Month: January 2011

Trying to Focus on Specimen Databasing

This post will explore some fairly specific topics, but I hope the thought process will be instructional (or inspiring) to others. Additionally I think it’s worthwhile to talk about the concepts of specimen/biological collection database management with reference to funding, not schemas and platforms.

At the UND Department of Geology and Geological Engineering, a small number of us have been pursuing an overall upgrade of the paleontological specimen and lithotype collection consisting of improved facilities (compactor cabinets) and a comprehensive online database. We’ve applied for funding from NSF and been denied twice, and the project would be dead in the water except for the quarter-time assistantship I’m receiving from the Dean’s office at the School of Engineering and Mines. Development has been slow, mostly due to the conversion between the existing databases (stored as flat text files) and the online system (I will not mention the name of the new system because events today have made me question (again) the cost/benefit ratio of utilizing it), and I’ve been importing locality data so we can use the new system to analyze locality distribution, among other things.

The question today is how to proceed. As useful as locality data are to paleontological and geological researchers, locality information is, at its core, supplementary to the specimens themselves. (I’ll avoid an argument right here: I believe that locality data are essential to proper context, and I’m not advocating the dissociation of these data from specimens.) Specimens are the core of the paleontological sciences, and it is from specimens and their assigned taxonomic identities that researchers work toward understanding past life. Rather than browsing locality lists and then looking at specimens, given a database most researchers will search by taxon or in special cases by specimen number, and then they will look at the associated locality data. In my opinion, we’ve been doing it wrong.

The above point regards usability, and I promised to talk about funding issues, so here we go: in order for such an online database (and more importantly, the effort to digitize specimen data and provide specimen imagery) to keep getting funding, it needs to be usable so it will be used! That’s the whole point. If the Dean (or any other UND administrator) wants to put us on the map for having a world-class collection, we need to get the data out there that people want, we need to tell them about it, and we need to encourage them to use it. From the administration’s perspective, numbers are going to determine how successful we are: number of unique visitors the online database gets every year, number of publications that reference specimens held in our collections, and number of researchers who visit or request material loans.

What can I do today that will improve our chances? In my opinion, we need to improve usability by others before we can improve usability by ourselves. This means a focus on specimen-data entry, the postponement of certain analytical capabilities we (as UND researchers) would like, and beginning with those specimens referenced in peer-reviewed articles, dissertations, and theses. These specimens have already gotten the most attention and they are likely to get more attention in the future because of their “published” status. The associated material can come next, and then we can start adding data systematically. At this point, to show that this is possible and that it shows our research collections in a good light, we need to get the bare bones online first and follow with everything else later.

That’s what I think, and what I will discuss with others here later today. Has anyone else come across such a crux of funding issues? How about with specimen collections that are even less sexy than ours (which are primarily freshwater mollusks, and are pretty darn sexy in my opinion)? Am I on the right track, or should we back this train up again?

Things It Would Be Great To See Finished (part 1) – Bicycle Routing Map for GFK

I’m always trying to organize myself a little bit more. Unfortunately, this means not being able to take the time to do (all by myself) everything I think would be totally awesome.

Excuse me while I take my pizza out of the oven.

The first of the many things I would like to highlight is my idea for an online bicycle routing map for the cities of Grand Forks, ND and East Grand Forks, MN. Obviously, these are small towns and don’t warrant the attention of people who like to develop maps on a large scale, but since I’m here and want to encourage people to ride more, I’d like to see it happen. This idea stems from my Advanced GIS class at the University of North Dakota, during which I built an (offline) routing map that integrated the streets, paths, and bike lanes to try to study specific routing problems in the city.

I discovered (last spring semester it was) that the mapping service Cloudmade was utilizing Open Street Map to do similar things for projects such as Ride the City. I began adding things via GPS tracks towards the end of the summer (I spent most of the spring and summer in a wheelchair/on crutches), but have not had time recently to add any more or work on the specific details. It would be great if anyone else in the area was interested in collaborating (competing?) on making OSM as reflective of the real-world as possible.

Not only this, but I’ve constructed an example of what the interface could do, with some tweaking. See the example here, [seems not to work in Opera 11, hmm] and remember that I’m not responsible if you get lost. Specifically, I’m not sure how the routing algorithm works, how it weights streets, and how it can be customized. It would be great if someone could help me with this or explain it to me. The link above has a list of project goals.

The point? For anyone who is not a bicycle commuter, even in a city as small as Grand Forks, the prospect of getting between points A and B is a daunting one. You need to deal with traffic, you need to look at intersections a certain way, and you may need to bend the law (or think you need to) in order to get where you are going (to follow a sidewalk under an underpass, for example). A routing map specifically aimed at cyclists and took into account some of the oddities of this particular city (“mistakes in planning”, let’s say) would make this step of commuting a little bit easier and maybe help people make it farther than down the block.

I’m open to questions, comments, and offers of help (or desires to take over the project completely). I don’t know of a similar project in the works for this area.