Most open source advocates like to believe that the migration from proprietary software products over to open source alternatives is fairly straight forward. Unfortunately, it's never that simple.
The problem is that proprietary legacy software is a lot like a barbed fishing hook. Removing it to make things better tends to hurt more than most people are comfortable with.
Microsoft Office Products
Considered to be the bane of any successful open source migration, Microsoft Office products make transitions a real nightmare.
In a perfect world, employees would simply make the switch over to the Open Office suite. Bundled with most Linux distros, Open Office has a clone application for nearly every Microsoft Office application available, with the exception of Outlook.
Despite these similarities, Microsoft introduces two distinctive barriers to migration that are difficult to overcome. First is the ribbon user interface found in modern versions of Microsoft Office. Second and the biggest of the two: file format lock-in with docx.
Open Office currently does support reading and writing to docx file formats. But this non-Microsoft office suite does not maintain formatting of most original docx documents that were created with Microsoft Word.
As you might imagine, this renders Open Office's support for docx nearly useless in most enterprise situations. So while it's doable to get users to "let go" of the ribbon interface of Microsoft Office, it's impractical to re-work all the existing docx documents using specific formatting that is not likely to transition well to Open Office.
Then there is the Excel conundrum. Open Office provides a reasonably useful spreadsheet program right out of the box. Yet despite its inclusion, I have met very few Excel users who have tried using the Open Office alternative who were fully happy. Most immediately found themselves missing a specific function or path to a function. This is a case open to individual interpretation to be sure, but nonetheless it has many new Open Office users feeling flustered.
Long story short: Open Office doesn't do well in the enterprise environment if newer versions of Office have already been introduced. Simple as that.
Before diving into this area, let me acknowledge one thing. Locally installed software is slowly becoming less of an issue than it once was. In many cases, I have seen firsthand evidence of network/Web based solutions that businesses are using as it makes IT management easier.
The downside, however, is that most of it remains proprietary in nature. Even worse, some of it requires the Windows OS due to its ActiveX needs or simply looks to work with Internet Explorer browsers only. Obviously this is not the case in every instance, but it's common enough to prove a problem more times than I can count.
So now we have a question that is begging to be asked: What if a company wished to migrate to a newly discovered open source solution? Sounds like a fantastic plan until the harsh reality of vendor lock-in rears its ugly head.
I know of a Veterinary office that have ran into this recently. They wanted to explore an open source alternative that was available free of charge. Unfortunately the existing software records could not be exported except into the newer version of the same proprietary application they were currently using. Needless to say, they are still not able to make the switch even to this day due to the lock-in problem.
The take-away from this is that vendor lock-in alone is reason enough to avoid proprietary software. No one likes being locked into something without the option of changing later on.