On 03.09.2014, at 18:26 , Adrian Custer <[email protected]> wrote: > […] which means we must be able to go from the exported database to an > estimate of horizontal accuracy, something my earlier message suggested was > missing and could only really be done well from the raw data you do not share > so should be present in the export. But perhaps I am wrong. Do you actually > have a running instance of your service using data exported via your API? If > so, how do you generate the horizontal accuracy estimate in the API from the > export database values?
Our database schema has a field called range, which is the same value we export as the range field. We use this value as the accuracy value if there’s only one cell/wifi network being used and use the distance to the farthest away cell/wifi position if we get multiple values. We than put in a minimum threshold of 5km (cell) / 100m (wifi). The code is fairly simple at https://github.com/mozilla/ichnaea/blob/master/ichnaea/service/locate.py#L39 This isn’t a terribly good approximation, but you get the exact same via the export data as what we have today. > So, with the OpenCellID web service, I can individually identify the quality > of each location estimate. I don't see how to do anything vaguely like that > with the exported data format. As I try to explain in the response to Felix, we have a different view on privacy than OpenCellID has. This view means we cannot share the raw observation data, which does indeed severely limit how much the community can help in data quality assessment. But we have already gotten good ideas from two users looking over the existing data and finding oddities. I’m hoping to find better ways to allow collaboration on the data level in the future, but that’s not been a focus to date. > Good luck in your work and have great fun, Thank you and thank you for all your contributions. If you are indeed interested more in working with the underlying data, at present the OpenCellID project might be more interesting to you. They are also planning to collect WiFi data later this year. Hanno _______________________________________________ dev-geolocation mailing list [email protected] https://lists.mozilla.org/listinfo/dev-geolocation
