For some time now, rumors have been circulating about a possible NSF-supported initiative to increase on-line access to biological collections. Now it seems like this might actually be a possibility; you can read more by following this link. This blog constitutes a community engagement process, so if you actually care about this issue (which I'm sure some of you do) I suggest scrolling down to the bottom of the blog and leaving a comment. Being an opinionated sort of chap, I already did. Suffice to say that while I think this is an excellent idea and while there are some good ideas circulating, I did feel obliged to strike a cautionary note.
First, you'll notice that a lot of the supportive comments come from botanists. This isn't particularly surprising because the botanical community has always been at the forefront of thinking about digitization. But there's also a good reason for this. Botany collections have one huge advantage when it comes to specimen imaging - the specimens are two dimensional. There is a world of difference between photographing a herbarium sheet and shooting the many irregularly-shaped lumps that make up the average vertebrate fossil, and this is reflected in the time and cost of doing so.
Second, as I've mentioned in other posts, cataloging (and yes, believe it or not there are significant swaths of our biological collections that have never been cataloged) is laborious, time-consuming, and unglamorous. It requires people, working systematically for many years, and there are various key aspects that are resistant to automation. Take, for example, labelling specimens. Yes, you can print off a label directly from your database, but you have to find a way of attaching the label to the specimen, or putting a catalog number on the specimen that links it to the label and to the original data. If you don't do this, and specimen and label part company, then the specimen becomes all but useless and all the effort that you have devoted to digitizing it - not to mention collecting, preparing, studying, and housing it - has been wasted.
This is not to say that NSF shouldn't be developing new technologies as part of this initiative - far from it. We badly need on-line tools to help people access, use, and enhance specimen data. But (to return to one of my enduring hobbyhorses) museums also need to commit their own resources to supporting the core activities of collection management, and that includes cataloging.