Jesper Tverskov wrote:
> 1) What would you consider Best Practice to use as mime-type for user
> agents, web spiders, etc., not sending an HTTP accept-header?

Assume text/html, unless told otherwise.

> I have decided to give them the best I have to offer,
> application/xhtml+xml. I have two arguments. A) Since they don't send an
> http accept-header they either don't need to know about mime-types in
> order to work or they are not worth dealing with. B) Google is not
> sending an accept-header, and I would like Google to be able to detect
> that I offer my XHTML as XML.

Google does not support XHTML served as XML.  Although it will still 
index the documents, it will simply state "File Format Unrecognised" 
within search results.

> 2) Is it a bug for Mozilla/FireFox suddenly to require that we also
> style the html element with background-color similar to the body
> element?

No, it's not a bug.  It's simply the way rendering real XHTML works. 
That is described within the Mozilla FAQ document you linked to, but in 
the following question [1]:
| # CSS works according to the XML+CSS rules.
|   * HTML-specific CSS exceptions do not apply. For example, the body
|     element gets no special treatment.

[2] http://www.mozilla.org/docs/web-developer/faq.html#xhtmldiff 

-- 
Lachlan Hunt
http://lachy.id.au/ 
http://GetFirefox.com/     Rediscover the Web
http://SpreadFirefox.com/    Igniting the Web