Sorry for the repost, my *cluelessness* led me hit send before I was finished. Ironic, eh?
> From: Vince Sabio [mailto:[EMAIL PROTECTED]] > > But no more. I'm becoming more and more convinced that this is An > Idea Whose Time Has Come(tm). I thought so too, but was hesitant to say so in SPAM-L :) Just call me meek. Btw - nice name Vince - "CURES". > My comments on SPAM-L: > > The lists could easily be downloadable via e-mail (what else?) using > an AR, and then just dropped into the list server's > "banned" database. I was envisioning some slightly less drastic, and more palatable to privacy watchdogs than an outright "blacklist". I definitely agree that access to the list should be restricted. And perhaps that restriction should be limited to people who belong to (tada!) - "The Association of List Owners". Just another idea floating in the back of my head. But I digress; I think such a database should different types of listings & corresponding actions. Examples: Type ==> Action 1. Cluelessness => flag posts for attention by owner/moderator/editor. Maybe extra instruction is needed (ie - learn how to use your vacation program correctly) 2. Technical issues => flag posts for attention my owner/moderator/editor; user may have trouble with unsubscribes 3. Trusted users => flag users that can be trusted not to spam. (on my system, I can individually control whether members are moderated or not; all new members are moderated until I see fit to let them post freely. A "trusted user" flag would let me let these people post freely right from the start) 4. Abused users => flag people who have been subscription-bombed, joe-jobbed, or otherwise attacked; this would permit the list-owner to identify (in real-time) these people and help prevent their systems from being used against the person. 5. Abusive behaviour => whatever the list owner likes #5 is where this begins to break down. Saying someone is "abusive" is a judgement call that may bring civil liabilities to bear on either the person who submitted the user, or on the CURES operators. And "abusive" is contextual, and in the eyes of the beholder; what constitutes normal rhetoric on SPAM-L most certainly is abusive in any other forum. I still think such a system is useful, but its purpose should be limited to identifying those users who need additional care and attention. Also, on most lists, people need the freedom to express themselves without worry that they may fall into some secret ALO blacklist. So there are other issues as well. I also think the system should be query-based; if it extends past being a "banned users" list, for example, to include #3 above, then it becomes a target for unscrupulous people who want to get their hands on a list of addresses. #4 above would be most useful in a real-time environment: for every subscription request, do a query against a central server. The server checks it's database to see if other subscriptions have occurred in the past nnn hours/days for this user, and returns a rating that indicates how likely it is that the subscription is not legitimate. To make it fast and effective through firewalls, the lookup would need to use HTTP. I've got lots of other ideas about this that expand the concept even more, but I need to get back to work now. I may get down to business in the next few weeks and develop #4, if anyone thinks it's useful. Let me know what you think. -- Rick Vandenberg
