I have had a fairly long discussion via email with the AARnnet administrator involved. He points out that wrongly or rightly (he believes wrongly) the AARNet does not have an "open access" policy, and the network is setup exclusively to service the university community. Public access systems are tolerated, but barely, and mainly through the grace of those who administer the system rather than those who fund it. The complaint in question was actually not at all specific, and came not from the NSF but from one of the NASA Internet officers who is responsible for the US end of the link to Australia (and pay for some of it). Essentially the complaint was one of increasing mail traffic on an already congested link to the US, as well as concern about the "hiding people's identities so they cannot be responsible for what they say". Personally I disagree with the second complaint, but cannot dispute the first, without statistsics about what component of the link was being consumed by the posting service. I suspect it was very small but these things all add up. It seems a shame that the anonymous system is being terminated "on principle" but the AARNet person has been friendly about it, in fact positively graceful in view of my somewhat inflammatory post, and so I guess I just have to leave it there. Hopefully eventually commercial vendors will provide an alternative channel to the university-based network here currently, much as has happened in the US over the years, and these questions will be less of a concern. david