Moved from talk to dev.<br><br><div><span class="gmail_quote">On 2/11/07, <b class="gmail_sendername">Jochen Topf</b> <<a href="mailto:jochen@remote.org">jochen@remote.org</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
On Sun, Feb 11, 2007 at 12:09:42PM +0000, 80n wrote:<br>> On 2/11/07, Tom Chance <<a href="mailto:tom@acrewoods.net">tom@acrewoods.net</a>> wrote:<br>> ><br>> >Ahoy,<br>> ><br>> >On Sunday 11 February 2007 00:53:25 Ben Robbins wrote:
<br>> >> 5 problems I have...please split up into seperate topics for<br>> >reply. Check<br>> >> image for refernces..<br>> >><br>> >> <a href="http://i57.photobucket.com/albums/g226/ben_robbins_/osm3things.png">
http://i57.photobucket.com/albums/g226/ben_robbins_/osm3things.png</a><br>> >><br>> >> 1) Where different highway types meet at junctions it usually renders<br>> >> quite a messy overlap of the core where 1 of the cores renders later and
<br>> >> has a rounded end. In just adding an additional tag (for the purpose of<br>> >> the example I have added core=yes) you can make selected cores render<br>> >again<br>> >> later and clean up junctions. What would a good tag for this be?
<br>> >or...are<br>> >> there better methods of sorting out this asthetical fault?.<br>> ><br>> >If things look wrong in the slippy map or your own Osmarender renders,<br>> >don't<br>> >tag your way around the problems! Report the problems with mapnik and
<br>> >osmarender and wait for them to be fixed, but keep the tags in the<br>> >database<br>> >clean.<br>><br>><br>> Actually this kind of experimentation and innovation should be encouraged.<br>
> If people play with rules files and demonstrate potential solutions, that's<br>> a much more constructive approach than submitting a fault to trac and then<br>> wondering when it will get fixed.<br>><br>
> I think the core=yes modification is quite interesting. The end result<br>> looks good and is an improvement - but maybe there is a better way to<br>> implement it. Now that Ben has demonstrated how good it looks, I think we
<br>> are at least one step closer to a solution.<br><br>Generally there is no need for the core=yes element. As far as I can see<br>the problem only occurs in T-shaped junctions (or topological similar)<br>where the through road has a "lower classification" than the ending road.
<br>So if there is a primary road ending at a T-junction and the<br>unclassified road goes through the same junction, you'll see the ugly<br>round knob in the junction. But it certainly is computationally possible<br>
to find those T-junction cases and re-order the drawing of the ways.<br><br>Problem is: This gets messy once we are not only looking at one junction<br>but at a whole road network, and it gets really messy once we take layers
<br>into account. But I don't see how the core=yes helps here. It only makes<br>things more complicated. Certainly for the person doing the tagging.</blockquote><div><br>Ben's solution is to tag a short section of the unclassified road as core=yes. This causes it to be re-rendered after the core of other more major roads have been rendered.
<br><br>The approach is not dissimilar to the way that layers work, except its only within the road rendering section.<br><br>It's elegant because it keeps the rendering engine simple, the results are better and the presence of an extra tag is non-invasive. Other renderers can ignore it and osmarender can ignore it when a better solution is found.
<br><br>I think I'd prefer the tag to be renderCoreLast=yes as this makes it a little clearer than just core=yes.<br><br>It's unwise to be critical of people who are doing innovative things with osmarender rules files unless we actually have something better to offer. I know Ben has a very fine eye for detail and cares a lot about the appearance of his maps. He has done something interesting here to solve a small problem that we have not yet paid attention to in osmarender.
<br><br></div><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">And another reason why the core-thing is an ugly hack that doesn't even
<br>work: It only works as long as the highway casing is always the same<br>size. See the enclosed png for an illustration of the problem.<br><br>The core of the problems seems to be that we assume that road junctions<br>
are point objects and have no size. But thats not the case once you draw<br>roads which have a width.<br><br>btw: The google map rendering has the same problem. Sometimes it is even<br>worse, because sometimes they not only have the ugly round core thing,
<br>but the casing is drawn on top also. See for instance just south of the<br>Golden Gate bridge the short stretch of highway 1. (Problem appears only<br>in some zoom levels.)</blockquote><div><br></div><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Jochen<br>--<br>Jochen Topf <a href="mailto:jochen@remote.org">jochen@remote.org</a> <a href="http://www.remote.org/jochen/">http://www.remote.org/jochen/</a> +49-721-388298<br><br></blockquote></div><br>