IL2009: Cloud Computing in Practice: Creating Digital Services & Collections

Speakers: Amy Buckland, Kendra K. Levine, & Laura Harris (icanhaz.com/cloudylibs)

Cloud computing is a slightly complicated concept. Everyone approaches defining it from different perspectives. It’s about data and storage. For the purposes of this session, they mean any service that is on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service.

Cloud computing frees people to collaborate in many ways. Infrastructure is messy, so let someone else take care of that so you can focus on what you really need to do. USB sticks can do a lot of that, but they’re easy to lose, and data in the cloud will hopefully be migrated to new formats.

The downside of cloud computing is that it is so dependent upon constant connection and uptime. If your cloud computing source or network goes down, you’re SOL until it get fixed. Privacy can also be a legitimate concern, and the data could be vulnerable to hacking or leaks. Nothing lasts forever — for example, today, Geocities is closing.

Libraries are already in the cloud. We often store our ILS data, ILL, citation management, resource guides, institutional repositories, and electronic resource management tools on servers and services that do not live in the library. Should we be concerned about our vendors making money from us on a "recurring, perpetual basis" (Cory Doctorow)? Should we be concerned about losing the "face" of the library in all of these cloud services? Should we be concerned about the reliability of the services we are paying for?

Libraries can use the cloud for data storage (i.e. DuraSpace, Dropbox). They could also replace OS services & programs, allowing patron-access computers to b run using cloud applications.

Presentation slides are available at icanhaz.com/cloudylibs.

Speaker: Jason Clark

His library is using four applications to serve video from the library, and one of them is TerraPod, which is for students to create, upload, and distribute videos. They outsourced the player to Blip.tv. This way, they don’t have to encode files or develop a player.

The way you can do mashups of cloud applications and locally developed applications is through the APIs that defines the rules for talking to the remote server. The cloud becomes the infrastructure that enables webscaling of projects. Request the data, receive it in some sort of structured format, and then parse it out into whatever you want to do with it.

Best practices for cloud computing: use the cloud architecture do the heavy lifting (file conversion, storage, distribution, etc.), archive locally if you must, and outsource conversion. Don’t be afraid. This is the future.

Presentation slides will be available later on his website.