(when I saw the title “How Is a Scarf Like a Dataset?” I was really hoping it would connect the process of knitting to this topic, which is an accessible analogy for me. I wasn’t disappointed!) In any case, while this article was wholly entertaining, I do not feel that I learned a lot about the creation of datasets, and am now really curious about what the “bind off” of datasets is. I did find it to be generally applicable to how I USED datasets in other classes. It was a fun article, though.
I also found the Heidorn article very interesting. As someone who works with government documents, and documents produced by government entitities (which do not always seem to be the same thing – the judicial branch never seems to be categorized with govdocs), I am so interested in how proprietary databases and products making these documents and data accessible work.
“Many scholars are unaware of the coming changes in the sociology of science and do not have the required skill sets to address the requirements in their new proposals (Cragin, Palmer, Carlson, & Witt, 2010). Worse, librarians know relatively little about current data management practices of scholars. Institutions have not yet established who will conduct data curation work.”
This is precisely why this kind of data ends up only functionally accessible by something like ProQuest. Equating scholars to government agencies and entitites is perhaps ignoring nuance, particularly that government data is publicly funded and that which is not classified should be free and easily accessable, but training librarians in specialized data management practices can only make information more accessible to the public.
In direct contrast to my last comment, I think it is so interesting that data collection (as in, making data part of the collection) by academic libraries is even happening. It is something that had never occurred to me. The idea that libraries should be involved from the start of data collection is so intriguing, and I see where the idea is coming from, but is that standard feasible for scientists who are not working under an academic umbrella? I am sure that many scientific organizations do have librarians, but, my point is, could a rise in librarian-aided research lead to preference for that data, and therefore research from smaller organizations may become even less represented? Is data collection with the help of a librarian better data, or just easier to integrate into a library?