Begin AGU!
Yesterday marked the start of the AGU (American Geophysical Union)’s 2017 fall meeting, which we will be attending all week! We’ve set up a booth in the exhibit hall for anyone interested in talking with us, where we’ll try to get face to face interactions and codesigns with as many people as possible. We’re planning on focusing codesign on 5 things this week: generation of a brand new “dashboard,” creating a resource, filtering and searching for resources, general vocabulary of HydroShare, and group functionality.
The convention center covers almost 11 square blocks and has 3 floors, so we have a lot of ground to cover when getting between posters, talks, and our exhibition hall booth. We did find some great posters and talks about data uploading, sharing, and managing platforms, and we’ll be continuing to follow up with different groups to see if they have found strategies or successes that could help our project. We’re going to keep attending talks and poster sessions, exchanging business cards as necessary.
One of our interesting observations yesterday: There are a huge number of data tools at this conference! Some are super niche to a specific topic, some are more broad, and some almost seemed specific to a university. Here’s a quick list of data-related solutions we found in only the first couple hours at AGU:
- MagIC → geomagnetic data community, data integration, and workspace
- StraboSpot → Geologic data collection tool
- GeoHub → Geologic data storage and sharing tool
- Zenodo → Research sharing + citing
- BCO-DMO → more curated data storage
- IEDA → curated data library, research has to go through submission process
- SubSurface Insights → electrophysical geosciences data
The most exciting part of the day was our 3 codesigns. 1 was with a member of our sponsor team at HydroShare, but the other 2 were active phD students who really would be potential users for HydroShare. We’ve made a system with our codesigns that we hope will optimize however much time our codesign partners can give. If they only have 5 minutes or so, we’ve made a shorter, streamlined session that revolves around word cards. If our partners have 45 minutes to an hour, we have paper prototype mockups of the HydroShare site itself, and how we potentially see it evolving. For the new, envisioned dashboard, this means that we’ve taken a huge conglomeration of ideas from other successful websites and created mini paper prototype index cards of features that can be added to or taken away from the dashboard. We tested this 45 min codesign with two of our partners, and gained a ton of insight into what’s actually useful; it turns out having the “quick actions” of always being able to create a resource or open Jupyter Hub are more sought after than we first realized.
We also used 1 of our sessions to talk about the sort/filter functionality, and determine what categories and attributes actually matter. We used colored index cards to sort by sub category and larger, nested topic and asked our partner to select what criteria mattered more to them. It’s simple, but a powerful exercise that can physically show us what users would potentially care about about how we can slim down our UX to fit their needs. Today, our biggest takeaway was that data is best sorted first by type of data (temperature, pressure, etc.), and then by attribute ( elevation, longitude, temporal resolution, etc.). It’s a more nuanced distinction, but an important one in how users think about organizing their data.
