.

In this particular migration I believe about a third of the budget is dedicated to training. But yes, it's one of the things taking time. They moved most employees to Open Office, Firefox and Thunderbird on Windows to get them used to the new applications they'd be using on Linux.

In this particular case, I think it's mainly sheer volume and interdependencies that cause the bulk of the transition time. With that many machines any kind of transition is hard even if you discount the human factor.