Re: [Fwd: I-D ACTION:draft-hall-mime-app-mbox-00.txt]

On Tuesday, May 4, 2004, 1:17:35 AM, Steve wrote: SD> I think mbox format that uses binary content-types is probably less SD> interoperable than one that uses text only. That amounts to an assertion that eol conversions and on-the-fly encoding conversion generally do more good than harm. It may be true, but it may not be. Tests using an mbox that contains messages in a variety of encodings (including some multibyte and code shifting encodings) would be needed to determin interoperability in practice. If conversion to a local eol convention is considered desirable, I would rather this was done by the mbox content type handler rather than bringing along the rest of the text/* baggage. -- Chris Lilley mailto:chris@w3.org Chair, W3C SVG Working Group Member, W3C Technical Architecture Group

At 1:32 AM +0200 5/4/04, Chris Lilley wrote:
If conversion to a local eol convention is considered desirable, I would rather this was done by the mbox content type handler rather than bringing along the rest of the text/* baggage.
One could register text/mbox and application/mbox and let the sender choose based on what was actually in the file. Always bearing in mind that the "right" way to do this is multipart/digest anyway. :-)
participants (2)
-
Chris Lilley
-
Steve Dorner