The Woes of Munich's Linux Migration 314
mikrorechner writes "The H Online has a writeup of the problems encountered by LiMux (Wikipedia entry), one of the most prominent Linux migration projects in the world, trying to introduce free software into the highly heterogenous IT infrastructure of the City of Munich. Quoting: 'Florian Schiessl, deputy head of Munich's LiMux project for migrating the city's public administration to Linux, has, for the first time, explained why migrating the city's computing landscape to open source software has taken longer than originally planned.'" Here is Shiessl's blog, in which he details some of the transition problems.
Re:I find this interesting (Score:5, Informative)
It's not ambiguous in the spec, it's undefined in the spec. But one thing is defined in the spec: a way to do application-specific spreadsheet formulas without breaking the standard and without conflicting with a standardized way of expressing formulas when it's finally standardized. The expectation is that applications will do formulas their own way, possibly recognizing other application-specific formulas (there actually aren't that many different formats). When formulas are finally standardized applications will begin using the standard and will convert any non-standard formulas they recognize into the standard form when the spreadsheet's read in, resulting in a quiet upgrade to the standard form.
And in the meantime, ODF can be used for things like word-processing documents that don't require formulas without having to wait for one spreadsheet-specific feature to be completed.
Similar stories (Score:5, Informative)
Regional government of the autonomous community of Valencia (Spain) also switched [lwn.net] to free software, last year they released a detailed report [gvpontis.gva.es] (english) of the problems they found and how they fixed it. It took a lot of time to complete it (4 years) and they still depend on propietary software for some systems. These migrations need a lot of work...
Re:how much did this all cost? (Score:5, Informative)
They aren't trying to make "everything work like it did before with the same functionality". They could have
We could have switched to linux clients in just a few months, giving the order to all 21 IT units to set up a linux client until end of 2008. No further specifications, no standardization and no consolidation. I’m pretty sure they would have done this excellent and then I would have published great news in 2007 or 2008 “LiMux done, Munich completely on free software”.
but the aim is/was to move from a very heterogeneous network (in terms of used OS and software solutions) to some overall standard, which is why it takes so long.
Can I still keep my geek card if I actually read TFA?
Re:Bad title is bad. (Score:5, Informative)
Myself, I'm certainly no Microsoft shill - I'm a Linux proponent, and interested in the LiMux project because I live in Munich.
If the title seems overly negative, I apologize - I'm no native speaker and might have chosen the wrong words.
Re:The real reason its late (Score:2, Informative)
I thought that was Gentoo?
Re:There is no free lunch (Score:3, Informative)
>>>Also, floppies?!? Really?
If you know how else to install Damn Small Linux or Kolibros onto a 386 machine, which only has a floppy for external input, please share.
Re:As usual (Score:3, Informative)
The real problem then was that they didn't made an in-depth analysis of what they were using originally. It's always the same.
That is not how I understand the blog. They started the transition, and realised that yes, they could do a transition in the allocated time frame, but they wouldn't get the maximum benefit that way. So the plan changed. Instead of saying "we planned to do it in X months, so we do it in X months", they said "we could do it in X months, but we could get much better long term results if we do a better job that takes 2X months".
Re:No free lunch, but a range of benefits. (Score:3, Informative)
The development costs will be a one off...
Having maintained windows, linux, solaris and novell based networks my experience is basically...
You require competent staff to manage any system properly, microsoft marketing says otherwise so windows networks often end up being operated very badly by incompetent staff (and have major security and stability problems as a result)... Generally only more competent people even know linux exists, so the cheaper less competent staff will never even think to try linux - if they did the results would still be bad but probably not as bad as a poorly deployed windows setup.
If you don't mind a poor setup, windows will cost more than linux but you will probably not be able to find as many extremely cheap low skilled staff pretending to have linux skills as windows...
If you want a good secure linux setup you need decent staff...
If you want a good secure windows setup you not only need decent staff but also a lot of third party software...
It's also my experience that you need more staff to maintain a windows setup unless you cut corners...
The problem is corner cutting, people think they can cut corners with windows but the end result is a huge insecure mess.. 99% of the companies i've ever been to simply don't have the budget to maintain a windows network properly..