<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    Am 28.04.2012 11:29, schrieb Thomas Davie:
    <blockquote
      cite="mid:F21B132C-F670-4993-897D-C6D486E9021E@gmail.com"
      type="cite">I like the idea, though I'm not 100% certain how the
      tool would call out to the various renderers – there is at least
      one (mine) which reqires a runtime only available on one OS.</blockquote>
    I think about the tool being written in Java (as that should be
    availlable on most OSs, I think)<br>
    The Interface to the renderers could be different<br>
    1) "server-client style": define some kind of a http request
    defining data and stylesheet, and a url and port that the renderer
    can support (would even allow to use renderers running on different
    OSes, like in a virtual machine)<br>
    2) "exec" style: define a shell command to run (slightly different
    depending on OS, but there should be libraries to do that more or
    less OS independent from Java<br>
    3) "include" style: for java implementations even running the code
    included as a jar file could work, if a common
    mapcss-renderer-interface would be implemented.<br>
    <br>
    Which renderers to compare and test could be decided by the user
    (e.g. with some presets), depending on the Operating System and
    defined using a config file, that can be exchanged arbitrarily.<br>
    <blockquote
      cite="mid:F21B132C-F670-4993-897D-C6D486E9021E@gmail.com"
      type="cite">
      <div><br>
      </div>
      <div>I'd also toyed with the idea of having reference renderings,
        like the CSS implementers do, but this becomes problematic
        without specifying things like label positioning algorithms, and
        conflicting label choice algorithms.</div>
    </blockquote>
    Sure: There may be differences, but comparing renderers does not
    (only) mean validating them. Validation is not easily possible on
    unspecified issues, and definitively not automatically possible, but
    thats similar to the needs designers have for websites: You can
    follow the HTML and CSS standards, but that doesn't make sure that
    the website looks good or even acceptable in all browsers.<br>
    <br>
    That's what map designers should get at hand, I think, so that's how
    it's a useful tool for mapcss design.<br>
    For you as mapcss renderer developers it's a useful tool as you can
    compare between renderers and decide on your own, which differences
    are acceptable, and which look like a conflict according to the
    specs and common sense.<br>
    <br>
    This may include judgements like: "these highways look different,
    but only, as A did not support multi-color-dotted casings, yet" or
    something like that, and that's fine, too.<br>
    <br>
    regards<br>
    Peter<br>
    <br>
    <blockquote
      cite="mid:F21B132C-F670-4993-897D-C6D486E9021E@gmail.com"
      type="cite">
      <div><br>
      </div>
      <div>Bob<br>
        <div>
          <span class="Apple-style-span" style="border-collapse:
            separate; color: rgb(0, 0, 0); font-family: Helvetica;
            font-style: normal; font-variant: normal; font-weight:
            normal; letter-spacing: normal; line-height: normal;
            orphans: 2; text-align: -webkit-auto; text-indent: 0px;
            text-transform: none; white-space: normal; widows: 2;
            word-spacing: 0px; -webkit-border-horizontal-spacing: 0px;
            -webkit-border-vertical-spacing: 0px;
            -webkit-text-decorations-in-effect: none;
            -webkit-text-size-adjust: auto; -webkit-text-stroke-width:
            0px; font-size: medium; ">
            <div><span class="Apple-style-span" style="font-family:
                Arial; ">
                <pre>if (*ra4 != 0xffc78948) { return false; }</pre>
              </span></div>
          </span>
        </div>
        <br>
        <div>
          <div>On 28 Apr 2012, at 10:25, Peter Wendorff wrote:</div>
          <br class="Apple-interchange-newline">
          <blockquote type="cite">
            <div>Hi.<br>
              I'm not inside MapCSS-Developing currently, but what do
              you all think about the idea of implementing a "mapcss
              test center".<br>
              It's a idea that came to my mind while reading the last
              messages on the list, so probably it's a bad one, but who
              knows...<br>
              <br>
              I guess, all mapcss implementations should be able to take<br>
              1) a mapcss file or string<br>
              2) map data (not sure, if here osm.xml is or easily could
              be a common base format)<br>
              and produce an image out of that.<br>
              <br>
              What about implementing this into e.g. a java toolkit
              (java, because it's running on any major OS usually), that
              enables the user to<br>
              - write a mapcss-style<br>
              - get rendering results of all registered renderers<br>
              <br>
              This could be used<br>
              1) to check rendering results<br>
              2) to compare renderings and identify compatibility
