Re: [Talk-GB] UK street addressing

[David Woolley]

On 21/12/2020 11:14, Richard Fairhurst wrote:
More philosophically, post towns violate the “on the ground” principle. 
No one here writes their address as Chipping


Addresses used by local people can also violate the on the ground 
principle. The place name I was given when I moved in appears to have 
been invented by estate agents, and doesn't appear on maps.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] UK street addressing

[David Woolley]

On 20/12/2020 23:21, SK53 wrote:
I'm aware of a number of terraces which are discontinuous, demonstrating 
that individual houses in a terrace are not building:part.


There is a set of maisonettes, which are both semi-detached 
horizontally, and split into four groups, with roads between them, 
around a roundabout, near me.  The individual maisonettes are numbered 
separately from the main street numbers, the development as a whole has 
no street number of its own, and actually has presences on roads with 
three different names!


It actually confuses the council's fly tip and street defect reporting 
app, which tends to reverse geocode it as though the maisonette's were 
directly numbered on the relevant road.  They failed to find one broken 
sign because of that.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Tagging bike ramp/ bike path down steps

[David Woolley]

On 13/12/2020 19:05, Edward Catmur via Talk-GB wrote:
Also, the steps should have bicycle=dismount, not =yes. This will allow 
people who can't dismount to go around by the road.


Only if it is illegal to try to cycle up and down the steps.  Otherwise 
it is the duty of the renderer (router) to infer that this will be 
necessary because of the steps.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] driveway-becomes-track

[David Woolley]

On 12/12/2020 21:11, Martin Wynne wrote:
What I'm wondering is how the typical recreational country walker would 
find that map, 


Your first problem would be establishing a funding model for it; OSM, in 
general, is not funded to a level that would support large scale end 
user use.


> or get it on their mobile phone app in place of the awful

Google maps? It's a lot of work to create if no-one ever uses it?


Google maps, as used in phones, are vector maps, with the rendering done 
in the phone itself


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] driveway-becomes-track

[David Woolley]

On 12/12/2020 13:20, Nick wrote:

Would changing this to Tag:highway=bridleway be a starting point?



I think the OP was saying there is a separate bridleway, almost parallel 
to the feature in question.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Bridleway across field

[David Woolley]

On 08/12/2020 15:11, nathan case wrote:

I am interested as a path I recently mapped is a PROW but is very dangerous to 
cross. It is now marked as disused:highway=path with 
access=discourged;designated but it is stilla PROW (byway open to all traffic 
in this case):https://www.openstreetmap.org/changeset/93427676


In that example, "Cross Bay Walk - DO NOT ATTEMPT" violates "name is 
only the name".  It may or may not be possible to justify "Cross Bay 
Walk", but the "DO NOT ATTEMPT" is not going to be a valid part of the name.


Unless there is a sign saying "unsuitable for pedestrians, horses, and 
vehicles", or similar, I would say "access=discouraged" violates "do not 
tag for the renderer".  The wiki specifically says that an official sign 
is required before using "access=discouraged".


"warning" appears to be non-standarised, and also subjective.

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] No U-turn on a long stretch of road

[David Woolley]

On 05/12/2020 12:39, Edward Bainton wrote:

Any established tagging system?

The turn restriction wiki 
 envisages 
turn restrictions at junctions only; my case is along the length of a 
major road (~3km). There's no barrier to prevent it, but presumably 
routing engines ought to know that route correction after a wrong turn 
will have to wait until the next roundabout.




That's a subjective judgement, so would be tagging for the renderer. 
The renderer (in this case a router) should be using some sort of 
heuristic like only  permitting U turns on residential or service roads, 
and giving a heavy weighting to the use of formal junctions at the the 
expense of distance travelled.  A real human would consider actual 
traffic levels and sight distances but they would be difficult to 
capture on the map and time and season dependent.


I think No U Turn signs tend only to be used where traffic volumes or 
junction structures, might otherwise suggest U turns were acceptable.


I don't think any driver (or autonomous vehicle) should be making U 
Turns based solely on the instructions of an automated router.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Idea - OSMUK walkers' map application

[David Woolley]

On 04/12/2020 16:38, Nick Whitelegg via Talk-GB wrote:
However as you say council take up could be problematic. Maybe we could 
provide a link to FixMyStreet?


Some councils insist that problem reports only come through their own 
web sites, or reluctantly, by phone, and will ignore emails (which is 
the default presentation for FixMyStreet).


The web sites generally provide structured input, whereas FixMyStreet is 
generally free text, and also, the web site sometimes bypasses the 
council contact centre, and goes direct to the out sourced contractor.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] High quality NLS imagery of buildings and HOUSENUMBERS (!) available in London (and Scotland). Create a tasking manger to add this?

[David Woolley]

On 01/12/2020 10:11, Tom Hughes via Talk-GB wrote:

Of course it's only claiming they do have a copyright that they
can make such a license necessary.


Not necessarily.  If you can establish a contract at every stage in the 
chain, you may be able to impose restrictions that go beyond copyright.


More interesting here is whether it is actually copyright or database 
rights that are at stake.  I think scanning only creates a new copyright 
on the typographical arrangement, and doesn't affect the database 
rights.  However, because there may be contractual restrictions, and 
because it might put OS into a bad position with their supplier, I would 
suggest that one obeys the restrictions that OS are imposing.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Indicating an information board is broken?

[David Woolley]

On 24/11/2020 17:28, Ken Kilfedder wrote:


They'll probably fix it, and the map can stay unchanged.



Although, either way, I don't think this is something to map in OSM, 
councils seem to consider fixing street name signs very low priority.  I 
think one argument they use now is that everyone has sat navs.  That 
causes a chicken and egg problem if OSM is the source for the sat nav!


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] electric fences

[David Woolley]

On 23/11/2020 22:13, Gruff Owen wrote:
For Public RIghts of Way, it is highly unlikely that this structure has 
been authorised by the Highways Authority.


Some West Country counties seem to accept electric fences across public 
footpaths, see the last item in 
.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Footways bikes can go on

[David Woolley]

On 21/11/2020 15:46, Mateusz Konieczny via Talk-GB wrote:

there is also bicycle=permissive (based on access=permissive) for
"permitted right now but can be revoked/changed at any time"


The way seems to be in a park, and, in general, permissive is the 
maximum legal status of any path in a park, unless it is also a 
bridleway or public footpath, in the definitive map.




In general modelling "clearly illegal but accepted and normal" is 
problematic

for access/parking tagging in OSM.



There is a modal filter near me, on a temporary traffic regulation 
order.  It has been flouted for all the three months that it has 
existed.  However it is clearly signed as emergency vehicles (and 
non-motor vehicles) only.  In that case accepted use shouldn't represent 
how it is mapped.  (It also has enforcement camera signs, and it might 
be interesting to find how many fines they collect if they do install 
the cameras.  I suspect the abuse will stop until they are moved elsewhere.)


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Service road with private locked gate and routing apps

[David Woolley]

On 16/11/2020 11:18, Mat Attlee wrote:
Upon surveying this service road it is very much closed to the public 
with locked gates which I marked as thus 
https://www.openstreetmap.org/changeset/93935943


However these routing apps still use this service road. Have I missed 
something or does it take a while for the changes to propagate?


It takes time for routing engines to update.  However, I would also have 
put access tags on the service road.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Multi-lingual tagging in Wales

[David Woolley]

On 16/10/2020 14:08, Gruff Owen wrote:
With that in mind, and admittedly polemicising the debate a little. If 
we accept the premise that the native language of Wales is Welsh and 
that OSM is a community mapping project where we have an opportunity to 
respect native communities in a way that past colonial mapmakers didn't. 
Could we take this as an opportunity to prioritise authentic Welsh place 
names where that's possible? I understand that there will be objections 
to this, but I'm not sure we can disregard it completely as an option?




