I found the following to be an interesting read: http://www.cdt.org/spam/
It shows that even five years ago or so most ligitimate businesses advertising legitimate services through spam employed header forgery. ...
It is an article of faith for many people that most spam involves header forgery, but no one seems to have better support than intuition for that faith. Where in the report at http://www.cdt.org/spam/ does it say that "most ligitimate businesses advertising legitimate services through spam employed header forgery?"
Ok, I can't find it right now, and the thing is too big to completely reread.
A lot of spam does involve header forgery, but a lot clearly does not. The problem with concluding that "most" spam uses header forgery is that it encourages looking for solutions to header forgery instead of stopping unsolicited bulk mail.
Stopping header forgery would be useful in and of itself, but regardless of that it will also help against unsollicitated bulk email. I downloaded the list with known spam address blocks from spews.org. It lists around 1600 spammers and 14000 addresses or address blocks. Obviously spammers are trying hard to cover their tracks. Filtering out 1600 spammers is easier than filtering out many more thousands of individual addresses. I'm assuming we can come up with some identifier that's harder to change than an IP address.
A quick look at a week's worth of email for an account I've used to post to Usenet for nearly 10 years (370, 98% spam or mailinglists I can't unsubscribe) tells me around 75% of the spam I received either has obvious header problems or employs some kind of anti-anti-spam measure. Also around 75% is of a pharmaceutical, sexual or financial nature (often at least two of those at the same time). There is no obvious correlation.
That leads to a major problem in dealing with spam. Most people who say they want to stop spam in fact have other goals that they value more.
Yes, it seems like many of them are more interested in perpetuating the status quo.