              problems (between implementations and compared to the
              "standard")<br>
              3) as a design tool for style designers, including mapcss
              validation, probably giving hints or helping with
              "shortcomings" like the idea proposed by Thomas about
              nesting rules - as the tool could give an option for "make
              more specialized rule".<br>
              <br>
              Just an idea - even if I don't have time to work on it.<br>
              <br>
              regards<br>
              Peter<br>
              <br>
              Am 28.04.2012 10:59, schrieb Komяpa:<br>
              <blockquote type="cite">Hello everyone,<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">We all know everyone is willing to
                make their renderer the best one.<br>
              </blockquote>
              <blockquote type="cite">Everyone has own purpuose - some
                are targeting on speed, some are<br>
              </blockquote>
              <blockquote type="cite">willing to provide good editing
                experience, some want to just show<br>
              </blockquote>
              <blockquote type="cite">something to the user...<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">So, what do I propose:<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">1. Count and list active renderer
                developers. Those who really have<br>
              </blockquote>
              <blockquote type="cite">time to fix minor things and want
                to have a badge "Supports MapCSS 0.2<br>
              </blockquote>
              <blockquote type="cite">final".<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">2. Set up a deadline. Something
                real, like 1st of June (or July),<br>
              </blockquote>
              <blockquote type="cite">2012, when a spec will be marked
                "final", and all things post this<br>
              </blockquote>
              <blockquote type="cite">date will go to further
                versions/extensions/discussions.<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">3. Chop the current draft to
                really really basic things. As we see,<br>
              </blockquote>
              <blockquote type="cite">almost noone supports extrude, not
                everyone has support for eval(),<br>
              </blockquote>
              <blockquote type="cite">.pseudoclasses aren't widely used,
                shields aren't implemented in most<br>
              </blockquote>
              <blockquote type="cite">software, set and exit; ...<br>
              </blockquote>
              <blockquote type="cite">So, we make a list of core
                features that everyone supports or will be<br>
              </blockquote>
              <blockquote type="cite">able to support in the nearest
                time.<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">4. Split all the stuff that's
                chopped from 0.2 spec into some separate<br>
              </blockquote>
              <blockquote type="cite">features for further discussion.<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">5. Draw beautiful badges "Powered
                by MapCSS" for sites and renderers.<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">Are there any supporters for this
                initiative? :3<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">Why:<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">Lately, I've tried to make a
                stylesheet that will look the same in<br>
              </blockquote>
              <blockquote type="cite">kothic, komap, potlatch and josm,
                and wasn't able to do it, because<br>
              </blockquote>
              <blockquote type="cite">even basic things like
                casing-width are handled differently in all of<br>
              </blockquote>
              <blockquote type="cite">these.<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite">I just want to have a strict
                subset of MapCSS that will work<br>
              </blockquote>
              <blockquote type="cite">everywhere for sure. :3<br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <blockquote type="cite"><br>
              </blockquote>
              <br>
              <br>
              _______________________________________________<br>
              Mapcss mailing list<br>
              <a moz-do-not-send="true"
                href="mailto:Mapcss@openstreetmap.org">Mapcss@openstreetmap.org</a><br>
              <a class="moz-txt-link-freetext" href="http://lists.openstreetmap.org/listinfo/mapcss">http://lists.openstreetmap.org/listinfo/mapcss</a><br>
            </div>
          </blockquote>
        </div>
        <br>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
Mapcss mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Mapcss@openstreetmap.org">Mapcss@openstreetmap.org</a>
<a class="moz-txt-link-freetext" href="http://lists.openstreetmap.org/listinfo/mapcss">http://lists.openstreetmap.org/listinfo/mapcss</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>