[Talk-GB] Gates

[Ian Spencer]

Gregory wrote on 28/05/2010 14:19:
> You are just thinking about cars, what about foot 
> routing/planning(sorry I meant to make that clear in my last e-mail)
>
> If we are creating a specification for open or unlocked, we might as 
> well include locked. Especially on foot it can be helpful to know the 
> gate you see up ahead is usually locked, but if you go another way 
> there is an unlocked gate. Don't restrict the uses of OSM by 
> restricting the data.
>
Stepping back - why is it useful to map gates?

* They can be landmarks.
* They represent a potential to restrict access.
* They potentially segregate foot traffic, who can access at some times, 
from vehicle traffic, who may not enter.

Is it useful to mark the status of a gate?
* If it is a gate that can be locked and there is no pattern, then 
probably not - access permissive might cover it.
* If the locking has a pattern, e.g. an alleyway that is locked at 
night, or access to an business park that is locked out of hours, 
(especially that might otherwise be a useful through way or access to 
other facilities, like a nature reserve), then that is useful.
* If the gate is clearly permanently locked to a private access, e.g. a 
farm gate, then this is barely useful information.
* If there is something like a public footpath that runs along a track, 
then there may be a permanently closed gate to bar vehicle traffic, yet 
there may be permanent foot or bike access. We see that often on 
abandoned railways converted into cycleways.
* The fact that there is a locked or unlocked gate where there is no 
public access should not make any difference to accessibility, so I 
think would generally infer access from the status of the adjoining ways 
rather than mark it, unless it is exceptional, e.g. a public road 
arriving at a junction with a locked gate. I can think of an example 
where there is a bus access to a business park, with footpath and 
cycleway alongside, but it is not available for a car. The fact it is 
open, closed, locked or unlocked makes no difference as to its 
accessibility to cars, yet cars can access either side of the gate, one 
as a road passing by, and one as a turning point.

Aside from that, I don't think the map should attempt to document 
something transient.

So, I'd say that if the gate makes a useful landmark, as in a country 
lane, or on a footpath, then it is worth mapping. If it is just an 
entrance gate that is there for security reasons, then I don't think 
people should get overly excited about mapping them. If it is a known 
obstruction to an otherwise useful way, then documenting the nature of 
the obstruction would seem sensible. How this in documented in part 
needs agreement from the routing team, because whatever is mapped needs 
to be understood by routing programs. I'm not sure how sophisticated 
they are at the moment for considering things like time restrictions.

So perhaps that is the test: think about the impact to a routing 
program, be it a program or someone examining the map. Would the 
information help decide how to get somewhere? If yes, then it is clearly 
useful, if it would not, then there is the answer.

Spenny