Many people have spent many nights lying awake trying to figure out what to do about spam. The the extent that when a person believes they have come up with an idea that is both new and useful, they are usually wrong. This results in some hostile attitudes toward new ideas - I have certainly felt that hostility myself enough times.
You may find this entertaining: http://www.rhyolite.com/anti-spam/you-might-be.html This mailing list exists for discussing the overall problem of spam: http://irtf.org/mailman/listinfo/asrg On 11/23, Christian Grunfeld wrote: > *check spam folder always This is your biggest problem. Most people aren't willing to spend their time reading all their spam to make sure it's all actually spam. That's kind of the whole point of spam filters. > False positives....yes, ONLY the first time for each sender! just I've spent years helping with dnswl.org, trying to list all legitimate sender IPs, and it still gets false positives (and false negatives). I've recently started a personal project to assemble similar data in a different fashion, http://www.chaosreigns.com/iprep/ The problem is, while most people might get email from most spammers, most people only get mail from a small percentage of non-spammer mail servers. There are many of them. Tiny companies and personal servers all over the world, that you will *never* get an email from. So even if you have a large group of people trying to collectively come up with a list of places that don't send spam, it's still hard, you will still get some false positives. Do some analysis on your own email to see it for yourself. > So, what do we have to waste resources on tons of rules, tons of perl > code, tons of regex if we know that 80% is spam? lets mark all of them > as spam and let this method work! I'd love to develop a public whitelist thorough enough that that could become even a remote consideration. It's not going to work on an individual user level. By all means, please prove me wrong. Come up with a clean implementation for spamassassin, submit it as a patch (against trunk) via https://issues.apache.org/SpamAssassin/ On 11/23, Christian Grunfeld wrote: > > Well, if I have to do *that*, I might as well not do any filtering at all. > > The whole purpose of anti-spam software is to shield me from spam. > > Not 100% correct. Now I always check spam folder, dont you? You and I and David are not the issue. Our use is not representative of a statistically significant percentage of users. > Do you advise your people not to check spam folders? Are you 100% sure > that machines can sort 100% efectively what is spam and what is not? What is advised and what can be expected are not the same. On 11/23, Christian Grunfeld wrote: > let people who wants spam to answer spam ! if you dont want spam dont > reply. Easy ! No. The problem is everyone else who is still providing a profit to spammers. Not easy to fix. > There are a lot of people who wants to sell viagra and send > spam....but I dont answer to them, dou you? :p Again, the behavior of people on this list is insignificant compared to the behavior of average users. > > Many users do this only sporadically, if they do. Some users don't know > > where to find the spam folder. Some organizations do not deploy per-user > > quarantine area or spam folders. Etc. etc. > > bad thing ! Sure. Fix it. > Just answering with your MUA. The email goes back through the MTA. > Then it can put the sender in the whitelist. That's not something that's particularly well implemented. -- "Every man, woman and child on the face of this earth is at the mercy of chaos." - a maxwell smart movie http://www.ChaosReigns.com
