ietf-mta-filters
[Top] [All Lists]

Re: vacation

1997-05-26 21:48:28
On Thu, 22 May 1997, Jack De Winter wrote:

At 07:02 PM 5/22/97 -0400, Tim Showalter wrote:
Should the base spec provide a way to do vacation-style forwarding?
What should this mechanism be?

I thought that sieve kind of made a vacation system surpufluous (sp?)?
Are are you talking about the mechanics behind vacation?  If so,
could you spell out what you think those mechanics entail so we can
get a clear picture?

I'm looking for a mechanism to
* specify how frequently a reply comes out
* provide support for storing ids at the filtering agent level

You can't do this in the draft as written, and the reasons you can't do it
are pretty good and have already been discussed.  You can approximate it,
but this has two problems:
* what I just stated is completely impossible, so you have vacation spam
* people won't get it right and you'll have lots of replies to mailing lists

I'd like to avoid the mailing list problem. (What addresses does this guy
have?  I have roughly three addresses that people send mail to, and would
like to accomidate all of them.  That's what I get for using legacy mailers,
I guess.)

A first pass at a comamnd would be
        vacation [-days NN] [-addresses <STRINGLIST>] <REPLY-TEXT>

This is the first command that would have optional arguments.  Maybe it's
appropriate to squeeze them in the draft now so that they'll be consistant;
people will add them.

REPLY-TEXT is the text of the vaction message, as a string (probably
multi-line); -addresses specifies a list of addresses to treat as "mine",
which must be in the to:/cc: lines to generate a reply; -days NN gives a
number to specify the "You'll get this message only every NN days", probably
14, the way the UNIX vacation program does.

I think the above paragraph makes sense, but it probably only works if you
are thinking what I am.  Please gripe at me if it's not clear.

-- 
                                           Tim Showalter 
tjs(_at_)andrew(_dot_)cmu(_dot_)edu


<Prev in Thread] Current Thread [Next in Thread>