Data, data, every where, nor any point to map
Having spent more time over the last 2-3 years than I care to quantify trawling the internet for public domain datasets to use on various projects, I’ve reached some disappointing conclusions: I will rarely find exactly what I’m after, working with data portals can be very frustrating, adding and maintaining metadata is a frequently neglected chore, and all too often data are published but not adequately promoted.
Co-author Joseph Kerski recently described a particularly frustrating day spent trying to find some geology data for a presentation he was working on. Having failed to locate a data source on-line, he ended up contacting a colleague at a state geological survey office who was able to provide access to some data. Joseph was also given permission to publish the data on ArcGIS Online for others to use, but he was left wondering why the organisation hadn’t chosen to publish and promote the data source itself.
In this era of open access, when organisations are rushing to liberate their data stores, many of these new on-line repositories often lack an effective interface to the data. Increasingly sophisticated search engines quickly return a list of candidates for each inquiry, but the success of a query can often rely on the syntax of the search instruction. Depending on how each query is worded, the whole process may be very hit or miss. From my own experiences, it would be good to see a more task-based approach to searching for, and providing access to, data. When people go looking for data, they generally have specific requirements for a given task and they’re looking for specific data, in a specific location, over a specific time period, with a view to producing a specific information product as the output. Few people have the time or inclination to trawl through everything every to do with ‘housing’ or ‘planning’; they also need the tools to allow them to extract the information that will serve their purpose.
Frank Biasi (National Geographic Maps) once commented on the demise of a conservation geoportal noting, amongst other things, that “.. the concept of sharing data is much more advanced than the practice“. Seems like we still need a lot more practise.