[Talk-GB] UK address project update

[Rob Nickerson]

I want to pick up the conversation on which points we should use. To do so,
a quoted section of a thread (me then Gregory in reply) follows:

=== Start quote ===

>* Finally, where the red points fall short are in dense city centres
*>* and areas with a high concentration of council houses. This is
*>* because of a lack of cadastral parcels in these areas (e.g. one large
*>* cadastral parcel for the whole area rather than individual cadastral
*>* parcels for each home or business unit). We are keen to get your
*>* ideas as to how to work around this. So far our ideas are:
*>*    1. Identify these locations and direct experienced OSMers to them
*>* (leaving the easy places for the new mappers we hope to attract) to
*>* add the address by ground survey.
*>*    2. Or same as 1 but instead of the experienced OSMer having to do
*>* a ground survey (as might not live locally) allow them to "pull" the
*>* green dots through into the final user interface so that local people
*>* new to OSM can collect the address. This could be done by adding
*>* these to OSM; that is adding points to OSM that just have a UPRN or
*>* similar tag at first so that they can be made available in the map
*>* editor we are working on.
*>* Feedback and alternate ideas welcomed. I am particularly keen to know
*>* if you support idea 2 as typically we have avoided this in the past.
*>* For example we did not add the postcode centroids to OSM. This time
*>* we can add an addressable location with a UPRN, city, postcode and/or
*>* street name. The final details (house name / number) would then need
*>* adding via the crowdsource project.
*> I'm not particularly keen on option 2. I think there's a risk that it'd
open the floodgates to a large amount of data which may not get
reviewed. The resulting nodes might also get moved without the
consequences of that being fully understood by all mappers, risking
points being linked with the wrong address in the future. I wouldn't
want to risk the overall data quality of OSM; I'd generally favour a
slower approach, but keeping the quality bar higher.

Gregory

=== End quote ===

I'm not sure I agree that other mappers will move these points or
contribute to them in the wrong way. Not more than any mapper could move
any other node or way in OSM anyway.

The reason I like option 2 is that it keeps quality high whilst allowing
new mappers to work with experienced mappers. Let me expand:

The aim is to create a really simple interface just for collecting address
tags. We want to see a new group of mappers contributing who might not be
interested in learning the other OSM editing software / apps. As we want to
keep the quality high, we need to make sure the points they are presented
are high quality. Unfortunately I think the Open UPRN dataset has too much
noise and risks false positives. I did consider allowing users to toggle
between the red and green points, but this just makes it harder and doesn't
get around the high noise in the green points anyway. But using just the
red points is not ideal either - contributors will be confused to see some
addressable locations missing. Option 2 gives those experience OSM
contributors a option to "pull through" some of the green points so that
those new contributors get a better experience and ultimately can help us
collect more addresses.

Now one option would be for experienced OSM members to "pull through" these
green points in to the (non-OSM) database backend that the app will get the
red points from. However this creates another challenge;- the app will use
Land Registry polygons to check for any duplication with existing OSM data.
All the green points we'd want to "pull through" fall outside of the Land
Registry polygons (we use LR polygons of sizes appropriate for homes, not
the mega large LR polygons for council owned land) so don't then get
covered by our de-deduplication process. This is why I think it makes much
more sense for experienced OSMers to add these green points to the OSM
database directly where we can check for any existing OSM address data.

A compromise option might be for the experienced OSM mappers to keep an eye
on where any new address mappers are contributing and pull these extra
green points through only where we feel we have an active mapper. Or
perhaps we can have the address mappers click a button to ask for help
preparing an area they plan to survey.

How does that sound? Might keep us on our toes as the aim is to attract
enough new mappers that we are adding addresses much quicker than the less
than 10% we have added in 17 years of OSM :-) [That's not meant negatively,
we have clearly excelled in many many other places but for addresses there
is a long way to go and I want to crack that]

P.S. @Michael: As noted above the aim is to pick red or green (or
combination) for use in the app. As such the colours won't last beyond this
decision. But thank you for the reminder on colour blindness. Will try to
keep it in mind for final AND draft work going forward.

P.P.S. @Tony: What to collect is an important topic, but not one I can yet
put my mind to. I would like to decide on the approach to getting the
addressable points into the app and avoiding duplication with existing OSM
data first. Once we've got that, I can free up my mind to the nuances of
what to collect (if anything) beyond house number, name, streetname. One
thing at a time :-)

Thank you
*Rob*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-gb/attachments/20211115/371276d7/attachment.htm>