On Fri, 27 Jan 2006, Andrew McKeon wrote: [snip long argument that is unrelated to the current discussion]
First you quote from the RFC to show that placing the new headers at the top is valid, and SHOULD be maintained in that order throughout delivery. So far so good... you're proving that the -I option is plausible.
Then you go on to claim that somewhere far downstream the headers could potentially be rearranged by a mailserver that doesn't follow the recommendations of the RFC, and decides to reorder the headers. Nice claim, and it's even valid, but it's entirely irrelevant to the discussion. We're talking about receiving mail here, and as such *we* control all downstream mailservers. Since they're under our control, it's easy enough to ensure that they *don't* reorder the headers.
I don't understand the strong resistance to functionality that may be useful for some. Claiming that it's a Bad Idea just because you personally don't have an immediate need for it is just silly. I don't add the headers, so this thread doesn't affect my setup, but I can certainly see how it makes sense to add them at the top whenever possible. It's a best-effort sort of thing. Nobody depends on it (all sensible admins reject at smtp time) but it's nice to have an audit trail so you can see which filters have been applied, by which mailservers.
Damian Menscher -- -=#| <[EMAIL PROTECTED]> www.uiuc.edu/~menscher/ Ofc:(650)253-2757 |#=- -=#| The above opinions are not necessarily those of my employers. |#=- _______________________________________________ http://lurker.clamav.net/list/clamav-devel.html
