Manuel Lemos wrote: > Hello, > > I thought that some of you would like to know that I just posted this > design proposal for an anti-spam add-on to Mozilla Mail/News in > news:news.mozdev.org/public.mozdev.mozspam . > > Like I just realized about the existence of this news groups (also > available as mailing list with Web archive), I think some of you would > like to follow up there.
I just discovered this newsgroup and thought I would chime in with what another project is doing regarding spam. I apologize for the brevity of description but I'm pressed for time this morning and I'll follow up with folks later. I'm the primary goad for the camram project (http://www.camram.org). It's an antispam project based around the proof of work postage technique known as hashcash. The system was designed to be decentralized and operate on a peer-to-peer basis without requiring direct connectivity between the peers. in a nutshell, any inbound e-mail must must satisfy a challenge otherwise it is placed in a jail. The challenges are either a proof of work postage stamp or a digital signature from a known key. Correspondingly, any outbound e-mail must contain a proof of work postage stamp or signed mail. now obviously, such a system won't work unless "everyone" participates therefore we've developed a variety of accommodations for the transitional phase such as automatic white listing[1] and challenge e-mails informed the sender that there is "postage due" and gives them a mechanism for generating this postage. you can try out the prototype filter and postage due mechanism by sending e-mail to [EMAIL PROTECTED] and following the instructions. as you have already discussed, most of the existing antispam systems have serious failings including unpredictable false positives, false negatives, dependency on centralized infrastructure, denial of service, and breaking end to end connectivity of the net. while camram is not infallible, it doesn't have the same types of problems. Yes, you can get a false positive if there is no one at the senders address to respond to the challenge message (i.e. e-mail robots sending out invoices) but you can't get a false negative unless the spammer goes through the process of generating postage. Another challenges that camram suffers from Moore's law inflation. Known problem but solvable. having worked with all of the other methods of spam control, I am encouraged by the early results with camram. ---eric [1] automatic white listing is not implemented in the same way others have been. A user is white listed only if they have answered a challenge message with appropriate postage or if you have sent mail to the user.
