Reply-To munging is bad (Was: Benchmarking stuff)

On 12 Mar 2005, at 23:33, pps wrote:

> What's wrong with the list?

Nothing is wrong with the mailing list.

> It doesn't have Reply-To headers.

Overwriting Reply-To is wrong.  Mailing lists that set Reply-To are 
bad.  If a mailing list sets 'Reply-To', then I cannot reply to 
somebody in private easily.  (Such as yourself, which I would, if you 
had an email address that looked like it would guarantee you would get 
my email.)  Instead I have to do extra work to replace the list address 
with your email address.  That's lame.

Also, I may send a mail to the list from one address, but want replies 
redirected to a separate address.  Overwriting the Reply-To destroys 
this functionality completely.  Who is the list to say where I want my 
mail sent?

> All reaplies I wrote went to personal mailboxes.

You hit 'Reply', right?  How is this broken?  It seems to be exactly 
what you asked to happen.

My mailer has a 'Reply-to-all' button.  Perhaps yours is missing this 
critical (and ancient) functionality?

The biggest thing I can say against Reply-To munging is that a rather 
well-known person wrote a very personal message to somebody by using 
the Reply button of their mailer, expecting it to go to the original 
sender.  (Perhaps they thought "I'll just replace the mailing list with 
the person I want to email's address" then forgot to.)

Since the list used Reply-To munging, instead this very personal 
message went to the list!

The list admins said "This is never to be spoken of again", but I have 
a copy of that very message and I imagine a number of other list 
subscribers also have that message, and this very personal email could 
surface somewhere on the internet again, much to this well-known 
person's embarrassment.  (Although it is unlikely, since it was a fan 
list.)

Oh, you may want to read this:

http://www.unicom.com/pw/reply-to-harmful.html

-- 
Eric Hodel - [email protected] - http://segment7.net
FEC2 57F1 D465 EB15 5D6E  7C11 332A 551C 796C 9F04

___________________________________
fastcgi-developers mailing list
http://fastcgi.com/fastcgi-developers/