On 28 June 2011 15:34, Ed Loach <e...@loach.me.uk> wrote: >> All this from a quick look at my village. We cannot rely on this > data >> without a survey. > > Near here I zoomed in on the ways that are in VMD and not OSM. We > have footways along the seafront where the black lines are (not > worked out what type that is in VMD - perhaps I need to look at the > VMD layer for a colour code). There is also a black line in the > middle of the area which I'm sure isn't a way. We have a footpath > following part of it, so perhaps a farm tack. I'll need to resurvey > to be sure though. > > It isn't just the roads layer which is a bit dubious. I was looking > at one of the other layers (I forget which now) which marks nodes > for points of interest (perhaps churches, schools and leisure > centres IIRC). I thought this might be interesting to compare VMD > with OSM but picking a few nodes at random in whichever package I > opened the VMD layer in (QGIS perhaps) I stumbled across a "leisure > centre" which is in reality a garden centre called "Outdoor > Leisure". > > So yes, I'd agree, while the OS data can be useful in seeing what > needs surveying, I'd be a bit worried about trusting it without one > (though in combination with Bing imagery some of the obvious > discrepancies can be ignored), though you do seem to have > encountered more issues with it than I have, so perhaps quality > varies from area to area. > > I'll confess when OS Locator first became available I tagged a few > road names using that and Streetview but have tried to visit them > all since to confirm they are correct. It's a bit like anything > tagged source=NPE; I'm still trying to track all those items down > and either survey them where they are on publically accessible land > or the best I can from Bing if not (one highway=road was a farm > track with a gate at the road end I couldn't get past, and only > recently did I get around to updating the way from Bing).
I think it will be worth looking at the data and the views quite closely over the next couple of days. I have found a few small errors in our interpretation in the past hour. We were missing OS pedestrianised streets for one view. One thing I do find curious is that service roads that are missing on VMD appear on OS Streetview. For example here: http://www.itoworld.com/product/data/ito_map/main?view=136&lat=51.88970170371981&lon=0.932458821175924&zoom=16 So again we are finding inconsistencies between different OS datasets (ie OS Streetview/OS Locator and VMD). As with any other data source this one isn't perfect and it should be used alongside our local knowledge and other available sources. Regards, Peter > > Ed > > > _______________________________________________ > Talk-GB mailing list > Talk-GB@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk-gb > _______________________________________________ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
