A lot of effort has been put to stop spam -- mostly unsuccessful. Sadly, spam is not a problem that can be stopped at a single site; every administrator of a SMTP-enabled server must do his best to stop spam, because it takes only one host that allows relaying to spam hundreds of other servers.
The first large-scale attempt to stop spam is found in version 8.8.8 of the popular sendmail SMTP server. With the standard sendmail distribution came a set of anti-relaying rules -- most importantly, they were all turned on by default. Although this caused a lot of headaches for system administrators (who did not know whom to blame when their users could not send mail anymore from their desktop computers until they modified the configuration files for sendmail), it has helped significantly to slow down --although not stop-- the problem.
Many sites decided to fine-tune their SMTP servers to disallow relaying, but many did not do so. The problem remained the same as it was: a spammer knew he would not be able to send unwanted mail using a given server, but thousands of open servers remained all over the Internet. This led to a second move: The RBL -- Real-time Blackhole List. This is a list of known sites that do not have anti-relaying rules installed, and since these sites are a favorite target for spammers, they are considered untrustworthy. Not trusting a site means discarding all the mail coming from it. This is not necessarily the best way out of the problem, because if the offending site has a large audience of users (such as the very well known case of the largest ISP in Mexico: Telmex), one will dump thousands of legitimate users' mail, only because they come from the wrong server. This is, however, one of the most popular ways of blocking unsolicited bulk mail nowadays.
Managing anti-spamming rules in sendmail, however, has proven difficult and intimidating for all but the most expert administrators. Sendmail is infamous for its difficult to understand configuration file, and rules are probably the most obscure part of this file. Using a wrapper, this can become an easy task -- thanks to Perl's powerful pattern-matching capabilities, rules can be written in a format much easier to manage and understand.