My understanding of how it works is that it is up to the local 
communities to ensure that road signs, etc., in the local area, reflect 
the community preferences, and OSM will reflect whatever the signage 
says.  This is even more important in areas where people are shelling 
each other over such issues.  Using what is on the ground is the only 
way that OSM can avoid taking sides.


There is nothing to stop a Welsh language supporter running a map tile 
server that uses name:cy, in preference to name, where it exists.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Blocked / overgrown / inaccessible footpaths and bridleways

[David Woolley]

On 29/09/2020 14:29, Gareth L wrote:
I’ve prioritised tagging width values on canal towpaths in some 
locations where, whilst legal, it’s precarious to try and cycle along as 
they’re practically less than a metre wide.


Unless you are just tagging with the actual width, I'd suggest that 
would be a dangerous precedent.  I've seen people wanting to invalidate 
official cycle routes because they think the traffic is too dangerous.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] UPRN Locations Map

[David Woolley]

On 26/09/2020 13:06, Russ Garrett wrote:

There is no legal obligation for FoI responses to be openly licensed.
The point of FoI is to make information available for inspection, but
not (necessarily) for reuse.


To expand on that.

Larger UK companies tend to be very intellectual property based, so 
making information freely available is never going to be a government 
objective.


The actual objective will be more towards open government; ensuring that 
decisions, and the information behind them are open to scrutiny.  A 
secondary purpose is probably to try to ensure that the taxpayers have 
the results of work paid for from their taxes.


OS are in a funny position, in that they are in the public sector, but 
are expected to be self funding.  To the extent that they succeed in the 
latter, they don't owe a duty to the taxpayer.


Although FoI is often used as a tactic for obtaining information for 
republication, as the response points out, that republication isn't 
actually authorised by the FoIA; the information is provided for the 
personal use of the requestor.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Jewson - is it shop=doityourself or shop=trade?

[David Woolley]

On 18/09/2020 21:55, Mark Goodge wrote:

but B, Wilko and Wickes are consumer


My impression is that Wilko is genuine consumer, but B, is mix of 
consumer and informal economy trade (aka handymen) and Wickes is mainly 
a mix of formal and informal economy trade.


(I'm not sure to what extent handymen are really in the informal 
economy, or simply do properly taxed small B2C jobs.)


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Jewson - is it shop=doityourself or shop=trade?

[David Woolley]

On 19/09/2020 13:32, Phil Endecott via Talk-GB wrote:

i.e.
some chains may describe themselves in a way that allows
them to get permission to operate on cheaper industrial
estates rather than more expensive retail parks.  I don't
think that's very useful information for map users.


I think you are assuming the typical OSM user is the man in the street. 
My impression is that the heavy users are actually academic and 
professional, for whom planning type issues may actually be very important.


OSM isn't trying to be a retail directory, and the very patchy nature of 
its coverage for initial addition and updates, means it will not be a 
good one, except in small areas, in the foreseeable future.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Brexit and OpenStreetMap

[David Woolley]

On 14/09/2020 14:41, Andy Mabbett wrote:

Change sets and item histories contain user names, for example.


If those don't fall under some sort of exemption, you have rather more 
fundamental problems than Brexit; you probably can't make the map 
available outside the EU without some sort of NDA.


An analogy would seem to be film credits.  Google didn't provide 
anything helpful on that.  Most hits were about tax credits!


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Brexit and OpenStreetMap

[David Woolley]

On 14/09/2020 13:51, Tony Shield wrote:
By thinking of moving OSMF from UK to EU because of Brexit are you 
saying that OSMF may never be able to function outside the EU - what 
about Switzerland where many international organisations are based, or 
United States. These are respected countries which should be considered 
if relocation is deemed necessary.




The United States is generally understood to have very weak data 
protection laws, but still manages to operate within Europe, although 
sometimes using Irish or Luxembourg proxies.


With respect to data privacy what is to stop OSMF mandating in its 
contracts and operation that the relevant EU data laws are adhered to. 
Maintaining data integrity and security is a function of OSMF, these 
functions are mandated by EU law, OSMF wherever it is domiciled can base 
its operations on the implementation of EU law.


I imagine that is the way that most big organisations will go, not just 
on privacy.


Having said that, is it actually the case that data protection law is 
being revoked at the end of the year.  I doubt it.


Of course, the map itself should not contain any personal data, as, even 
when based in the UK, I don't see how adequate controls could be applied 
to unpaid volunteers.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] New Bing Imagery

[David Woolley]

On 19/08/2020 16:21, Russ Garrett wrote:

5m accuracy


You'll accumulate that error in a couple of centuries, just from 
continental drift: 
, 
given that OSM is referenced to WGS-84, not the British mainland.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Proposal: Import EV charging point data

[David Woolley]

On 18/08/2020 00:11, Steven Hirschorn wrote:
I'm hoping to import a dataset of EV vehicle charging points in London. 
I've created a wiki page here: 
https://wiki.openstreetmap.org/wiki/SourceLondon




There was a lot of discussion about importing EV charging station data 
recently, so the first thing you should do is go through the list archives.


Also, note that you need the agreement of the community before you start 
any import.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] National Cycle Network removal/reclassification

[David Woolley]

On 14/08/2020 19:14, Simon Still wrote:
I’m not sure that’s actually a legal status that changes anything - 
pedestrians have priority on all shared use paths so not sure that tag 
would add anything


Towpaths are privately paths (currently owned by the Canals and Rivers 
Trust), so the rules for public paths don't apply.  At one time  you had 
to apply for a free licence to cycle on them, the quid for quo for which 
was a promise to do things like give pedestrians priority.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] National Cycle Network removal/reclassification

[David Woolley]

On 14/08/2020 12:46, Mateusz Konieczny via Talk-GB wrote:
If signage on the ground is gone or never existed then route relation 
should not be mapped in OSM*.


In the long term, this could make OSM useless for motor traffic as there 
is a general policy of decluttering signs.  One of the arguments for 
that is that everyone uses satellite navigators, so they don't need the 
signs.  I think is also used as an argument for why it can take councils 
years to fix missing street name signs.


If OSM relies on on the ground signage, when the authorities rely on 
virtual signage in online maps, it could lose a lot of roads!


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Using OSM as a base for my own fictional map?

[David Woolley]

On 24/07/2020 23:54, Martin Wynne wrote:


The downloaded PDF files are vector files which can be zoomed to any 
level without pixelating, and can have the internal records modified as 
required.


The only vector files where you can reliably detect things like railways 
are those produced by the OSM APIs.  The PDFs will change as ideas on 
how to render the map change.  On the other hand, they contain no 
rendering, just wire frames.


Also, generating a PDF server side is a relatively expensive, so don't 
expect to welcomed if you start doing this on the fly.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Surveying rural buildings

[David Woolley]

On 23/07/2020 10:12, Nick wrote:

Do we actually know what the general public use OSM for?


My impression is that the target for a lot of the material in OSM is 
professional users of maps, rather than the general public.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] POI files of Pub/Restaurant chain

[David Woolley]

On 15/07/2020 14:00, o...@poppe.dev wrote:

As this data is pretty much openly accessible, I think there'd be no major 
issue with asking them if this data could be used to check all the places 
against OSM data and, if needed correct and/or create them, right?


