[Tagging] Two side-of-road parking questions

Matthew Woehlke mwoehlke.floss at gmail.com
Fri Jul 24 14:00:49 UTC 2020

On 24/07/2020 08.19, Paul Allen wrote:
> On Fri, 24 Jul 2020 at 00:06, Matthew Woehlke wrote:
>> Currently, I have the non-parallel spots marked as a lot. To my mind,
>> parallel parking and on-street parking are nearly synonymous.
> I'm not entirely clear what you mean by those terms as you're
> American.

That's okay; that makes two of us ;-).

> The image in the wiki for parking lanes matches
> what I expect of it.  As in this situation near me:
> https://goo.gl/maps/WUZKmhQTDSRsgnDx7 on the right
> of the road are double yellow lines, which mean "no parking
> or waiting at any time" (but there are exceptions) and on
> the left is a single yellow line which means "parking and
> waiting permitted some of the time" (though there are
> exceptions and provisions and it gets complicated).  The
> left is a parking lane, as I understand it.  There are no
> parking spaces marked.

AFAIK there is nothing exactly like that in the US. People do park on 
streets (note 5th, 4th and 3rd Avenues, as previously mentioned), and 
there is sometimes signage regulating this.

Actually, this might answer one of my prior questions; is =marked 
supposed to be used for stretches that alternate between parking allowed 
and parking forbidden?

BTW, this is what NYC apparently considers a "parking lane":

Based on this discussion, it seems to me that that is *not* a "parking 
lane" as OSM uses it.

That said, I think the US definition may be "a lane which is for 
parking, rather than through traffic, and which may be intermittently 
present" (e.g. the above video). However, I am much *less* convinced 
that it is useful to model them that way, at least in the current state 
of things.

>> Well, there's an easy solution to that; map the spaces, also ;-)
> Yeah, but the spaces don't render.  Oh wow!  I just checked one of your
> later examples and parking spaces now render.  I'd given up on hoping that
> they would render.  Doesn't fix the example I'm thinking of, though - it's
> clearly a pregnant bulge that is for parking, but no spaces are marked.

Ah, yes, that would be an issue. Not sure what was with the not 
rendering, unless you happened to look at it soon enough after I 
uploaded the changes. There does seem to be a variable delay between the 
database changing and the rendered tiles being regenerated.

>> it's normal for a parking lot area to include at least parts of the
>> aisles.
> In one sense it's correct.  At a level of highway modelling we don't do and
> may never do.  In terms of what gets rendered (where the renderer draws
> roads on a layer above parking lots), it's perfectly comprehensible.  Since
> we don't have a better way of representing what's there, I ignored the
> objection.

Agreed; if the road was mapped also as an area, it would make sense.

>> ...and to be honest, another argument for modeling as lots is that the
>> parking_lane tagging is rather more obtuse...
> There is that.  Which is why I tend not to bother with it.  Especially as it
> means surveying and finding out the restrictions on times.

Meh, I'm happy to omit the restrictions :-).


More information about the Tagging mailing list