If it were me....  I would include a DOCTYPE and ensure that the 
fragment was actually legal xml (the example below is not).  Somehting like:


<!DOCTYPE p
     PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
     "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd ">
<p>This is <em>really</em> cool</p>

And use a content type of application/xml-external-parsed-entity.  Then my parser can look at the DOCTYPE 
if it chooses, but no one will think that it is actually XHTML.  



Garret Wilson wrote:

> Thanks, but that's not a full solution---how does the application (the 
> Wiki, the newsfeed---whatever) know whether to process and interpret 
> tags? If it were really plain text, the "this is <em>really</em> cool" 
> could be a plain text example of how to use to use markup, rather than 
> actual use of markup.
>
> Garret
>
> Spartanicus wrote:
>
>>Garret Wilson <garret@globalmentor.com> wrote:
>>
>>  
>>
>>>A common use nowadays, especially with wikis and newsfeeds, is to store 
>>>XHTML fragments (such as "this is <em>really</em> cool") to be later 
>>>integrated into a larger XHTML document.
>>>
>>>What content type should we use for XHTML fragments?
>>>    
>>>
>>
>>As long as the application that processes the fragments handles it
>>properly it shouldn't matter, as these fragments should never be
>>accessible in other ways.
>>
>>I'd use text/plain
>>
>>  
>>

-- 
Shane P. McCarron                          Phone: +1 763 786-8160 x120
Managing Director                            Fax: +1 763 786-8180
ApTest Minnesota                            Inet: shane@aptest.com