[Talk-GB] man_made=survey_point

[SK53]

This approach has been advocated in other European countries, and the
Spanish community imported all the points of the national geodesic network
(e.g., for Extremadura
<https://www.openstreetmap.org/changeset/6041229#map=7/39.254/-6.124>).
They more or less violate the idea of OSM as something which is community
contributed (IIRC each point has "DO NOT MOVE") and often interfere with
objects which do need mapping (churches are a particular point). It's not
clear that this import has assisted improved accuracy of mapping in Spain.

Many trig pillars are now way out of alignment and mainly of interest as an
artefact. Even benchmarks might not have much relevance as OS surveying
mainly uses differential GPS with reference to their own base network (OS
Net
<https://www.ordnancesurvey.co.uk/business-government/tools-support/os-net/positioning>).
(From the OS website "Ordnance Survey (OS) benchmarks and their heights
haven't been regularly maintained for over 40 years.").

OS Net is effectively proprietary, there are a limited number of open base
stations for differential GPS in the UK. I do believe differential GPS
(RTK) has a role to play in OSM surveying, although for specific purposes
rather than generic improvement of feature alignment.

Regards,

Jerry

On Sun, 23 Aug 2020 at 10:05, Nick <nick at foresters.org> wrote:

> I have been looking at what is recorded under this tag in my area. I see
> that there aren't that many and those that are on OSM refer to trig
> points (see also http://trigpointing.uk/). My thinking is that if these
> are accurate and precisely marked on OSM then perhaps they could be used
> for resolving issue such as aerial imagery offsets.
>
> I therefore wondered if it was worth using other data under this tag -
> specifically benchmarks (https://www.ordnancesurvey.co.uk/benchmarks/)
> as there are huge numbers in the UK. If these were marked on OSM and
> their accuracy and precision verified (OS open data is to the nearest
> 10m square and transforming that adds errors), they could be helpful in
> local surveys where they are less than accurate but also for ensuring
> that moving all nodes in an area is valid (not just to match aerial
> imagery). A possible linked organisation with data is
> https://www.bench-marks.org.uk/
>
> Incidentally, the benchmarks can be helpful if you need to align
> historical maps which have benchmarks shown.
>
> Any thoughts?
>
>
> _______________________________________________
> Talk-GB mailing list
> Talk-GB at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-gb
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-gb/attachments/20200823/909af92c/attachment.htm>