Due to a disk crash and backup failure, this site has been restored from an old backup with a number of more recent articles missing. The missing site content is being restored as time permits. We apologise for any inconvenience.
Nutters.org The Nutter Log
MAPS: The Phantom Menace? Entry id: maps
By The Famous Brett Watson
On Wed, 23 May 2001 15:22:00 +1000

In Are Anti-Spam Policies Censorship? I discussed some of the questions surrounding the borders of spam-blocking and censorship, with particular reference to the case of John Gilmore and his open mail relay. Since then there have been a couple of media-fusses about MAPS versus free speech advocates like Peacefire (which wound up in the RBL because MAPS had a bone to pick with their spammer-friendly ISP, not Peacefire themselves). The latest such media-fuss I've read is an article at The Register, and in this particular instance I was moved to write a response, an edited version of which follows.

I believe the report contains a number of factual inaccuracies, but it may be that I am misinformed myself. I here list my main points of contention in the hopes that we can come to some agreement with regards to the facts.

"MAPS hunts for open mail relays that will allow spammers to use companies' servers and drive millions of people mad with worthless messages. If it finds one, it blacklists it by putting it on its RBL list."

To my knowledge, MAPS hunts for nothing. They accept nominations for the RBL only after every other reasonable method of resolution has been tried. Only one branch of MAPS has anything to do with open relays, and that's not the RBL, it's the Relay Spam Stopper (RSS). The RBL is intended primarily for sites and service providers that have unacceptable policies with regards to the use of email and have made a nuisance of themselves.

"They will blacklist someone but not tell them - it's their fault if people have open servers."

As far as I am aware, all the MAPS initiatives make reasonable attempts at informing the target of the blacklist of the action. In the case of open mail relays (RSS) this is often impossible: the same slackness or cluelessness that caused the mail relay to be open in the first place means that the standard channels for contacting the postmaster don't work. RBL listings are considered severe, and serious effort is put into resolving the issue before the block is enforced. No doubt the block is put in place if the targeted party can not be contacted or will not cooperate.

"But ahead of that (at least from our perspective) is that they will never talk to anyone, never explain or defend themselves and certainly never apologise."

If they talk to the targeted party about the issue, I consider that enough. They have explained themselves sufficiently that it should be obvious why any particular offender is given a listing. They have defended themselves from legal action quite well on a number of occasions. To whom and for what do they owe an apology?

"What will hopefully happen though is that people will have the option to decide their level of freedom on the Internet."

The existing protocols simply do not allow this. No perfect solution exists. MAPS, by its own admission, is sometimes obliged to throw out some baby with the bathwater. The alternative is to keep the bathwater. I'm also hopeful that new protocols will arise that address the issue and allow greater individual control, but history has shown that improved protocols are slow on the uptake. I cite IPv6 as an example.

"The only thing that will stop these frequent battles though is if MAPS decides to improve its public relations and stops offending people with its attitude."

Better PR on the part of MAPS will do nothing. This is an ideological divide that there is no point in arguing: the alternatives are incommensurable and irreconcilable. The peacefire-type people think that its offensive to block anything. The MAPS-type people see spam as network abuse and therefore a fair target for blocking. The peacefire-type people think that it's unacceptable to cause any collateral damage in the fight against network abuse. The MAPS people regret the collateral damage, but believe that it's unavoidable in the general case and sometimes necessary to get the required attention of the network abusers.

Put simply, the peacefire-camp disagree with the very means the MAPS-camp believes is necessary to combat the problem. Reconciliation will not happen by better PR between these two camps. If there's to be a resolution, it will have to happen at the level of RBL subscribers. If the peacefire-camp really want to win this war, then they will have to persuade all the RBL-subscribing ISPs out there that spam is a worthwhile price to pay for complete interconnectivity. If nobody subscribes to the RBL, then who cares what they list?

I am an RBL subscriber in my own small way, and from a pragmatic viewpoint I'd much rather have my (free) subcription than not. I'm all for freedom of speech in a general sense, but not when it uses my network and personal resources in ways of which I don't approve (such as spamming). If MAPS really are the sinister and untrustworthy force that they are often painted to be, then I want to hear about it, but so far it just looks like the extremist free-speech crowd have been name-calling; nothing more.

That was the extent of what I wrote to The Register on the matter, but on re-reading it, a couple more thoughts occur to me. Many peacefire-types (as I call them) are in fact anti-spam, and I don't wish to imply otherwise. In my experience, however, they are very touchy about the collateral damage issue. I cite John Gilmore as an example: he takes the view that the first goal of spam blocking should be to "allow and assist every non-spam message to reach its recipients." (In other words the first goal of spam blockers should be not to block.)

Finally, the peacefire-types have put up a co-signed statement against the practice of "stealth blocking". This is what they call the practice of blackholing a network, only they express it in terms of failure to notify the end user that the network address in question has been blocked. I'd agree with this if I thought it were practical, but see my earlier paragraph about protocols and the fact that it can't be done. The blackhole may well be in place at some transit network (neither the originating or destination network), and the best that network can do is report "destination unreachable". From the perspective of the end-user, this looks like a temporary network outage, not a policy decision. If peacefire really want an improvement on this situation, they should suggest a realistic mechanism for implementing it.

Assume for the moment that the reason they aren't suggesting an implementation is because none exists. Their tacit implication is, "if you can't implement blocking nicely, don't do it at all," so their request that blocking be done nicely amounts to one requesting blocking not be done at all. Like I said, I don't see much chance for agreement here.

Public Domain: the author waives copyright on this log entry. Other sources (if any) are quoted with permission or on the principle of "fair dealing" and retain their original copyrights.