Calling all OS OpenData users!
Have you struggled to get to grips with GML? Have Shapefiles left you feeling positively un-shapely? Do you dream of a Dump file? Maybe you get giddy over a GeoPackage? Then you may want to take part in our OS OpenData trial that’s being announced at FOSS4G in London today…
Since the launch of OS OpenData in 2010, we have supplied our data in a GML and/or SHP format. Any open data we publish must include open source formats to meet open standards, and GML fits that bill. But over the years, the open data community have told us that they sometimes struggle to manipulate the data in GML and SHP. With such a varied user base of our data and a nation full of scenarios to tackle, we want to find a format which can work wonders for the masses.
We’ve been investigating exporting our data in a range of different formats. For this trial we’ve chosen nine OS OpenData products and, while we have tested internally, we would really like to have your feedback. Your feedback will help us assess the trial and decide whether to take the data formats any further.
The products you can trial with different formats are OS Open Names, OS Terrain 50, OS Open Roads, OS Open Rivers, OS Open Map-Local, OS Open Greenspace, Code-Point Open, OS VectorMap District and Boundary-Line.
Regular OS OpenData users will notice that some of the formats are different. Some will enable you to access the whole of Great Britain in one file rather than in the smaller tiles you have used previously. We’d appreciate feedback on this too.
Data format trial
Access the trial here by the 18 May 2018 deadline: http://data-format-trial-osonline.opendata.arcgis.com/
You will then be able to download any of the nine products listed above, which are available in GeoPackage, PostGIS Dump, Zipped CSV, GeoTIFF or ASCII Grid, depending on the product.
After you explore the product(s), we would greatly appreciate you completing a quick online survey for us. We want to know your thoughts on these trial formats, especially if it speeds up your processes, is more user friendly and/or would make you more likely to use OS data more regularly.
Thank you in advance for helping us improve the products you use.