[Tagging] [Talk-us] stop signs

[Anthony]

On Wed, Oct 27, 2010 at 9:14 AM,  <john at jfeldredge.com> wrote:
> I have never seen a stop sign at a railroad crossing.

I have.  They used to be quite prevalent.  Now they're more rare, as
most places have installed active warning devices and gotten rid of
the stop signs.

Unfortunately, I don't remember exactly where I was last time I saw
one, or I'd check it out on Google Street View.  On the other hand, it
only took a couple minutes of searching to find this news story:
http://www.abbevillenow.com/view/full_story/9581389/article-New-stop-signs--markings-installed-at-railroad-crossing-on-Hospital-Drive?instance=home_news_lead

> Buses are required by law to stop before a railroad crossing, and open the bus door so that the driver can better
> hear if a train is approaching.  Some other commercial vehicles routinely stop as well, but private vehicles aren't
> required to stop.

Illinois Rules of the Road: "When approaching a railroad crossing you
must stop within 50 feet to 15 feet if there is a posted STOP sign,
the electric signal is flashing or the crossing gate is lowered."
(http://www.cyberdriveillinois.com/publications/rules_of_the_road/rr_chap10.html)

> If there is a jurisdiction that places stop signs at each railroad crossing, I would be interested in learning where it
> is.

I don't know of any jurisdictions that place them at *each* railroad crossing.

Anyway, I thought we were saying that railroad crossings *were*
defined as intersections.  They certainly would fit that definition of
"a node shared by three or more way segments".

On Wed, Oct 27, 2010 at 9:26 AM, M∡rtin Koppenhoefer
<dieterdreist at gmail.com> wrote:
> really I don't see the point of this discussion anymore: I already
> question the benefit of tagged stop signs in general, as a stop sign
> itself requires very few seconds of travel time, while a unregulated
> crossing with a lot of traffic from the right might require a lot
> more, it all depends merely on traffic density (which itself is quite
> dependent on the time).

Routers are not the only users of OSM data.

> But why should we conduct research on
> "jurisdiction that places stop signs at each railroad crossing" or
> stuff like this?

No one's forcing you to do any research on anything.

> Is our way of mapping stop signs (or better the "requirement to stop") depending on this?

One proposal for mapping stop signs is that the stop sign always faces
opposite the nearest intersection.  In order for a computer to
determine the nearest intersection, it has to have a precise
definition of intersection.  Whether or not railroad crossings count
as intersections would be a key part of that definition.