I am worried about the tools we are developing and deploying to control spam. Some of them are esentially centralsied methods of controlling Internet content. Paul's anti-spam feed for instance prevents users of some providers from seeing spam. The user has no choice; they cannot opt to receive spam other than by switching to another provider. Even worse: they may not even be aware that they are "missing" some content. Combatting spam is considered a Good Thing(TM) by almost everybody here, including myself. However the same technology could just as easily be used to do Bad Things(TM). Even worse: if it works it demonstrates that *centralised control* of the content of Internet services like e-mail is *feasible*. This will give some people ideas we may not like, and sometime in the future we may ask ourselves why we have done this. The end does not always justify the means. I hope that methods like the anti-spam feed will not be taken up widely. Please consider the consequences before you use them. I stress that I do not question the morality or good intentions of those involved. I am just concerned about the almost ubiquitous and apparently unreflected zeal that spam seems to evoke and the danger of it making us accept methods we would otherwise despise. I would prefer to see more work in technology that is less centralised and gives the users a choice of the content they wish to see. Yes this may be harder to do, but the consequences of deploying the easier methods may be just too severe. Waehret den Anfaengen (beware of the beginnings) Daniel PS: I hope this is more coherent than my contribution at the meeting yesterday when my brain failed due to jet-lag while my mouth was still working perfectly ;-).