Unit 12- Preconfigured VMs

This week's question is an interesting one, as I never considered the possibility of installing a pre-configured VM. My first instinct is that building a VM from scratch is a worthwhile learning opportunity, plus seeing the end result of what you've created is very satisfying. The other side of the argument is that my computer skills are adequate but I fear I lack the ability to build a high quality digital repository. The question put forth is that if I didn't have to spend a good amount of time learning these computer skills, I could focus more on the content of the collection. In my current job, our library uses software that we are unable to edit or add to, to meet our needs. The software was designed very nicely, but we have technical questions we can't receive help because our IT folks are unfamiliar with the system (and the licensing support has expired). I feel that building our own program would allow us to at least have a better working knowledge of how to make adjustments as we see fit. This would definitely benefit the collection. If a pre-configured VM is supported by an open access community, that would be good as well.

Unit 10 - Service Providers

The first service provider I looked at was Citebase Search. This is a self-proclaimed "work in progress" database which, while stating they've harvested metadate from different OAI repositories, they don't specify which ones. For some reason, this is information I like to see, and service providers such as Clio-i provides. I do, however like where Citebase Search is going with their setup. Once completed they'll have information on how to use the database, which organizations are harvesting from them, etc.
The University of Illinois OAI-PMH Data Provider Registry has a very informative consortium of information repositories. As a example, I looked at the repository for Princeton University Library. The service provider not only lists which repositories this one harvested from, but describes them as well (such as the subjects their records are on). The provider also states the OAI base URL which, I have learned, is essential for harvesting the metadata into my own repository. This service provider (as opposed to the "Registered Service Providers" website) outlines much of what I need to know before actually going into the repository and searching for it myself. This is very helpful as there have been some repositories where I could not find the OAI base URL, so I couldn't figure out how to harvest the metadata.
Sites that harvest a large number of metadata could potentially be more helpful; more likely to hold information that a user is seeking. I worry, though, about the technical implications of storing so much information on one site. If the repository somehow exhausts the amount of memory it can hold, the site could potentially lose service- the instability of such a repository might result in users losing faith/turning to other sources.