I often find that business locations are way off target.  I think head 
offices often just plug the postcode into Google Maps, rather than 
asking someone to read them on their GPS capable mobile, or match them 
to actual features on the local map.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Holiday camp tagging

[David Woolley]

On 13/07/2020 18:19, Cj Malone wrote:

2 - The site is split into areas with names, I've started adding
address details to aid in routing, mapping them as addr:housename=area
and addr:unit=#. I was originally planning to use addr:place instead of
addr:housename, but this specific camp site uses real local places


I would consider that a misuse of addr:place.  Addresses are basically 
things to which you could send mail.


If separate areas have been mapped for tents and caravans, you should 
not coalesce them.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Paths on Wimbledon Common

[David Woolley]

On 10/07/2020 13:11, Colin Smale wrote:
What does "legally accessible" mean? Are they Public Footpaths? Do we 
tag all Public Footpaths with an explicit "foot=yes" or is 
"designation=public_footpath" enough?




I don't know the situation in Wimbledon Common, but most footpaths in 
public park are more correctly described as access=permissive.


My understanding is that designated only has meaning if combined with an 
access type tag with a value of designated.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Bus Routes on OSM

[David Woolley]

the Open Government Licence which I believe is compatible with OSM 
https://wiki.openstreetmap.org/wiki/Open_Government_Licence.




As the wiki says, you can't rely on this without further enquiry.  The 
problem is that it only gives rights to intellectual property owned by 
the government, and datasets often include intellectual property from 
other organisations, e.g. TfL in this case.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Bus Routes on OSM

[David Woolley]

On 06/07/2020 14:02, Matthew Scanlon wrote:
How are Bus Routes added into OSM? I have noticed that bus routes in 
Basildon (my local area) are a few years out of date with some service 
such as the 5 and 8Ahaving been  withdrawn and the route 2 being 
renumbered 28




My understanding is that TfL doesn't licence the information on a basis 
that would allow it to be directly used, so OSM rely on members of the 
public using the buses, or tracing the routes between bus stops.


How up to date the information is depends on how enthusiastic people are 
in an area, and there tends to be a preference for mapping things for 
the first time over maintaining existing mapping.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] UPRN Locations Map

[David Woolley]

On 04/07/2020 18:24, Lester Caine wrote:
The current 'OHM' is not a layer that can be easily combined with the 
current 'OSM' layer. Large sections of the current data are simply 
cloned into OHM


I'm not referring to OHM; I'm referring to the main OSM map.  At least 
since September 2012, OSM has the complete back history, and, as far as 
I can see, you can use overpass API to retrieve the map as of any date 
and time since then.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] UPRN Locations Map

[David Woolley]

On 04/07/2020 11:45, Lester Caine wrote:
At the very least data currently live in on a 'current' view should be 
automatically filed to an historic layer when it is replaced


How does this differ from how OSM already works?  You can already create 
versions of the map at any point in its history, except where data has 
been redacted for legal reasons.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] UPRN Locations Map

[David Woolley]

On 03/07/2020 11:23, Tony OSM wrote:


There was a reference to £1000 worth of data being made free each month 
to individual users - can't find out how this works yet. This may allow 
us as individuals to populate OSM and OSM essentially aggregates the 
data - rather like postcode data.


Generally reconstructing a database piecemeal is considered no different 
from copying a substantial part in one go.


I suspect the reality with postcodes is that the the Post Office doesn't 
see collection of individual codes from the relevant business or 
resident as a significant risk, as they don't believe anyone will 
construct a sufficiently complete database that way.


Unlike natural features, both postcodes and telephone numbers start 
their life as entries in databases, but both of them need to be known to 
contacts of a person without the need for formally entering into a 
licence agreement.


However, if every user used their free quota of postcode searches each 
day, I think the Post Office would take action.


Normally when limited searches are allowed, there is also a personal use 
only restriction applied.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] "secret" site

[David Woolley]

On 27/06/2020 23:37, Dave Love wrote:

I was going to map a covered reservoir round here that I've known from
my youth, but I happened to find an article about it from the local
paper suggesting the location is secret, though it's listed in
Historic England.  (It's not far from a "sensitive government
establishment" that no-one locally knew about before it ceased to be
secret :-/)
Should I not map it, or pretend I didn't see the article?



How would another mapper verify this?  I normally think that people 
interpret the map only what's on the ground rule too literally, but in 
this case, I'd suggest the moral thing to do is to ignore your local 
knowledge an only map what can be seen, without guessing its significance.


(There are some road signs entitled "secret bunker", e.g. for Kelvedon 
Hatch, but I don't think this is the case here.


On the other hand, there used to be part of the North Yorkshire moors 
that had "undefined" written over it on OS maps.)


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] TfL Cycle Infrastructure Database - matching against OSM

[David Woolley]

On 21/06/2020 13:38, Mateusz Konieczny via Talk-GB wrote:

Is it ok for pedestrians to walk on
the carriageway and cross the road
together with cyclists in place
marked by bicycle paintings?


It's legal for them to do so, which is what determines access.  They 
don't get any priority.




___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Farmfoods clean up

[David Woolley]

On 27/05/2020 11:48, Cj Malone wrote:

their homepage exclusively shows ambient products


All self service stores, unless completely sold out, have ambient 
products, pretty much by definition, even open air ones at the South 
pole, if such existed, and including ones that only stock frozen 
products.  Catalogue stores might be an example of ones that didn't.


Did you mean room temperature?



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Ficticious embankments?

[David Woolley]

On 14/03/2020 18:09, ael wrote:

I have just noticed some new "Embankments" added around a fortnight ago.
These were added to some stone circles in Cornwall which I know well and
have extensively surveyed. There is no trace of any embankments. No
source was given and the user does not appear to be local.


I suspect this is a case of tagging for the renderer, which is, of 
course, wrong.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] railway=halt

[David Woolley]

On 01/02/2020 09:26, Philip Barnes wrote:

In GB halt has been used to indicate request stops.


That was certainly the context in which I came across them, in this 
case, Warrenby Halt, on the, then British Steel iron works at Redcar, 
which has now been mothballed, after only a short life.


That one was actually on private land.  It used to be about here 
. 
 It may well have gone, but, also just might never have been mapped. 
(Please note that the relationship to features now mapped may be wrong 
by up to a few hundred metres.)


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Landuse between fences?

[David Woolley]

On 01/01/2020 11:58, Martin Wynne wrote:
So where is the definitive specification? The only practical way to 
discover if something is valid seems to be to see how the standard map 
renders it.


There are a large number of things that are perfectly valid that are nor 
rendered by the standard Mapnik layer.




If it renders ok, the assumption must be that it is acceptable mapping. 


That pretty much defines tagging for the renderer.  The map isn't any 
two dimensional image, but the actual data from which that image was 
generated.



Otherwise, why call it the "Standard" map?


Loose terminology, a convenient shorthand.



It is rather more than a demonstrator, it is used in many places for 
actual use, see for example:


That's because people make a choice that it is adequate for their 
purpose.  Part of that choice may  be related to not wanting to spend 
time searching for something better, or to use someone else's tile 
server.  (In practice, the site you quote is probably using a snapshot 
of the "standard" style, and not the current version.  They need to 
create their own tiles for fair usage reasons.)


The standard rendering tends to follow, rather than lead, actual usage.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Landuse between fences?

[David Woolley]

On 01/01/2020 07:13, Martin Wynne wrote:


However, I have discovered that highway=track, *area=yes* is valid - as 
evidence of that it is rendered on the standard map as a light brown 
infill between the fences with the existing highway=track as a routable 
way superimposed over it, in darker brown.


That's tagging for the renderer, which you should not do.  If you map a 
highway as an area, you should not also map it as a line.  It is up to 
the router to deal with that.  If  you double map, there is no incentive 
for routers to deal with areas properly.




It seems odd to have highway=track twice, but if that's what it takes to 
have a meaningful mapping for an area of land, I'm happy to do it that 
way. Presumably the developers of the standard map know what they are 
doing.


