[Talk-GB] UK addressing project

[Steven Hirschorn]

Just a quick note to say I'm really excited by the idea! The address base
shouldn't be locked up as a proprietary dataset and if it takes
crowdsourcing it to open it up, this looks like the best available approach.
I tried using multiple colocated UPRNs as a way to identify flat
conversions in my local area, but other than tagging the building as being
flat conversions, the effort of actually identifying whether flats were
addressed as "Ground Floor/First Floor" or "32A/32B" etc proved quite
convoluted. I don't suppose you have a way of getting the list of addresses
that need to be applied in any area from an open/license compatible dataset?

Steven



On Mon, 6 Sept 2021 at 23:04, Rob Nickerson <rob.j.nickerson at gmail.com>
wrote:

> Hi all,
>
> I've mentioned this on and off over the last few months, but with all the
> pieces now in place, I am pleased to say that OSM UK is kicking off a UK
> addressing project. Ed beat me to this email so rather than me retyping it
> all here, head over to
> https://blog.opencagedata.com/post/supporting-the-osmuk-addressing-project
> to find out more.
>
> In short, we are working with Russ Garrett <https://russ.garrett.co.uk/>
> and Pieter Vander Vennet
> <https://openstreetmap.be/en/motm/2018/01/12/pieter-vdv.html> (both long
> standing OSMers) to develop a new way to crowdsource address data within a
> specialist, simple editor. We are planning to use OGL data as a starting
> point to get lat/lon coordinates of suspected addressable locations and ask
> the crowdsource community to check and add all the tags.
>
> Below are the topic areas that we are aware that we will need to address
> (pun not intended!):
>
>    1. How to get the best quality set of suspected addressable locations
>    (Land Registry and Generalised buildings vs UPRN data)
>    2. How to manage integration with existing OSM data (filtering).
>    3. How to make the user interface as simple / intuitive as possible.
>    (It will be based on MapComplete)
>    4. What data we actually want to collect (e.g. easy, observable data)
>    5. How to promote the crowdsource project to as wide an audience as
>    possible...
>    6. ... whilst maintaining good quality contributions that are
>    compliant with OSM (e.g. not copying from protected databases)
>
> Firstly, have we missed anything? And secondly feel free to start chipping
> in with answers / comments on these. It could get too much for this mailing
> list so feel free to get something up and running in a collaborative doc or
> on OSM UK's loomio site. (If nobody beats me to that, I'll get that done in
> the coming days).
>
> I hope our approach is welcomed and does not fall into some of the traps
> that cause worry. For example, we are crowdsourcing not importing, and by
> keeping it simple we should minimise / avoid bad edits by newbies, etc. All
> comments, questions, and so on are welcome.
>
> Thanks,
> *Rob*
> _______________________________________________
> 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/20210907/38196619/attachment.htm>