It is now common wisdom that wiki spammers are spamming wikis because they want to influence the results that search engines will return. That's why many wiki enignes now take care about the robots meta tag. That's why many wiki admins have hand-crafted robots.txt files, etc. If spam on wikis doesn't end up in a search engine, the reason for spamming is gone and wiki spamming should come to a halt a some later point.
If you clean your wiki on a regular basis, wiki spam should have a hard time being found by a search engine spider on the current revisions of your pages. It still could be found on your kept pages, though. Even though those kept pages will eventually expire, search engine robots might find them and it will raise the position of the spammer's site in the search results for the spammer's keyword. We don't want that, but it definitely makes the spammers happy. They don't have to care whether you clean the wiki or not. Two weeks till the kept (spammed) page expires should be good enough for them.
When looking at the server logs of the chongqed.org wiki after it was spammed a couple of times, I noticed a consistent pattern used at least by the Chinese wiki spammers: They ask Google for pages containing spammy URLs. In other words, they are trying to find spammed wiki pages, so they can spam them again, this time with different URLs or keywords or just to make sure that there spam stays fresh. This means that spammy pages (even if they are old revisions) on your wiki that are indexed by Google and friends, will attract more spammers!
That's why I'm suggesting that wiki admins allow only the most current revision of any page on their wiki to be indexed by search engine spiders. Even if you don't care about the spammer's page rank, you sure don't want to attract more spammers.
The Register published a very interesting letter today: China's IT: an inside outsider's view.
It's a good and very scary read.
Not much to say here. The wire mesh spammer was back at the chongqed.org wiki and here are his keywords: