Re: media type?

[Nathaniel Borenstein <nsb@xxxxxxxxxxxxx>]

There is ample precedent for confusion here.  Most applications that 
want to make use of MIME media types will want at least form #2, but 
naive implementors may stop at type/subtype and ignore parameters.  In 
particular, I think JSR 168 (the portal/portlet specification) is going 
to have to be amended to clarify this very point, as there are 
apparently implementations that won't support parameters in the media 
type.  -- Nathaniel

On Thursday, November 27, 2003, at 11:36  PM, Jeroen Bekaert wrote:

> Dear Valdis,
>
>
> Thanks for your reply! If I understand this correctly, defining Y as:
>
> 1. "a MIME media type value indicating the type of a resource" results 
> in:
> "text"
>
> 2. "a concatenation of MIME media-type, subtype and parameters 
> indicating
> the type of a resource" results in: "text/xml; charset=utf-8"
>
> 3. "a MIME content-type value indicating the type of a resource" 
> results in:
> "Content-Type: text/xml; charset=utf-8"
>
>
> Regards
> Jeroen Bekaert
>
>
> > -----Original Message-----
> > From: owner-ietf@xxxxxxxx [mailto:owner-ietf@xxxxxxxx] On Behalf Of
> > Valdis.Kletnieks@xxxxxx
> > Sent: Thursday, November 27, 2003 8:34 PM
> > To: Jeroen Bekaert
> > Cc: ietf@xxxxxxxx
> > Subject: Re: media type?
> >
> > On Thu, 27 Nov 2003 18:32:50 MST, Jeroen Bekaert
> > <jeroen.bekaert@xxxxxxxx>  said:
> >
> > > As such, a possible value of Y could be "text/xml; charset=utf-8"?
> >
> > > As such, a possible value of Y could be: "text"
> >
> > You want media-type to be "text" with a  subtype of "xml" and
> > a parameter "charset=utf-8".
> >
> > I don't see any way to use the BNF of RFC2046, section 5.1 to get any
> > other parse.  In fact, further down in section 5, it says:
> >
> >    type of data.  Thus, a media type of "image/xyz" is enough to tell
> > a
> >    user agent that the data is an image, even if the user agent has
> > no
> >    knowledge of the specific image format "xyz".  Such information
> > can
> >    be used, for example, to decide whether or not to show a user the
> > raw
> >    data from an unrecognized subtype -- such an action might be
> >    reasonable for unrecognized subtypes of text, but not for
> >    unrecognized subtypes of image or audio.  For this reason,
> > registered
> >
> > So the intent is to have the media-type say "this is text" or "this
> > is an
> > image" or "this is a sound" or "this is a composite", and so on, so
> > that an MUA
> > can make a vague guess as to what do with it even if it has no clue
> > at all what
> > to do with the specific subtype or parameters. For instance, an MUA
> > would be
> > within it's rights to refused to deal with *any* flavor of text/*
> > (including
> > even text/plain) if it included a 'charset=' parameter it didn't know
> > how to
> > deal with.  As an example, displaying *any* sort of Big5 text on a
> > display that
> > is ascii/8859-* only is basically doomed...