Carl Grant posted this week on his blog about a kerfuffle with Serials Solutions, over who owns a library’s data, and makes some solid points about what libraries can and should do to help prevent this kind of problem. I agree enthusiastically with everything he says in there, surely: establish policies, get it in writing, hold vendors to their commitments…all good admonishments, to be sure.

Serials Solutions has apparently told libraries that their data presents “competitive advantage,” which is vendor-speak for “vendor lock-in.” It’s easier for them to sell you renewals on their services, if it’s hard to get your data out, you see?

One thing Carl doesn’t say here, though, is that one useful preventative measure for this sort of problem is to use systems that are open-source, and based on open-source technologies. This not only prevents vendor lock-in, as there can be other vendors that support the same software, but it also makes movement of the data for purposes other than vendor change much easier.

Take Koha, for instance. (You knew that was coming, didn’t you?) Koha’s data storage mechanism is in MySQL, itself an open-source database engine. There is development toward support of other open engines as well, including PostgreSQL. Since Koha is open source, you have a choice of vendors to support your ILS, or go it alone, and a change in vendors should not result in a lot of needless pain and suffering–or expense–of a migration.

Getting a copy of your data is super-easy with Koha, too. Since just about everything is stored in the database, a dump of the database taken with mysqldump is all you need. ByWater partners already have this happening every night, automatically, with backups being copied to multiple places to eliminate single points of failure. If a partner wants a copy of their data for testing on another server, it’s easy to give them. We also use these “last-night” backups occasionally to spin up on a test server on our end to help debug problems you may be having.

I had an experience a couple of years ago where a library’s server crashed from a dead hard drive. They had their MySQL backups on a USB thumb drive plugged into the server, so after they got a new hard drive plugged in, I was able to walk them through reinstalling the operating system enough for me to get to the server over the network, reinstall Koha, reload their data, reindex, and declare them back up and running. Start to finish–about four hours, including the drive to Best Buy for the library director to get the new hard drive.

Naturally, we all hope that vendor switches are infrequent, and crashes are never, but the use of open-source systems and open-source underlying technology can make either of these situations much less painful for you, your library staff, and your patrons.

[Originally posted by D. Ruth Bavousett]