Nettitietojen mukaan Sveitsin Solothurin kantonissa koetun Debian Linuxin fiaskon takana olisi ollut projektin jo alussa kokemat ongelmat siinä, että data lukittiin Microsoftin tietokantaan.
Tästä tapauksesta on kirjoittanut Robert Pogson:
http://pogson.6k.ca/2010/09/23/a-failed-migration-to-gnulinux/There are right and wrong ways of doing everything and a Swiss canton did several things wrong:
* They let M$ lock-in their data in a database
* They danced to public pressure from rebels
* They did not fire the rebels and hire more flexible people
* They should have been a tad more diploamtic in the introduction
The result has been after several years, a decision to migrate to “7″ on the desktop with no guarantee it will be any less difficult.
I have done a bunch of migrations and they all worked. One does not dictate. That results in push-back. One demonstrates the advantages and reasonable people prefer the change to the status quo. The Swiss decision to migrate to GNU/Linux was made in 2001 when the GNU/Linux desktop was not clearly as capable as it is today and they continued to use M$’s database. It should have been the first thing to migrate. 9 years later, the hole is that much deeper. On top of that a bunch of fhings went wrong like bankruptcies of suppliers. They may well have been better off to have developed the necessary talent in-house. They had the time.
Munich had some of the same problems but they kept working and are arriving late but under-budget. Munich discovered they had more problems than the OS, a fragmented IT system barely able to interoperate with itself. They used their extra time to tune everything up. GNU/Linux is now icing on the cake, a solid cake filled with fuit and nuts, not an “air-cake”. Solothurn, it seems, had similar problems it did not address. The cause of the failure was a failure to plan portable/open-standards-based IT in the first place and not fixing the problems first.
The one thing I have learned about IT is that a migration to GNU/Linux is always possible and worthwhile in the long run. Sometimes it is a long run because there are so many problems but fixing one problem at a time works. You can either fix the whole system in a provably-correct method developed in advance in detail or you can fix one problem at a time as fast as you can do it. If you always opt for open standards, the end-result will be the same but the latter will cost you a lot less time and energy and money. Trying to keep closed systems and lock-in is a recipe for disaster sooner or later.
Tuossa kirjoituksessa mainitaan että myös Saksan kuuluisassa Munchen-projektissa ilmeni ongelmia mutta niitä on ratkaistu ja voitettu. Suljettujen systeemien pitäminen järjestelmän osana on Pogsonin mukaan tekijä, joka enemmin tai myöhemmin aikaansaa katastrofin. Näin kävi Sveitsin Solothurissa, näin on saattunut käydä myös Naantalissa. Sitä en tosin tiedä.