[Ietf-calsify] Character set restriction in section 4.3.11 Text

[Alexey Melnikov]

Mark Crispin wrote:

> Well, then, which is more important; having calendar fall back to 
> TEXT/PLAIN behavior (in which case you need CHARSET), or having 
> calendar objects be required to be UTF-8?
>
> Another thing that you could do is that TEXT/CALENDAR could require 
> that ;CHARSET=UTF-8 be present as a parameter.

I think this is the best way.

> Declare that a TEXT/CALENDAR which omits CHARSET, or has any charset 
> other than UTF-8, is undefined and not to be interpreted according to 
> the calendar specification.

I wouldn't go as far as saying that they are not covered by the calendar 
specification.
For example I find nothing wrong with a calendar that is entirely in 
US-ASCII and has to CHARSET parameter.

> By the way, all of these are to be treated as random ideas.  I'm not 
> pushing for any of this as a solution, although I believe that it 
> would be highly desirable to prevent any new application from getting 
> into the multi-charset business.

Right.