The "[listname]" insertion is clearly the "sore point" case. This is
natural because it seems to defy the otherwise reasonable-sounding rule
that i myself proposed, which is that the Subject field is for human
beings. However, I'm not sure that this is really as bad a layering
violation as it might seem. While the list processor is indeed
inserting something into the "human" part of the stream, it is (at
least in theory) doing so in accord with a policy set by a list owner.
A list owner should be able to choose to insert any strings he wants to
in messages that go through his list, to convey a meaning that he wants
to convey. The fact that list processors make it so easy to insert
such strings may be simply a reflection of the fact that so many list
owners want to be able to do that. But they want it for its human
meaning, not for any intended protocol consequences.
I would certainly be unhappy with a list processor that *required* the
insertion of such strings, but when a list administrator chooses to
have them inserted,this is the human moderator's decision to "put his
mark" on all messages to the list. Sort of like a dog urinating on a
tree -- it may not always be pretty or desirable, but it's going to be
awfully hard to design a tree that can prevent itself from being peed
on. -- Nathaniel
On Nov 30, 2004, at 7:39 PM, Steve Dorner wrote:
At 7:16 PM -0500 11/30/04, Nathaniel Borenstein wrote:
I think we need to distinguish between human-level conventions and
protocol elements. Any "harmfulness" in the former is not something
that can be addressed technically.
Ok, human bad, protocol element good. That's quite clear. Pesky
Is [nameoflist] inserted by a list processor a human-level convention?
If a machine puts it there, I think it's reasonable for another
machine to try to take it out. If we as an Internet community can
make that easier, we will have done a good thing.
Where are we going, and why are we in this handbasket?