<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000099">
I think that Richard may have a sensible compromise here that could
work for rights of way. Remembering that:<br>
<br>
1. The ID / reference for a particular public right of way (PROW)
usually applies to a single way - although occasionally to 2 or 3.<br>
2. Any particular way almost always belongs only to a single PROW ID /
reference (there are rare exceptions).<br>
<br>
the bridge / tunnel analogy works for me and avoids the destruction of
a lot of hard work while allowing the sort of revision that is being
discussed to move most ref tags to a route that exists at a higher
logical level. I already do this in fact for cases where there is a
'named' or 'reference numbered' long-distance path or cycle route -
usually involving many dozens (or hundreds) of ways. These are clearly
'routes' as normally understood and I fully agree that in these cases
it is the route that should carry the ref and not the numerous
individual underlying ways. In fact it is doing this that allows me to
use the ref tag for the PROW ID (sometimes signed and, even if not,
available in the public domain).<br>
<br>
The mention of bridges does remind me of another reason for supporting
the basic proposal - with the important proviso that bridges (and
PROWs? <span class="moz-smiley-s3"><span> ;-) </span></span>) are
treated as having "physical" (and often signed) IDs. In England it is
normal for bridges over canals (we have an extensive network of inland
waterways) to carry ID numbers unique for that particular canal - but
these bridges may also carry a highway that has a number reference. If
the highway has a national reference number e.g. "A51" it would be
convenient to assign this to the 'route' of the A51 and leave a ref tag
on the bridge for the number of that bridge. This does establish,
however, a hierarchy: what if the bridge carries a PROW? You may want
to tell me that the PROW should also be a route! (and we're back to my
original argument and concern).<br>
<br>
This leads me to two interim conclusions - one firm and one not:<br>
<br>
1. If the proposed change in use of the ref= tag is to be implemented
it MUST involve human intervention. It is too complex an issue to leave
to a bot (even a good one, if such exist!) - if I am raising one 'grey'
area it suggests that there may be many others out there. It is more
difficult to make these big changes now that OSM is maturing (well
adolescent anyway <span class="moz-smiley-s7"><span> :-\ </span></span>).<br>
<br>
2. A less drastic (?) change might be to subdivide the ref= tag (as I
have done on occasion when faced by complexity) e.g. ref=A51 on the
route of the vehicular highway (many ways); bridge_ref= (better?
ref_bridge= ) on the bridge (one or very few short ways; if very
complex, then likely already to be a relation that can be tagged
instead); PROW_ref= on the public right of way (one or very few fairly
short ways); ref=N99 on the cycleway route that uses the same bridge,
etc.<br>
<br>
The 'grey' area is well-exemplified by rights of way and perhaps by
complex bridges that have not been coded as a relation - there will
certainly be other examples. It might be a good idea to remove some /
most of the 'grey' cases by implementing a ref_whatever= protocol (much
easier to do for a specific set of cases - even maybe with a bot???).
After this clean-up step it would be safer to proceed with the main
proposal - so long as there is sufficient human intervention to avoid
unintended destruction / loss of data from the OSM database. It's worth
taking a little longer to get it right.<br>
<br>
A final thought: when I was a real newbie (how long does it take not to
be one still - a decade?) I thought relations looked difficult. They
are not - but perhaps the wiki could be more encouraging and get people
using them more e.g. for routes, which might reduce future problems of
a similar kind to this one).<br>
<br>
Cheers!<br>
<br>
On 09/03/2010 03:54, <a class="moz-txt-link-abbreviated" href="mailto:newbies-request@openstreetmap.org">newbies-request@openstreetmap.org</a> wrote:
<blockquote
cite="mid:mailman.1477.1268106859.20993.newbies@openstreetmap.org"
type="cite">
<pre wrap="">Send newbies mailing list submissions to
<a class="moz-txt-link-abbreviated" href="mailto:newbies@openstreetmap.org">newbies@openstreetmap.org</a>
To subscribe or unsubscribe via the World Wide Web, visit
<a class="moz-txt-link-freetext" href="http://lists.openstreetmap.org/listinfo/newbies">http://lists.openstreetmap.org/listinfo/newbies</a>
or, via email, send a message with subject or body 'help' to
<a class="moz-txt-link-abbreviated" href="mailto:newbies-request@openstreetmap.org">newbies-request@openstreetmap.org</a>
You can reach the person managing the list at
<a class="moz-txt-link-abbreviated" href="mailto:newbies-owner@openstreetmap.org">newbies-owner@openstreetmap.org</a>
When replying, please edit your Subject line so it is more specific
than "Re: Contents of newbies digest..."
</pre>
<pre wrap="">
<fieldset class="mimeAttachmentHeader"></fieldset>
Today's Topics:
1. Re: Time to retire ref= on ways? (Paul Johnson)
2. Fwd: Re: Re: Time to retire ref= on ways? (Mike Harris)
3. Re: Fwd: Re: Re: Time to retire ref= on ways? (Zeke Farwell)
4. Re: [Talk-us] Fwd: Re: Re: Time to retire ref= on ways?
(Richard Welty)
</pre>
<pre wrap="">
<fieldset class="mimeAttachmentHeader"></fieldset>
_______________________________________________
newbies mailing list
<a class="moz-txt-link-abbreviated" href="mailto:newbies@openstreetmap.org">newbies@openstreetmap.org</a>
<a class="moz-txt-link-freetext" href="http://lists.openstreetmap.org/listinfo/newbies">http://lists.openstreetmap.org/listinfo/newbies</a>
</pre>
</blockquote>
<br>
<div class="moz-signature">-- <br>
<b><i>Mike Harris</i></b></div>
</body>
</html>