[Talk-GB] Administrative boundaries (from Re: traffic island mapping / harmful detail?)

[Colin Smale]

>     On 04/09/2021 11:43 AM Edward Bainton <bainton.ete at gmail.com> wrote:
> 
> 
>     Colin, where you find boundaries have been pushed about, do you have any sense of which editors are responsible?
> 

I have to say that the majority of cases are done using iD. That's not to say that iD is a bad editor, it is just the most popular.

I am starting to suspect that there must be something in iD that makes it easy to make certain mistakes. Is it possible that boundaries look similar to paths? I see a lot of new footpaths being drawn in that end on a node shared with boundary instead of linking to the highway a couple of metres further. I see boundaries being split unneccesarily where they are crossed by a footpath (the footpath may need splitting at the boundary due to different PRoW ref's, but there is no need to split the boundary). My hypothesis is that somehow people are confusing boundaries with footpaths.

>     I ask because I know I've been guilty of it in the past. I use iD exclusively, which is presumably the editor of choice for new mappers.
> 
>     So on this topic of layering, I wonder if a quick fix would simply be to have iD default to *not* showing boundaries (ie, its check-box isn't ticked), so you have to opt in to seeing/editing them. (Possibly the checkbox could also be accompanied by a polite warning about the ease with which they're changed, and the difficulty of putting them back where they were.) 
> 
As most OSM boundaries in the UK originate from OS Boundary-Line the geometries can be considered a useful reference frame for the alignment of objects positioned by other means (tracing from imagery, consumer GPS traces). Instead of hiding them, how about displaying them but read-only? If a user attempts to edit a boundary, either refuse, or at least prompt "are you sure?". Perhaps advanced users might be able to disable this check and warning. On the other hand advanced users will be more likely to use JOSM or Potlatch, so why not just make boundaries read-only in iD. How should it behave where objects have been linked (e.g. boundary shares a way with a waterway or highway) and the user has reason to move the highway/waterway but should leave the boundary in place?

>     It was only after I was proficient enough not to move boundaries (I hope) that I was proficient enough to use those selectors.
> 
> 
>     On Fri, 9 Apr 2021, 09:46 Tom Crocker, <tomcrockermail at gmail.com mailto:tomcrockermail at gmail.com > wrote:
> 
>         > > 
> >         On Fri, 9 Apr 2021, 09:19 Colin Smale, <colin.smale at xs4all.nl mailto:colin.smale at xs4all.nl > wrote:
> > 
> >             > > >             I would just like to describe the normal paradigm when dealing with layers, as used by many renowned programs (e.g. Photoshop, QGIS). I am not suggesting that all these points are essential or even desirable for a layer-aware OSM editor; it is intended solely as "food for thought."<snip>
> > >             Have I missed anything?
> > > 
> > >         > > 
> >         Layers can be grouped so their visibility can be altered together (and some other features like blending I think, not sure what of relevance to OSM). I'd like to see this in JOSM (don't think it's implemented?) regardless of data layers.
> > 
> >             > > > 
> > > 
> > >         > >         _______________________________________________
> >         Talk-GB mailing list
> >         Talk-GB at openstreetmap.org mailto: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/20210409/2e68c740/attachment-0001.htm>