The standard map doesn't claim to be a definitive specification of what 
is allowable.  It is a combination of a technology demonstrator and a 
background map to assist mappers in placing other features.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Landuse between fences?

[David Woolley]

On 01/01/2020 00:49, Dave F via Talk-GB wrote:
I was trying to intimate, *personally*, I wouldn't bother obsessing with 
mapping every *square inch* of land.


I also don't think you should be mapping in that detail, but if you 
really want to, I would suggest that you map the wide area with just 
landuse, and then do nested mappings of the fields, with the crop, etc., 
type as well.


Similarly, I wouldn't want a residential estate broken up by road corridors.

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] New Entertainment venue - what tags?

[David Woolley]

On 29/12/2019 10:40, Tony OSM wrote:

What is the best way to tag? One node or three nodes?


Three *areas*, plus the building outline.

Really it depends on how much detail you have about the internal layout. 
(If there are individual administrative offices, in a different part of 
the building, and you know where those are, multipolygons might even be 
appropriate.)


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Roundabouts one piece or segregated

[David Woolley]

On 23/12/2019 18:15, Nick Allen wrote:
I may be missing something here, but 
https://www.openstreetmap.org/way/477263099 looks okay to me.





The OP was proposing that , 
, and 

Re: [Talk-GB] Roundabouts one piece or segregated

[David Woolley via Talk-GB]

On 23/12/2019 03:08, Warin wrote:

I'm looking at Wivenhoe B1028 way 477263099.
This is a segment of a roundabout.

Would it not be better for the way to be a single feature in OSM?


It is rarely a good idea to revert from more concrete to more abstract, 
so I would say no.


Incidentally, many things mapped as roundabouts aren't legally 
roundabouts, but rather traffic circles.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] No Through Road Ahead

[David Woolley]

On 19/12/2019 14:06, Martin Wynne wrote:

How to tag this road?

https://goo.gl/maps/B4kUxoR83ej9JXWQ8

There is no actual barrier, just a very sharp corner.


You tag the corner, not the road, as the sign is only advisory.

Unfortunately, I suspect the current tagging scheme may have difficulty 
encoding the turning circle restriction, especially in a way that 
routers understand.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] barrier=kerb on highways may be blocking OSRM (Car) routing

[David Woolley]

On 18/12/2019 15:59, Robert Skedgell wrote:

It's parking a car on a
footway which is illegal in London (an offence which is only subject to
civil enforcement), unless explicitly allowed by the local authority.


It's potentially a criminal offence anywhere see sub-paragraph 17 of 
.  I think the 
situation in London is just that the default position is reversed, and 
everywhere is assumed to have a notice by default.  The offence is 
crossing the kerb or verge, not parking on the footway, which is a 
separate offence.


I think issue of civil enforcement is just that the police have 
abdicated all this sort of thing to civil enforcement, rather that it 
isn't a crime.  About the only parking offences the police will 
prosecute are dangerous and obstructive parking, but they could 
prosecute any of them.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] barrier=kerb on highways may be blocking OSRM (Car) routing

[David Woolley]

On 18/12/2019 13:31, Edward Catmur via Talk-GB wrote:
That said, the same goes for cars - other than the lowest bodied sports 
cars, pretty much all motor vehicles are capable of taking a kerb at low 
speed.


Although raised kerbs are generally there to stop that happening and the 
resultant trespass on the footway can be illegal, e.g. in London.  As 
such routers should not be routing motor vehicles over kerbs.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Disused or empty apartments prior to demolition

[David Woolley]

On 17/12/2019 20:35, Warin wrote:


so
building=apartments
becomes
disused:building=apartments

or
building=yes
becomes
disused:building=yes


I disagree.  It is still a building.  In fact some of the most 
interesting buildings are disused ones.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] What is farmland?

[David Woolley]

On 16/12/2019 10:07, David Groom wrote:
I see no benefit to mapping individual fields as separate polygons 
tagged as farmland if adjacent fields are also farmland. Could you 
explain why you think this is best?




I see no reason why mapping individual fields would not be an objective 
for OSM.  In that case, tagging them as farmland seems the sensible 
thing to do.


As always, the level of detail will depend on the priorities of mappers, 
and if no-one in the area is interested in fields, the detail might be 
limited to tens of square kilometres.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Elections Online website - candidate for OSM?

[David Woolley]

On 03/12/2019 09:47, Edward Bainton wrote:


General Elections Online 
 (hosted 
at parliament.uk ) have got a failed page where 
the Google map is overlaid with "Development purposes only".


I was planning to suggest they use OSM instead.



The advantage to them of using Google is that Google provides the tile 
servers.  OSM tile servers aren't funded to support a mass market use 
like this, so the organisation will have to install and run their own 
tile servers.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Import of UK SSSI data

[David Woolley]

On 17/11/2019 22:37, Henry Bush wrote:
I am aware that SSSIs change, so my plan was that my bot would look for 
an existing entry first, and if it exists, either modify or delete it 
(if the latter, I'd verify the tags were the same or something first)


Delete and re-add is something that should not be done, as it destroys 
the history of the object.  Generally if you detect a duplicate, you 
should manually check it before doing anything.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] FIXME/fixme/OSm Notes Quarterly Project

[David Woolley]

On 06/11/2019 00:53, Dave F via Talk-GB wrote:


This one looks like a right mess given the loop with the bus stop is one 
way.

https://www.openstreetmap.org/relation/2580522#map=19/52.58839/-0.21216


It's also the wrong relation type.  route_masters should have route 
relations as their members.  This one is just a route.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Licensability of an employee's work

[David Woolley]

In particular, when submitting data obtained through employment, it has 
to be clear that the company is the one that is agreeing to the 
licensing terms, not the individual.



On 21/10/2019 12:47, David Woolley wrote:
I meant you should use an account that clearly belongs to the company. I 
guess you could have an account for each relevant employee, but that 
will cause problems when an employee changes job, either internally, or 
to another employer.


Definitely do not use the same account to submit personal contributions 
and company ones.



On 21/10/2019 12:37, Edward Bainton wrote:

Thanks, David.

Discussion ongoing on the legal list, but FYI from Frederick Ramm, who 
opines:


 > PS: I would strongly advise against using a "corporate account" that
 > groups the activities of many individuals as it makes communication
 > between the group/company members and other members difficult, and 
good

 > communication is a cornerstone of every successful organised editing
 > activity.

I don't know if that's precisely what you meant, but here for info 
(without judgment either way)


Edward

On Fri, 18 Oct 2019 at 20:08, David Woolley 
mailto:for...@david-woolley.me.uk>> wrote:


    On 18/10/2019 17:43, Edward Bainton wrote:
 > *If an employee edits the map in the course of their employment,
    has the
 > work been adequately licensed to OSM/the big wide Open?*
 >

    I think it is true worldwide that employers have the copyright in 
work
    for hire, and only they can licence the use of their copyright.  
If the

    map is being edited at the employers request, the employer should
    create
    an OSM account for such purposes.

    In the UK, if you day job is producing copyrighted maps, you will
    almost
    certainly find that anything you attempt to do on OSM comes under the
    employer's copyright.  California, in the USA, is a notable
    exception to
    this.


    ___
    Talk-GB mailing list
    Talk-GB@openstreetmap.org <mailto:Talk-GB@openstreetmap.org>
    https://lists.openstreetmap.org/listinfo/talk-gb





___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb




___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Licensability of an employee's work

[David Woolley]

I meant you should use an account that clearly belongs to the company. 
I guess you could have an account for each relevant employee, but that 
will cause problems when an employee changes job, either internally, or 
to another employer.


