Re: extbody URL/mozilla dnd

[Brian Stafford <brian stafford uklinux net>]

On Tue, 16 October 17:59 Carlos Morgado wrote:
>
>On 2001.10.16 17:50:38 +0100 Albrecht Dreß wrote:
>>Am 16.10.2001 00:43:30 schrieb(en) Carlos Morgado:
>>>just commited support for RFC2017 URL external-body type and dnd or
>>>netscape/mozilla urls to the attachment box.
>>>i'm not very sure about the rfc2017 bit, so if anyone knows another mua
>>>that supports it give it a interop test please.
>>
>>I think it is a good idea to implement that, but--- in RFC 2046, which 
>>defines all the message/external-body stuff (or at least this was *my* 
>>reference to implement the support for that) does not mention it. On the 
>>other hand, it does not state that it obsoletes RFC 2017. So I think this is 
>>a little mixed up here. Maybe Brian can comment on this issue?
>>
>
>as far as i can tell 2017 complements 2046 cause 2046 doesn't mention URL,
>and 2017 *is* "Definition of the URL MIME External-Body Access-Type".
>but then again as prolly you could tell from the previous post i'll deny
>any clue on this issues :)
>
>(it was brian who pointed me at 2017, i blame him :))

Just been checking the RFCs and IANA web pages.  RFC 2017 has proposed 
standard status and is not marked as obsolete.  Since RFCs are archival 
documents, once published they cannot be withdrawn, they may only be updated, 
obsoleted or reclassified as historic by a later RFC.

RFC 2046 updates and obsoletes RFC 1738 which defines the same external body 
access types as RFC 2046.  Since there is no mention of the URL access type in 
2046, presumably nothing in 2046 impacts 2017.  I also note that the 
publication dates of 2017 and 2046 differ by only one month and share authors.

My conclusion is that RFC 2017 is "safe" to use.  The caveat is that RFC 2017 
may not be as widely implemented as the rest of the external body stuff.  
OTOH, there is no HTTP access method that I can find, so 2017 is the only 
interoperable way to provide this feature.

The only iffy bit is that the IANA registry of access types does not mention 
RFC 2017.  I assume that this is an oversight, since everything else indicates 
there is no problem with 2017.

Brian