Re: ISSUE-41/ACTION-97 decentralized-extensibility

> Namespaces are one solution, but even a standardized prefixing 
> mechanism would meet this requirement.

Just a simple thought, but one other vehicle for extensiblity is/could 
be MIME.

For example, in the following html5, which would be served as 
text/html, @type could certaily give the browser a clue that it was 
dealing with data that at some level could be considered at least 
providing an intention that the contained code could be considered to 
be a nested context. In the case of MathML and Ruby, I am thinking 
they are integrated closely into the browser and thus need no special 

Again, I say served as text/html because it is the author's intent is 
that the browser plow through it and do the best it can, not necessary 
halting for 'simple' problems. If I wanted to find all problems I 
serve it as application/xhtml+xml and use all the latest XML aids.

Anyway, just tossinig this in in the spirit I saw in the HTML5 intro 
video and little production value.

<!DOCTYPE html>
    <meta http-equiv="Content-Type" content="text/html; 
    <title>Draft Example #4, served as text/html</title>
    <style type=text/css'>
      X3D { height:100%;width:100%; }
      svg { height:100%;width:100%; }
    <X3D name='x3dBlock' type='model/x3d+xml'
      version='3.3' profile='Interchange'
      allow-same-origin allow-scripts>
        <Viewpoint description='Start' />
          <Box size="4 4 4" />
          <Text string='"This is X3D Text"' />
    <svg name='svgBlock' type='image/svg+xml' version="1.1">
      <ellipse cx="2cm" cy="4cm" rx="2cm" ry="1cm" />
      <mrow><mi> x </mi><mo> + </mo>
      <mrow><mi> a </mi><mo> / </mo><mi> b </mi>
    <ruby> WWW <rt>World Wide Web</rt></ruby>

oops, forgot <canvas> 2D and <canvas> 3D examples.

Thank to All and Best Regards,

Received on Friday, 16 October 2009 19:35:54 UTC