Definitely do not use the same account to submit personal contributions 
and company ones.



On 21/10/2019 12:37, Edward Bainton wrote:

Thanks, David.

Discussion ongoing on the legal list, but FYI from Frederick Ramm, who 
opines:


 > PS: I would strongly advise against using a "corporate account" that
 > groups the activities of many individuals as it makes communication
 > between the group/company members and other members difficult, and good
 > communication is a cornerstone of every successful organised editing
 > activity.

I don't know if that's precisely what you meant, but here for info 
(without judgment either way)


Edward

On Fri, 18 Oct 2019 at 20:08, David Woolley <mailto:for...@david-woolley.me.uk>> wrote:


On 18/10/2019 17:43, Edward Bainton wrote:
 > *If an employee edits the map in the course of their employment,
has the
 > work been adequately licensed to OSM/the big wide Open?*
 >

I think it is true worldwide that employers have the copyright in work
for hire, and only they can licence the use of their copyright.  If the
map is being edited at the employers request, the employer should
create
an OSM account for such purposes.

In the UK, if you day job is producing copyrighted maps, you will
almost
certainly find that anything you attempt to do on OSM comes under the
employer's copyright.  California, in the USA, is a notable
exception to
this.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org <mailto:Talk-GB@openstreetmap.org>
https://lists.openstreetmap.org/listinfo/talk-gb





___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Licensability of an employee's work

[David Woolley]

On 18/10/2019 17:43, Edward Bainton wrote:
*If an employee edits the map in the course of their employment, has the 
work been adequately licensed to OSM/the big wide Open?*




I think it is true worldwide that employers have the copyright in work 
for hire, and only they can licence the use of their copyright.  If the 
map is being edited at the employers request, the employer should create 
an OSM account for such purposes.


In the UK, if you day job is producing copyrighted maps, you will almost 
certainly find that anything you attempt to do on OSM comes under the 
employer's copyright.  California, in the USA, is a notable exception to 
this.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] accurate GPS

[David Woolley]

On 09/10/2019 14:40, Simon Ritchie wrote:
They often leave objects in the ground to protect them, and then come 
back a few years later to have another look using new techniques.  It 
would be nice if they knew precisely where their target is.


For that, you really need to record one or more reference points that 
are likely to stay there and keep their relative position in the long 
term, as what you are interested is the location relative to the 
surrounding, not an absolute position on the globe.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Import UK postcode data?

[David Woolley]

On 07/10/2019 14:23, Mark Goodge wrote:

The ONS website explicitly states that their postcode products are OGL


The OGL only applies to the parts of the data that relevant government 
organisation has the ability to grant rights to.  It excepts "third 
party rights the Information Provider is not authorised to license;". 
As such being OGL doesn't meant that you have a right to any RM data it 
may contain.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Import UK postcode data?

[David Woolley]

On 07/10/2019 11:43, Simon Poole wrote:

It's the nature of the beast that when we are discussing OGL licensed
datasets that when something turns up that was previously thought to be
part of a proprietary dataset all alarm bells go off. Do you know how
they derived that flag and if there is really no residual proprietary IP
from RM in the data?


Something you have to consider is the threat to RM.  RM won't worry too 
much if they are not loosing revenue, but they probably make significant 
revenue from licensing the postcode database to commercial mapping 
services, so if OSM starts to effectively compete, RM may look rather 
closely at how OSM sourced its postcode data.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Import UK postcode data?

[David Woolley]

On 04/10/2019 13:47, Dave F via Talk-GB wrote:
No. The centre point is not associated with *any* delivery point. It is 
an arbitrary mean, calculated mathematically. it could, in theory, be 
located in the middle of a park.
Even postcodes unique to one property/business aren't accurate as their 
positions are misaligned by the effect of adjacent areas.


Although I don't have a primary source for this, my understanding is 
that the median is snapped to the nearest actual delivery point within 
the postcode.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Import UK postcode data?

[David Woolley]

On 02/10/2019 13:57, David Woolley wrote:
Whilst I'm not sure of the precise conclusions, this has been considered 
many many times before.  I think it may even have been done in some 
places.  I'd suggest a search of the list archives.


Also, the discussing tab on the wiki page you referenced contains 
discussion on this going back over 9 years, and expresses concern about 
the licensing.


As a minor point, although described as centroids, they are not actually 
centroids, but actually the nearest valid delivery point to the centroid.


My feeling is that, if considered acceptable, it would already have 
happened.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Import UK postcode data?

[David Woolley]

Whilst I'm not sure of the precise conclusions, this has been considered 
many many times before.  I think it may even have been done in some 
places.  I'd suggest a search of the list archives.


Note that this data is not suitable for reverse geo-coding, because I 
don't believe it distinguishes between single business and area codes.


(Better would be to convince the general public that a proprietary 
locator system is not the best way of providing a short code for an 
approximate location.)



On 02/10/2019 13:43, Russ Phillips via Talk-GB wrote:

Hi,

I'm wondering if it would be feasible and advisable to import the UK 
postcode data from OS OpenData Codepoint 
.


The licence is OSM compatible. My thinking was that we could create a 
node for each data point and set the addr:postcode tag. This would be 
useful for routing software like OsmAnd, since it would allow a user to 
enter a postcode as a destination.


I'm happy to do the work, but the import guidelines 
 say that imports 
should be discussed on the imports@ list and the appropriate local 
communities, hence this email.


Russ Phillips



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb




___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Resurrecting the 'find the missing paths for 2026' project

[David Woolley]

On 30/09/2019 18:25, Nick Whitelegg wrote:
I made a start on this about a year ago, here's a quck mock-up showing 
council data in colours and OSM paths shown in white as a 'tippex' 
effect. This allows the identification of historical 'F.P' footpaths on 
the historical maps which do not correspond either to current council 
RoWs or current OSM paths, and thus would be candidates for 
investigation to see if the path is in a usable state or there is 
evidence of use.


Such paths are not going to have finger boards with "public footpath" on 
them.  In other threads, I sense quite a strong lobby for only mapping 
rights of way that are so marked on the ground and ignoring any 
designation that only appears in a map.


As such, you will end up with at best a permissive status recorded on 
OSM.  Even that is actually likely to be subjective.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] non-squared buildings

[David Woolley]

On 30/09/2019 11:47, SK53 wrote:
I imagine for accurately surveyed & designed buildings JOSM's algorithm 
is likely to introduce additional errors because the 
architects/engineers will have used British Grid.


Squaring to the grid is something you could only reasonably do for 
buildings oriented to one of North, South, East or West.  Given that, if 
spherical geometry has a significant effect, you are going to have to do 
more complex calculations, I imagine things will be done in the way that 
is easiest to implement with theodolites, etc., not one that produced 
round grid references.




___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Subject: Re: Thomas Cook shops

[David Woolley]

On 29/09/2019 15:30, Dave F wrote:
Preventing the mass (hardly "mass" though) edit of Thomas Cook & instead 
relying on individuals to update *will* guarantee more shops will be 
"wrong".


It could well actually have the opposite effect by getting people to 
audit lesser known businesses on the same street.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Subject: Re: Thomas Cook shops

[David Woolley]

On 29/09/2019 14:03, Jez Nicholson wrote:


I'm not keen on bulk automated closing everything called Thomas Cook 
because the world is more complicated than it first seems to be. I 
favour visual confirmation.




I think too much effort goes into these big changes.  The real problem 
with business directory mapping on OSM is that people like doing the 
first time mapping of shops on a high street but no one likes 
maintaining them. The number of shops that are wrong because of churn or 
small businesses, or individual closures of chain shops is probably 
orders of magnitude more than the ones that get lots of publicity.


