Christina K. Pikas is a science and engineering librarian in a special library as well as a doctoral student in information studies.
There's a nice rebuttal of the Sirsi Dynix anti-open source white paper done by Mark Leggott that just came out (I found it via Jason Griffey). More thoughtful than some.
There are so many misconceptions on both sides of this. First, open source is not free. You do need people to install and maintain it and maybe customize it. Some open source projects have less documentation than others.
On the other hand, what's worse is when you pay hundreds of thousands of dollars for a large software product only to then have to pay more and more and more and more to buy additional modules, customizations, training, support, upgrades....
At least if you have some good programmers on staff you can fix the open source yourself.
Migrating a large database or content management system to another is always a hassle. That has nothing to do with whether its open source or not.
Technology
Information Science
Comments
Most of your post is spot-on, but I disagree with the last paragraph. Very often, proprietary systems will do what they can to lock data in. Open source systems tend to focus on having a clean infrastructure more than features; proprietary apps prefer features (which drive sales) over infrastructure (which can be compensated for by having full-time experts develop the system).
If you have two systems that use a SQL database to store application data (such as the web site content for a CMS), it's fairly likely that the open source system will have a cleaner schema, and more likely that the proprietary system vendor will not want to provide documentation for the schema. Both of these things make it easier to get data out of -- or in to -- an open source system.
Posted by: Michael | November 8, 2009 9:04 AM