Latest commit b1f0c11
Apr 5, 2011
|Failed to load latest commit information.|
INTRODUCTION ============ Penny Red is an Hashcash implementation for Mutt using Python. Penny Red combats spam email by making the sender pay for the email using CPU cycles. Outgoing mail is marked with a token identified with the "X-Hashcash" header. Every recipient receives their own token. Minting may take a second or two per token. Recipients verify the token is valid using the Hashcash binary, and keep a copy of the token in a local database. Users are notified whether or not the token is valid, spent or invalid. If the token is valid, the recipient can be convinced that the sender of the mail is genuine. ABOUT PENNY RED =============== Penny Red is a Hashcash frontend for the Mutt MUA. Thus, Penny Red assumes that you have Hashcash and Mutt installed. Because Penny Red is a Python solution, it also assumes you have Python installed to execute the scripts. Penny Red is not for other MUAs, such as Gnus, Alpine or others, although you could probably fork the code, and make it work with other MUAs as needed. Penny Red was developed as a need to get Mutt working with Hashcash. The solutions that currently existed at the time didn't fit well with using Mutt to connect to Gmail via IMAPS/SMTPS. Thus, the "hashcash-sendmail" wouldn't work, and Tim Ruddick's Perl solution did not come with any documentation on how to install it. So, Penny Red was created. Penny Red requires some changes to your ~/.muttrc and has some basic minimum software requirments to work. See the CONFIG and REQUIREMENTS files for details. WHY HASHCASH WORKS ================== Hashcash works, because it is not based on a trial-and-error approach like most spam filters. Instead, each sender must pay for the email by minting tokens for every recipient in the mail. This payment is made in terms of computer time rather than actual money. The additional computer time required is negligent, and the sender will likely not be bothered by the additional time it takes to send an email. Hashcash tokens cannot be forged, as they rely on a valid SHA1 sum to start with a specified number of bits to be zero. Further, tokens have a valid timestamp on the date of their mint. Hashcash expires tokens (28 days by default), so old tokens cannot be reused. Here's an example of minting a valid token: $ hashcash -m email@example.com hashcash stamp: 1:20:110330:firstname.lastname@example.org::q/Vef65O3EV46S/3:0057QH $ echo -n hashcash stamp: 1:20:110330:email@example.com::q/Vef65O3EV46S/3:0057QH | sha1sum 00b566a8acec0f4568b1e1e9fc5a35794e483c10 - Because the token "1:20:110330:firstname.lastname@example.org::q/Vef65O3EV46S/3:0057QH" has a valid date it was minted "110330 = 03/30/2011", and because the token itself produces the sha1sum of 00b566a8acec0f4568b1e1e9fc5a35794e483c10, the token is valid. Hashcash is designed to be slow to mint, but fast to verify. Thus, users who receive email with valid tokens in the header can verify the validity in a fraction of a second, where it may have taken the sender up to a full second to create. This helps mitigate spam as spammers depend on the ability to send billions of email a day. There are only 86,400 seconds in one day, so minting tokens for every mail would limit the spam ring operations to 86400 mails per computer, a far cry from the goal of sending billions. ADVANTAGES ========== Many antispam software solutions, such as SpamAssassin, support Hashcash tokens in the mail header. Because Hashcash is designed to be quick to verify, but slow to mint, SpamAssassin can do the verification on the MTA with fractional overhead. If the token verifies, the antispam software can then mark the mail as non-spam, and deliver the mail to the INBOX, even if it would have otherwise failed. Thus, users who send mail with valid Hashcash tokens in the mail header can have confidence that their mail can reach the INBOX of the recipient. Of course, the email administrator must configure the antispam software to accept tokens. So if the sender does send an email with valid minted tokens, that ends up in the SPAM folder of the recipient, the sender could send an email to the email administrator, asking for them to enable checking for valid tokens. Of course, the biggest advantage of Hashcash is the ability to virtually stop spam in its tracks. If more and more people use proof-of-work systems, such as Hashcash, then we can begin demanding that email have valid tokens attached, or mark the mail as spam. When we reach this point, spammers will have to come up with new hardware solutions for minting the tokens, and getting the mail into our INBOX. ADDITIONAL INFO =============== Please see http://hashcash.org/ for more information about Hashcash. HISTORICAL PENNY RED ==================== The name for this project comes from the "One Penny Red" British postal stamp of 1841. It succeeded the Penny Black stamp, and continued as the main type of postage stamp in the United Kingdom of Great Britain until 1879. The stamp was changed from black to red, because seeing the cancellation mark on a Penny Black stamp proved difficult, while it was easier to see on a Penny Red stamp. See http://en.wikipedia.org/wiki/Penny_Red for more historical info.