I doubt that many people are going to get misled by a Thomas Cook or 
Maplins store that remains mapped, but many may be misled by the loss of 
a specialist store that didn't make the national press.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Subject: Re: Thomas Cook shops

[David Woolley]

On 25/09/2019 13:24, Andy Mabbett wrote:

   opening_hours = none


I believe the correct syntax would be:

opening_hours=closed

or even

opening_hours=closed "tenant being liquidated"

The evaluator accepts both, although gripes about the lack of public 
holiday rules.  It interprets the latter as:


Facility is always closed in the (near) future, comment: "tenant being 
liquidated"


Whereas it throws out "none" on seeing the second "n".

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Adding buildings and addresses

[David Woolley]

On 16/09/2019 00:02, Luciën Greefkes via Talk-GB wrote:
On the other hand, there are the Community Maps. They contain very 
decent shapes/contours of buildings. I would like to work with a layer 
like that to be able to compare with aerial imagery.
The community map of Welwyn Hatfield gives Ordnance Survey as the 
source. I have sent a request to OS if they'd be willing to share the 
data used for the community maps, because that will be a massive aid in 
my current mapping project.




I would imagine the maps are based on OSes high resolution mapping data, 
which is the data from which they make their money.  I would be most 
surprised if they give permission.


Generally the commercial resources to which OSM has access are allowed 
to OSM because they are not the best information available, so there is 
little loss to their providers in releasing the data.


If you want accurate building outlines, without paying licence fees, you 
probably need to buy or hire a laser scanning surveying device.  I think 
the prices may be eye watering.


You may find that some of the more recent OS OpenData has more detail 
than StreetView data traditionally used for OSM.  You will never get the 
best data.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] 'Sources' tags

[David Woolley]

On 07/09/2019 17:54, Edward Bainton wrote:
I've noticed the changeset data includes the aerial image used 
(presumably at the moment you hit 'save', if you've referred to several?)


Does this mean I don't need to add a source=aerial_imagery tag before I 
save a changeset? iD doesn't have it available as default; you have to 
choose it under 'add field'.




This is a feature of particular editors and I think it was introduced 
because people were failing to add sources.  Of course it can result in 
a source being indicated that wasn't really used.


In any case, source=aerial_imagery is too vague, one should really be 
using the actual imagery source.


If it was the only source (i.e. armchair mapping) and is correct, it 
shouldn't really be necessary to add a source, but if there were other 
sources (survey and gps are gold standards) they should be included, and 
if it is wrong, it should be removed.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Copyright in OS-derived maps

[David Woolley]

On 05/09/2019 05:48, Warin wrote:

If they had derived their data from OSM .. then all would be fine.


As I hinted before, the use of a red line, and a custom printout from an 
OS detailed map, suggests this is a map for legal purposes.  For both 
the Land Registry and council planning applications, a red line is the 
convention for showing a property boundary.


Until you can get lawyers, the Land Registry, and councils to accept OSM 
derived mapping, this sort of map is always going to be OS derived.


(Actually, at least for the Land Registry, the outline on the map is 
only indicative, not definitive.)


Another issue with the Land Registry is that all the outline data is 
combined into an index map that can be searched, but not viewed, by the 
public.  Combining OS and OSM derived data might cause licensing issues 
for that.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Copyright in OS-derived maps

[David Woolley]

On 03/09/2019 12:31, Edward Bainton wrote:
I've been sent a map by a local charity that looks after large swathes 
of countryside near Peterborough. It's for their own internal use, 
showing the extent of their estate. It's based on an OS map, and comes 
with flags indicating Crown copyright thus:


/Reproduced by permission of Ordnance Survey on behalf of HMSO. ©Crown 
copyright and database rights 2010. All rights reserved. Ordnance Survey 
licence number 6035/


This sounds like a Land Registry or Planning map.  They are probably 
breaching the licence by even showing it to outsiders.  The red line 
will have been traced relative to OS features.


I would say definitely off limits, as this is the sort of map from which 
OS is now funded.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] National Trust Paths organised edit page

[David Woolley]

On 02/09/2019 23:13, Warin wrote:


On 3/9/19 2:53 am, Dave F via Talk-GB wrote:

On 02/09/2019 14:58, David Woolley wrote:
This could conflict with a trend that I believe is developing, at 
least for more formal roads, of removing signage, because it 
distracts drivers, and relying on satellite navigators to provide the 
information instead.


What evidence have you of this "trend"?



I too, would like to hear of evidence of this 'trend'.



Google "reducing sign clutter" for the general principle.  Use of sat 
nav as an alternative I might have heard on the radio, or in a local 
paper.  However 
<https://www.driverknowledgetests.com/resources/what-is-signage-clutter-and-how-do-we-reduce-it/> 
is the only reference I can find to that, online, in a quick search.


I think, in practice, it why local councils often don't bother to fix 
AWOL and broken street name signs, even when told about them.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] National Trust Paths organised edit page

[David Woolley]

On 02/09/2019 14:48, Frederik Ramm wrote:

Sometimes they want us to add a "vehicle=no" to a track that has
absolutely no signposts whatsoever locally, meaning that nobody can
verify that vehicles are forbidden and no local motorist would be turned
away


This could conflict with a trend that I believe is developing, at least 
for more formal roads, of removing signage, because it distracts 
drivers, and relying on satellite navigators to provide the information 
instead.


Whilst this probably doesn't currently apply to prohibitions, a logical 
extension, at some time in the near future, might be to make the 
electronic map definitive in all cases.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Valuation Office Agency council tax data (was postcode mapping (was Re: Automated Code-Point Open postcode editing (simple cases only))

[David Woolley]

On 30/07/2019 15:12, Stephen Colebourne wrote:

What about the Valuation Office Agency council tax data?
http://cti.voa.gov.uk/cti/inits.asp

I found this recently, and it allows you to lookup from a postcode to
the individual addresses (presumably in their standard recognised
form). I've not used the data, although I'd love to because it is so
good.


Anything giving standard form is going to be a derivative of PAF and/or 
UPRN databases.


The argument for the food safety data is the addresses are as provided 
by the business.  The reason you can't use Land Registry information is 
that the addresses have been validated.


I suspect the Post Office might challenge even the food safety data if 
OSM postcode coverages started to seriously compete.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] postcode mapping (was Re: Automated Code-Point Open postcode editing (simple cases only))

[David Woolley]

On 30/07/2019 14:19, Mateusz Konieczny wrote:
Is it typical for post codes to be posted like housenumbers? Either on 
buildings or postboxes?


Postcodes almost never.  The only time you would normally find them is 
where the building is a company's registered office.


Housenumbers seem to be only on (gut feeling) about 30% of houses these 
days, and maybe 5% of commercial premises.  (When they are on houses, 
they are typically unreadable after dark.  Almost no-one has front gates 
any longer, which might have carried them in the past.)


What appears on street signs and post boxes is the outbound postcode (up 
to the end of the first number).


Businesses are required to provide an address for service in 
advertising, including web sites.  This may well be their accountant's. 
Electronically, they need to provide a geographic address, but, on 
paper, I think a PO Box number is acceptable.  Non-compliance is 
endemic, e.g. businesses only contactable over mobile phone numbers.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Mapping at Moira, NW Leicestershire.

[David Woolley]

On 27/07/2019 08:15, Graham Bowers wrote:
Is anybody local mapping here that I could tag along with to learn the 
ropes please?


Everyone has different interests.  Just because they are local doesn't 
mean that someone is going to be interested in cycle infrastructure.


My activity thus far has been editing the cycling infrastructure to 
allow routing websites to route correctly.
There is construction activity in the area that does need mapping and 
does have cycling infrastructure impact but it's more than I's wish to 
tackle without knowing quite a bit more than I do now.


