Farewell to Munich's Linux project

End of Munich's Linux project should not be end of government IT experimentation

Tags: GermanyMicrosoft CorporationOpen source
  • E-Mail
Farewell to Munich's Linux project The City of Munich's project to move its computing platforms to Linux was a bold plan that should be an example to other government projects. (ITP Images)
By  Mark Sutton Published  November 16, 2017

The news that Munich City Council is to ditch Linux in favour of Windows marks a sad end to an experiment in open source software in the public sector. Munich was a long time champion of open source, opting to switch all of its systems from Windows to Linux and LibreOffice back in 2003, with the aim of controlling costs and reducing reliance on Microsoft and other commercial software vendors.

By 2013, the city had successfully migrated around 15,000 desktops to ‘LiMux', but now, with changing leadership and changing technology, the council has decided to ditch Linux and replace it with Windows 10 by 2020. The committee in charge cited costs, mainly from the complexity of making applications run on Linux, user resistance and communications problems between LiMux users and customers and partners on Windows, although not all stakeholders agree with the decision or the reason.

The project had already been slowly ramping down, with many users already switched to Windows systems, but now it seems the initiative is all but done - a sorry end to a bold plan to ensure that city government got the best value for money and had more control over its technology platform.

I'm not anti-Microsoft, but it seems to me that it is good to have a workable choice when it comes to technology suppliers. When a company has a de facto monopoly of so many of the applications we use in our work environment, it's good to have some other options, and projects like Munich can serve as a proof-point that there are alternatives. LiMux was an attempt to be more accountable for public spending, improve cybersecurity and enable the city to decide on things like upgrade cycles and support for itself.

That said, in the nearly 15 years since Munich opted for the open source route, a lot has changed. Microsoft has changed its approach (and improved security) and many of its applications are available at much better pricing for the public sector than they were 15 years ago. Large scale licensing agreements give a better return with add-ons like training and joint initiatives. With much more complexity in IT, having a standardised OS base for platforms and development is an important step to get the whole ecosystem working properly.

One of the biggest problems for Munich was getting public sector applications written for Windows to run on LiMux, and it could be argued that a city council is not a software developer, and budgets spent on porting applications could have been better spent on other things. Some might even say that a public sector body has no business conducting what was effectively an IT experiment.

In the UAE however, government organisations are most definitely leading experimentation with IT. Indeed, programs like Dubai 10X almost demand that government starts using technology that has not had extensive testing. This will require government departments to retain the ability to experiment, including staff skills and budget that will enable them to do so without disrupting normal service. In the 15 years since Munich announced its Linux plan, the IT sphere has also moved on in ways that make experimentation more feasible.

Cloud computing is now common place. In the past, government would never have considered using a web-based service for email, but now plenty of organisations are happy to use web mail and many applications in the cloud, although maybe not for sensitive communications or mission critical applications. Open source has also become much more common in many areas, and is now often the main source of tools in some of the most advanced technology deployments in fields such as analytics, machine learning and blockchain. Public sector organisations that want to experiment in any of these fields will likely use open source whether they want to or not.

Ultimately, Munich's decision to drop LiMux was driven in part by political allegiances, and rather than be seen as a failure, notwithstanding the cost of two migrations across operating systems, it should serve as an example that the public sector can, and should, be willing to look for new ways of doing things that challenge the status quo.

Add a Comment

Your display name This field is mandatory

Your e-mail address This field is mandatory (Your e-mail address won't be published)

Security code