Re: Feedback on 0.3 Format

[Graham <dtcd@xxxxxxx>]

> 1: I <3 NY
text/plain - escaped or xml
OR xhtml - xml

> 2: I <3 NY
html - escaped

> 3: I <strong>&lt;3</strong> NY
xhtml - xml
(it might also be valid text/plain - xml, but that's a matter for the 
courts)

> 4: I &lt;strong&gt;&amp;lt;3&lt;/strong&gt; NY
html - escaped

It's always been my understanding that when parsing, the string you get 
out of a standard XML processor (which does one level of unescaping) 
*is* the document as it would appear in disk. Your suggesting we use 
different rules depending on the media type, aren't you? Since in your 
scheme '<title mode="xml" type="text/plain">I &amp;lt;3 NY</title>' 
represents the document "I <3 NY" whereas '<title mode="xml" 
type="text/html">I &amp;lt;3 NY</title>' represents the document "I 
&lt;3 NY". There's no way to consistently reply such rules across 
arbitrary media types.

Or what Dare said.

(This exchange also explains why the current mode names are bonkers)

Graham