The main things to remember is not to make subjective judgements are 
about suitability - record the details that an ideal router would use to 
make its own judgements - and don't use guidebooks, etc., as as source 
of information - named routes may have database rights associated with them.


Don't expect to be able to get routers to take the decisions you would 
make.  If you go too far that way, you will end up tagging for the 
renderer (routers are considered renderers), which is not allowed.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Gates open/closed by default

[David Woolley]

On 26/07/2019 12:57, Stephen Colebourne wrote:

unless there is an explicit "private" sign


There is no legal need for "private" signs.  The default assumption 
should be that everything is private (even though the OSM default is 
mainly the opposite).


In my part of the country, garden front walls are an endangered species; 
that doesn't mean that the the front yard is public parking space.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Gates open/closed by default

[David Woolley]

On 26/07/2019 10:46, Stephen Colebourne wrote:

I'd like to distinguish between two kinds of gate on private roads:

- those where the gate is closed by default (eg automatic closing)
- those where the gate is open by default (the gate exists, but is
rarely if ever closed)


I'd suggest opening_hours with a narrative time (assuming that they are 
only used at very long intervals to avoid giving a public right of way.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Ground truth v legal truth

[David Woolley]

On 19/07/2019 15:37, Mark Goodge wrote:
There are, though, two potentially useful open data coordinate mapping 
systems that can be used by the likes of OSM. One is Mapcode, the other 
is Google's Open Location Code (aka Plus Codes). Both have the advantage 
of not only being entirely free and open to use, but can also be 
generated programmatically from a published algorithm - no need to hook 
into an API, just run some code locally.


There is also the extended form of the venerable Maidenhead Locator System.

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Ground truth v legal truth

[David Woolley]

On 19/07/2019 15:06, Tom Hughes wrote:

You then followed up by saying that the logical consequence
of it being a primary (which I was assuming was correct) was
that nothing was tertiary, which didn't seem  to make much
sense to me


The logical consequence of ignoring the official classification if it is 
not signposted, is that you cannot map tertiary, because with, very rare 
exceptions, they are not signposted and you can only distinguish them 
from residential by using the official sources, or by personal judgements.




___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Ground truth v legal truth

[David Woolley]

On 19/07/2019 13:37, Tom Hughes wrote:
I would say the logical consequence of that argument is that no road 
should be mapped as tertiary, as, unless taken from OS, it is a 
subjective judgement and can't be consistently verified.


That doesn't follow - in the UK we have always (with very rare
exceptions like Oxford High Street) mapped secondary, primary and
trunk to the official status of the road.


You seem to be rejecting the original proposal.  I was analysing the 
case where the original proposal is accepted, and therefore the official 
status must be ignored if it is not signposted.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Ground truth v legal truth

[David Woolley]

On 19/07/2019 12:36, Philip Barnes wrote:

I cannot dispute this is legally a primary, OS Opendata shows it.



I would say the logical consequence of that argument is that no road 
should be mapped as tertiary, as, unless taken from OS, it is a 
subjective judgement and can't be consistently verified.


I think mapping it in conflict with a published official designation 
will devalue OSM.


As you hint, the correct thing to do, in mapping, is to map verifiable 
attributes that would make a rational router want to avoid it. Beyond 
that, like any tagging for the renderer issue, it is up to routers to 
use more than the primary classification in deciding whether to route.


This is going to get more important, with the continuing trend is to 
reduce signage, on the basis that it distracts drivers, and they have 
access to GPS maps so don't need it.


(As a variation on the last point, one of my pet hates, these days, is 
how few houses now have house numbers in the UK.  It make it difficult 
to give accurate locations for fly tips - many of the apps use sources 
like Bing, which use address interpolation, and can be a long way out in 
some cases.  I believe some US cities have bye-laws requiring them to be 
displayed.)


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Automated Code-Point Open postcode editing (simple cases only)

[David Woolley]

On 16/07/2019 22:19, ndrw6 wrote:
3. Use a collation plugin to collate both datasets with "centroid 
distance" set to "< 15m". The condition is there to apply postcodes only 
to small buildings in direct vicinity of the codepoint centroid.


This algorithm will apply PO Box number postcodes to some buildings 
adjacent to the post office.  Similarly for other high use post codes, 
which are close to residential areas.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] UK coastline data

[David Woolley]

On 13/07/2019 22:21, Colin Smale wrote:
So what was your point again about internal waterways? The "extent of 
the realm" is not the 12-mile limit, it is ±MLW, isn't it?


Assuming it is mapped correctly, this is an example of an administrative 
boundary that is outside the low water mark: 





___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] UK coastline data

[David Woolley]

On 13/07/2019 21:38, Colin Smale wrote:

Have you got a reference for this, making the link between the boundary 
of the Realm and the MCA classification of an inland waterway?
What could be a consequence of this? Could you illustrate this with an 
example?
The MCA definition of "inland waters" would draw a line across the 
Thames at Gravesend and across the Dart at Battery Point. These lines 
don't correspond to any admin boundaries I am aware of.


The link is actually with territorial boundaries (12 mile limit).

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] UK coastline data

[David Woolley]

On 13/07/2019 20:53, Colin Smale wrote:
Another reason to want MLW in OSM: The "Extent of the Realm" is *for the 
most part* defined as MLWS. This is the limit of the jurisdiction of 
normal (local) government. Beyond MLWS, the local council no longer has 
any say - it's the UK laws of the sea, as applicable to territorial waters.


Low water mark is only a boundary of the realm when it doesn't fall 
within a "internal waterway".


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] ITOworld maps

[David Woolley]

On 13/07/2019 15:33, Brian Prangle wrote:
ITOworld  maps which showed a huge variety of visualisations of OSM data 
seems to have gone offline. For about a week now I've been getting



  503 Service Unavailable

No server is available to handle this request.

Does anybody know what's happenng here? - those maps were incredibly useful




I was under the impression that this free service was closed down many 
months ago.


I guess there were management changes and it was no longer considered a 
useful loss leader.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Importing NaPTAN Data

[David Woolley]

On 01/07/2019 16:02, Silent Spike wrote:
As far as I can tell, some progress was made previously on importing 
NaPTAN data for specific areas of the UK. However, the process for 
requesting an import on the wiki seems to have broken down somewhere 
along the line and I believe the python script mentioned on the wiki is 
outdated.




In may experience, a lot of the original NaPTAN imports have decayed:

- people have double mapped stops and the the NaPTAN one has been deleted;

- stop have moved and got double mapped and deleted as a result;

- stops have been temporarily out of service, e.g whilst redeveloping a 
bus station, and been deleted and then lost the NaPTAN associations when 
remapped.


Also, all bus stops need continual maintenance because names get changed 
 as landmarks come and go.


Given that few people like maintenance work, if you can't map all the 
stops from first principles, it is very unlikely that imported ones will 
get maintained.  Retaining the NaPTAN tagging is important in allowing 
any later remerge of the updated NaPTAN data.


Another problem with NaPTAN stops, which applies to non-OSM users as 
well is that they have virtual stops in Hail and Ride areas.  Routers 
seem to only like people boarding at those place, so, in my case, can 
take me about 7 minutes out of my way against the direction of travel, 
so tell me I have missed a bus that could be easily caught.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Adjacent nature reserves

[David Woolley]

On 28/06/2019 00:56, Warin wrote:
that are also holes in them (they usually omit making the hole, so an 
added car parking area will be covered by trees until I notice


I believe that is a renderer bug.  Generally smaller, fully nested, 
areas should cut out holes in incompatible backgrounds without explicit 
relations.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Adjacent nature reserves

[David Woolley]

On 27/06/2019 10:49, Robert Whittaker (OSM lists) wrote:

The two representations are identical in terms of the data, but
the latter requires 2.5 times as many objects and is much more of a
pain to work with in the editors. All to avoid having a common line
segment between two areas



I'd certainly say there should be a strong presumption against the use 
of relations.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Playground age limits

[David Woolley]

On 04/06/2019 16:09, Martin Wynne wrote:
The main reason for the fence would seem to be the several NO DOGS 
signs, which I have tagged.


