[talk-au] What gives with roundabouts?

[bluemm1975-osm at yahoo.com]

+1

I completely agree with all of Darrin's points.

I was unaware of the decision on the mailing list when I started mapping a about 1.5 years ago. I read the descriptions on the wiki and went with a proper roundabout for suburban roundabouts, since they don't fit the definition of the mini-roundabout. I don't even recall seeing one in Melbourne, you always have to deviate around even small ones. As a Potlatch user, it sucks a bit to add them, but Merkaartor & I think JOSM have a tool to make it easy.

I'm a big fan of "mapping what's on the ground" and "don't tag for the renderers/routers". But I like the idea of global consistency, it makes it easier on all users of the raw data. That's what I hope Map Features will provide (consistency), but the voting has it's issues as well. There's talk on the Talk mailing list of having a Core Features page. So for eg. I'd be in favour of using the wiki definitions of place=* tags.

I plan on submitting a proposal for the roundabout tag, where you can add it to a node like a mini_roundabout, for use in simple suburban type roundabouts. Something like junction:inner_width=3mcould specify the island size, making it possible for pretty rendering. Weird intersecting ways or large roundabouts would have to continue as is.
Anyone have any suggestions before I create the proposal?

PS. Was it me adding turning_circle to courts? The wiki page description seems to match my use of it (I waited many months for it to be proposed/accepted/added to renderers).

Cheers, BlueMM

--- On Thu, 11/12/08, Darrin Smith <beldin at beldin.org> wrote:
> From: Darrin Smith <beldin at beldin.org>
> Subject: Re: [talk-au] What gives with roundabouts?
> To: Talk-au at openstreetmap.org
> Received: Thursday, 11 December, 2008, 3:10 PM
> On Thu, 11 Dec 2008 12:55:13 +1100
> Ian Sergeant <isergean at hih.com.au> wrote:
> 
> > > I've looked back through the logs, found the one discussion, noted
> > > that it was basically a 4-3 split of contributors and since every
> > > discussion on it has been "we discussed it and decided this".
> > > Hardly a consensus in my mind.
> > 
> > Since you made the effort to go back through the logs, and re-read the
> > discussion that took place then, I'm surprised you would reach the
> > conclusion that peoples position was solely related to effort.
> 
> I suspect that's because you were in the discussion and supported the
> views of the first poster, hence you didn't look at how he expressed
> it. (see below)
> 
> > I disagree that there hasn't been consensus on its use.  There always
> > have been differences of opioion, but as you say, most people have
> > been happy to accept that it is the way it is.  That is consensus.
> 
> Ok I'll pay that in the technical definition of the word you are
> correct. However given that all new people approach a project
> like this with some trepidation (For example it's taken me 10 months
> and someone altering work I've done to make me raise this issue which
> I've thought was wrong almost as soon as I found out about it)
> it's not surprising the 'consensus' has been maintained.
> 
> OSM is littered with cases of things done badly to start with (which is
> not a problem in one sense because something needs to be started
> somewhere) and then carried on forever after (this is where it's a
> problem) in what appears to be consensus because no-ones been motivated
> to change it (The hideous is_in tag comes to mind). 
> 
> > I feel the approach you are taking is wrong.  There are reasonable
> > arguments to use a mini-roundabout tag in Australia where it is
> > currently being used.  If you want to convince people to not use it,
> > and to map using junction, take the time to understand and address
> > those arguments, and convince people that the best way is the way you
> > are suggesting.  Don't dismiss its proponents as lazy, or worse still
> > as disruptive.  Many of its these people have been valuable
> > contributors to getting the map done.
> 
> Ok, I could have approached it a better way I'll admit that. But this
> issue boils down to the fact there are no actual reasons given for why
> mini_roundabout should be used! The discussion just seems to assume that
> every roundabout is a mini until it has reason to be bigger, it's not
> even discussed whether this is valid. 
> 
> The discussion resolves around what benefits the roundabout tag offers
> OVER the mini_roundabout tag, ignoring the fact they actually imply 2
> quite different things in the first place.
> 
> Historically the roundabout tag predates the mini-roundabout tag by at
> least 10 months in the wiki pages. So in effect the original
> mini_roundabout tag was devise to handle a very special case of
> roundabout that doesn't fit well with the normal definition in size,
> shape, signage and the fact you can drive straight over it in
> extenuating circumstances (I can't help but wonder if mini is
> referring more to it's HEIGHT that it's radius).
> 
> But here in Australia we apparently want to turn that very specific case
> tag into the general tag and make the general tag for specific cases
> without actually providing any valid reason (since we are apparently not
> lazy). 
> 
> The general roundabout tag is still more accurate on the ground - You
> CAN'T drive straight through the center of the the average small
> suburban roundabout, you often can't even walk over it for the
> various things stuck on it. There is and island there, so why shouldn't
> there be an island on the map? 
> 
> So we in Australia have effectively reversed the precedence order of
> these roundabouts.
> 
> -- 
> 
> =b


      Start your day with Yahoo!7 and win a Sony Bravia TV. Enter now http://au.docs.yahoo.com/homepageset/?p1=other&p2=au&p3=tagline