The spam fighting scheme is not a workable idea. Sorry.
If the person doing the verifying is a human, they don't need the stamp -- they can (and will) judge the mail in an instant based on their own arbitrary and capricious criteria, such as "Mail from Mom is not spam" or "Mail from the merchant whose double opt-in list I just signed up for is spam if I don't know how to delete in Gmail yet". If the user doing verifying is a machine... well, in point of fact, there are no servers on the Internet who support your own one-off anti-spam measure. (If you're just wrapping a slightly-more-well-known measure in attractive mapping, you win marketing points and still lose on deliverability.) You have no buyin from Gmail, Microsoft, et al who have the email accounts that people actually want to deliver to.
Your plan fills no business need for anyone. If I want to pay $0.01 to make sure my email gets delivered, I have an option for that: MailChimp. (It works fairly well for me, incidentally.) They do it by keeping in good graces with the major mail providers, kicking off clients, staying away from banned lists, and aggressive list scrubbing. CentMail will have to do all these things better than the commercial providers to make sense as a commercial offering.
The $5 to charity is not improved by the included bundling of the spam fighting scheme. If you want to donate $5 to charity, you can do that right now.
P.S. I used to work in anti-spam research at my previous day job. I have a lot of natural sympathy... for approaches which work.
That was my thought too. Can it be improved? If there is an API to create a unique key based on the contents of the message (and the headers) that the client can then verify with centmail (or w/e).
Sure it wont do anything for mail NOT covered with centmail. BUT you can dump anything with a faked key (likely spam) and let in those with a real key. Then anything else goes to the spam filter as usual.
Kinda like a more worldly version of PGP signing.
Obviously that kind of scheme would require much more support from email vendors.
The main thing we're hoping for though, to achieve any kind of critical mass, is that you'll want to sign up initially as a sender simply to promote your favorite cause. Confer the popularity of Causes, the Facebook app.
We certainly don't imagine Centmail as a panacea; but you could imagine it becoming a reliable additional feature for your spam filter.
Think of it like hashcash, but instead of burning money/effort, you send it somewhere useful. Come to think of it, maybe this could be made more palatable by combining the hashcash concept with some seti@home-style distributed computation problem. Look for aliens in order to send email!
what if your idea of good isn't the same as mine? maybe i don't want my money going to xyz? and would you be sending 100% of the donation through? If I want to do good, I'll do what I already do, donate directly to the charities of my choice.