I've always assumed that such fences and the "adults must be accompanied 
by a child sign", that often accompany them, is to keep out adults who 
are not related to the children, as a child protection measure.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Farmland (crop or animals)?

[David Woolley]

On 24/05/2019 10:43, Gregory Marler wrote:
to me, meadow is a different to the common farm fields that have animals 
in. A meadow is likely longer grass, or encouraged to get long. It might 
be for flowers/wildlife rather than animals.


Meadows, in farms, in a land use context, are for producing hay. 
Deliberately encouraging wild flowers and animals, would be a park type 
usage, not a farm type usage.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Tagging a named building now used for a different purpose

[David Woolley]

On 23/05/2019 21:37, Mark Goodge wrote:
Normally, commercial 
buildings have the owner's name as the value of the 'name' key.


This is often wrong, as often the business does not occupy the whole 
building.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] Increase of mail size limit

[David Woolley]

On 14/05/2019 22:04, Rob Nickerson wrote:


Tried to send a mail to the list but it failed:

 >     Message body is too big: 533180 bytes with a limit of 40 KB

Any chance that the limit could be lifted? 40 KB seems very low given 
how cheap storage space is.




Please don't.  If people regularly start sending large attachments, I 
will need to remember to check for them and remove them from my personal 
archive, even if I defer reading the mail.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] What is a residential area?

[David Woolley]

On 08/05/2019 02:18, seirra blake wrote:
is there any articles on what does/doesn't get used on particular zoom 
levels? just figure it wouldn't hurt to double check my general 
understanding


That assumes that zoom level has any meaning to the tool being used to 
access the map.


For slippy maps, it depends on the  particular slippy map and will 
change from time to time, depending on the target audience and 
judgements about the level of clutter in the rendered image.


The only real way of finding out is to look at the code used.  As far as 
I can remember, for Mapnik, as used in the default slippy map on 
www.openstreetmap.org, it is quite easy to see the rules for when things 
get rendered.


The target audience for default slippy map is mappers, not the the 
general public.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] What is a residential area?

[David Woolley]

On 07/05/2019 15:22, Dave F via Talk-GB wrote:


If a pub has external land other than the building itself a similar 
schema to 'schools' should be used: draw an enclosing polygon around the 
extent of the grounds & tag it with amenity=pub & any other details such 
as name, address, website etc. All other areas parking, garden etc 
should be mapped with the boundary The actual buildings should only be 
tagged with building=yes/pub.


Agreed.  I was commenting that it is common to see the pub building 
tagged as the pub, when the whole grounds should be.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] What is a residential area?

[David Woolley]

On 07/05/2019 15:03, Martin Wynne wrote:



But is that what OSM is for -- to describe the *purpose* of a thing?


The original purpose of OSM was to break the monopoly on map data held 
by commercial mappers, by taking advantage of the ready availability of 
GPS equipment, so the purposes for which the data would be used would be 
all applications for which OS data was used, with a bias towards 
applications that needed a lot of data.


I thought the idea was describe the *physical* object and its location? 
Physically it is a house built in a garden.


Describing the physical object is a way of objectively mapping, but if 
that is all you do, you don't need a map; just use the aerial imagery 
directly.


Saying that something consists of a building within a garden is actually 
going beyond describing what can be seen, as it basically says that 
building is subordinate to the garden.  Whilst I wouldn't want you do do 
this, if you really want to map without making any inferences, map the 
garden as natural= and map it in just those areas that are not house, 
drive, hard standing, etc.


For all we know, it may not be anyone's residence -- it could be being 
used as offices, say. In that case landuse=residential would be wrong, 
it should be landuse=commercial. But it's still a house in a garden.


It's a building, partly surrounded by grass.

landuse and house both imply a level of abstraction based on judgements 
of the intended use that go beyond a literal description of what you can 
see.


Although I picked on the problem with having both fence and garden main 
tags in the housing estate link, that does correctly show how you would 
use garden, when micro-mapping.  (I didn't pick up the fact that the way 
the fences are mapped is making assertions about the legal ownership of 
the fences which are probably wrong.)



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] What is a residential area?

[David Woolley]

On 07/05/2019 13:30, Martin Wynne wrote:
This idea of primary and secondary tags is new to me. There is no such 
distinction in the iD editor -- all applied tags are simply listed in 
alphabetical order.


Things like name, height, and colour are normally considered secondary.

Things like landuse, building and leisure are considered primary.

Generally if you don't have a primary tag, the object won't get 
rendered.  If you have more than one primary tag, it becomes difficult 
for renderer to choose how to interpret the object.  However, if there 
is no great conflict, it is sometimes acceptable to combine, primary tags.


Often combining primary tags indicates that you are oversimplifying, 
e.g. it is common to have both shop and building, but that is probably 
wrong, because the shop often doesn't occupy the whole building. 
Similarly a pub may really be the gardens, as well as the building.


leisure and landuse conflict because they are both really specifying 
land use.


I think your biggest problem though is expecting that there is one 
correct way of doing things.  In a project with distributed decisions 
making, you are just not going to have everyone agree on what OSM is for 
and how things should be mapped.


However, I wouldn't say the primary purpose of the area you were asking 
about is to be a garden; I would say it is somewhere to reside, and the 
gardens form a subsidiary part of it, and should be represented with 
nested areas.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] What is a residential area?

[David Woolley]

On 07/05/2019 12:40, Martin Wynne wrote:


I think this goes to the heart of my (mis)understanding of what OSM is for?

Are we trying to create a legal reference document?

Or a description of what a visitor would see on the ground?



We are trying to create something that serves several purposes.

We are not actually trying to create a legal document, but we are trying 
create something that is useful to people making planning decisions and 
trying to understand the nature of the area.


The reason we cannot produce a legal document is that we cannot fulfil 
the requirements of producing something that is definitively correct, 
but for many purposes things only have to generally correct, not legally 
correct.


In this case, I wouldn't have cut out the farmland at all, around the 
road and the house, especially if the house turns out to be a farm 
house.  I might have nested landuse=residential, for the curtilage of 
the house.  It would be up to a data consumer to decide whether such an 
isolated landuse=residential was relevant to its needs to know about 
landuse.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] What is a residential area?

[David Woolley]

On 07/05/2019 12:17, Martin Wynne wrote:

On 07/05/2019 11:34, Dave F via Talk-GB wrote:
Your OSM example look fine to me - a single property is still where 
people reside. Any other details, such as garden, should be mapped 
individually within that area.


Thanks Dave. But in that case, why in the iD editor when I change 
"Residential Area" to "Garden" does it remove the landuse=residential tag?


To stop you having conflicting main tags.


Should I be creating a duplicate way as a "Garden" on the same nodes?


If you are going to micromap the garden, it should exclude the house and 
the car parking area, so it won't be using the same nodes.




Or should I leave it as "Residential Area" and add a leisure=garden tag? 
When I do that, the OSM standard map doesn't render the garden.


Again you are giving something two different main tags and that will 
confuse most renderers.  On the other hand, you shouldn't be basing your 
tagging on what gets rendered.  (Although I wouldn't advise mapping the 
average city front yard as parking space, or the average but to let 
front yard a recycling area, it would not be be technically wrong to do 
so, even though it would produce an ugly rendering.


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb