[OSM-talk] A warning about gates and other barriers

[sergio sevillano]

El 20/09/2010, a las 12:33, Elizabeth Dodd escribió:

> On Mon, 20 Sep 2010 11:21:21 +0100
> "Dave F." <davefox en madasafish.com> wrote:
> 
> The point is that tags should only accommodate one piece of information.
> A gate is a gate, access is something else. Defaults cause confusion
> and arguments - we should have a system in which information is
> specified and not assumed.


hi

im the one who wrote that barrier implied access=no back in the day.
but when the barrier tag was created i thought that that implication was forcing to always tag access along barrier.
(that was a common effort and no one else until now thought that this was wrong neither me)
also i never thought about routing softwares

now, after this years of mapping, i think that defaults are a wrong way to tag
if you have access tag, use it either way.

barrier=gate alone, inmo only means there is a gate, nothing else.
if there is no access tags there is no access info, period.

a mapper can see a gate but has no time to stop 
and check if the gate is closed, locked or has opening_hours, 
so he should map it so the info doesn't gets lost 
and let other with more time tag access.

same with satellite mapping, if i map a highway and i don't know what type is it, 
i must use highway=road and of course i don't know if there is access limits. 
until someone with more info than me tags it.
 
the assumptions that routing software must take 
should be done by common sense of coders and 
explained in their own pages
so the routing software user knows what to expect.

> It´s cold here and I´m going to sit back and wait for the flames to
> warm me up.

me too

cheers 
sergio