Research Data Alliance: Of data, water and lifeboats
Do this for scholars, not for you: First she stressed that we should approach data infrastructures, not with how we want them to work, but the way that scholars work! Her observation is that most fields won’t share their data so readily. Yes, we have some fields that do, e.g. genomics field. However as is the case with physics, which didn’t roll out as a model for literature repositories, it doesn’t follow suit that other fields will be so keen to embrace sharing data.
So, take the value proposition from position of individual researcher. She took a look at the functions of scholarly publications as follows: Legitimization, dissemination, access. There is an air of responsibility, publications are embodiments of intellectual arguments. But for data, it doesn’t stand alone, you need software for understanding the actual data and making it legitimate.
Data is like a dowry! Remember for a researcher that data is something they hold dear to their hearts. In terms of incentives, you bring it with you to your next project and promotion. Therefore researchers are aware that this needs to be governed as a resource. This element of harnessing it and backing it up, for themselves, or even for future generations and purposes will ring true for researchers.
So, what we can do for researchers?
Give them fresh water: Tools, services, skills, resources, incentives. Here, as we already know, libraries need to get involved as early in the process as possible, with a clear idea of what we want to give researchers.
Give them life boats: Repositories, governance, provenance, and how migrate their data know-how.
Carrots or sticks? A question from the audience highlighted that in some cases sticks do work, e.g with mandates such as Pubmedcentral deposit. Borgman pointed out however that while mandates will meet the letter of the law they won’t necessarily meet the spirit of the law in terms of researchers' attitudes. With data if the incentive is only driven by a mandate, you will end up with researchers depositing unlabeled columns, no code or lab books.
I've got a DOI. Good, I’m done. Another audience question: what is the role of ‘informal’ scholarship: by this they meant Figshare and generating DOIs without necessary any curation tasks involved. So, by uploading, the author is happy and thinks they have met funder requirements. Borgman’s argument is that while this fills a short term role the data hasn’t been made sustainable. Remember, there is no long term commitment from commercial systems, and revenue drives it.
And yes, infrastructure is messy: But it is a good necessity! A bit of the old, bit of the bottom up, some fancy stuff, knitted together, policy, free riders. But this cooperation is essential to serve researchers' needs.
Christine's slides can be found at her website: http://christineborgman.info/
Amsterdam, 24 September, 2014