Stephanie da Silva wrote:
>
> I don't mind if someone has list-request aimed at a person or an auto-
> responder, so long as you get something or someone that responds back
> (although there are too many listowners out there who simply do not
> read list admin mail/write back - different issue).
>
> What I don't like are implementations that point it at the listserver
> or at an autosubscriber.
I guess I'm not clear what the difference is between an 'autoresponder'
and an 'autosubscriber'. Some kind of automated intelligence at the -request
address strikes me as a Good Thing. And if that encompasses looking for
obvious subscribe and unsubscribe requests (and returning a confirmation
cookie for the former), then that strikes me as a Very Good Thing.
In my case, anything that doesn't readily parse returns the help file,
which among other things says that if you can't get the desired results
write to [my address]. But it also says to let me know that you've
already tried the -request address, because my default response to
that kind of e-mail is to send the exact same help file. (And I made it
easy for myself by setting up [EMAIL PROTECTED] as an address which does
that, so all I have to do is 'bounce' mail to it.)
Intelligence at the LIST address to try to catch some of the 'help me get
uns*bscribed from this &^%^%$ list' messages is also a Good Idea.
A question for Chuq: How do you build properly formatted action requests
by clueless/impatient subscribers into your new paradigm?
