Re: New Internet Draft: draft-duerst-archived-at-00.txt

> > I strongly disagree.  When a message is archived, it should be 
> > faithfully archived.  Or a least, if a message is going to point to 
> > an archive of itself, it should be a faithful archive.  Archives in 
> > other formats are less functional and often lose information as 
> > compared to the original.
> Archives that exist and work (as does the scheme described by Martin) 
> are infinitely more functional than ones that don't.
It's much easier to make the _entire system_ "work" if the archives use 
a format that is already supported by the email readers that will use 
this header field.  All of those email readers support rfc 822 and know 
how to do reasonable things with 822 messages.  Even those that support 
HTML don't know how to do things like reply to messages formatted as 
HTML because when you reformat a message into HTML all of the semantic 
content from the original message is lost.
Archives that preserve the original message format do exist and work, 
are trivial to implement, and used to be the norm.  When I ran several 
IETF mailing lists, every single one of them archived messages in 
original format.  It took around five lines of script to generate a 
unique file name, and write the message to that file.
The only reason we don't see mail archives in original format so much 
today is that many recent archive packages are designed to allow 
messages to be accessed from web browsers.  But there's no reason why 
an archive package can't make a message available in its original 
format (instead of, or in addition to, HTML).  And it's not as if it's 
difficult to tell an http server that an extension of (say) .MIME or 
.822 equates to message/rfc822.  You can even do content-negotiation so 
that a client gets HTML or 822 depending on its preferences.