From: Russ Allbery (rra@stanford.edu)
Date: Mon Sep 20 1999 - 16:10:03 CDT
Brad Templeton <brad@main.templetons.com> writes:
> A new semantic was proposed for Distribution which is similar to the
> original but fixes the problems.
> The concept was to supplant the use of distribution in the control of
> outgoing feeds with using it to control incoming feeds. (Outgoing feed
> control can still remain but it will always be unreliable over multi
> hops.)
> A news database or transport instead maintains a list of the
> distributions which it considers itself to be a "member" of.
This is the way that INN has dealt with Distribution for years and years
now. It still hasn't made the traditional semantics of Distribution as
representing a geographic area more useful; the fatal flaw in that system
lies with the fact that news servers often cannot be usefully linked to a
particular set of geographic areas and geographic distributions are almost
never used in any sensible fashion. I'm not going to reject messages
claiming to be limited in distribution to France when I have French users
who want to read them.
Distribution is useful for identifying and controlling news propagation
within a subnet. Beyond that, I think the realities of modern news
administration make it highly unlikely it will be useful, and more
complicated semantics doesn't really change that.
> Thus if I want to issue a Nocem stream, I would create a distribution
> and use it on all my messages. Sites not in my distribution would
> ignore or discard the articles.
It's not useful for me to have to continually add more new distributions
as the crop up so that my users can continue to receive all Usenet
articles in the groups that they read. My server is therefore configured
to receive all distributions, despite the fact that the code fully
supports your proposed semantics already. From discussing this with other
server administrators, this policy is very common, and the decision to
implement it has generally been made intentionally.
Given the existence of a large number of sites with that policy, the
semantics of Distribution break down to the point where it's really not
useful except as mentioned above, to control the propagation of articles
within restricted subnets.
-- Russ Allbery (rra@stanford.edu) <URL:http://www.eyrie.org/~eagle/>