Several days have passed since the first case of wiki spam and now we have a second case on the same wiki.
This is a Chinese spammer, as evidenced by the IP address (222.95.27.208). Now you wonder why administrators find it easier to "block China" than deal with this stuff.
This is the spam:
http://sandbox.merjis.com/index/diff?version=443
The spammer arrives through a Google search for "wiki sandbox". At the time that the spammer did the search, I was around #260, so obviously the spammer is going through the results, page after page after page, and spamming on all of those links.
The spammer is using a web browser (MSIE 6) again, just like the first spammer. So it looks like they're doing it by hand, again. Unlike the last one, this one didn't clean up their spam after them, but just left it there on the front page.
The fact that this isn't a script actually appears to make this more difficult to deal with. Captchas won't work because we're actually dealing with a human, albeit a human very low on the evolutionary scale.
The spammer didn't actually work out how to make a link, using the [link] form which doesn't work on this wiki. This indicates that the person responsible was working from a script, because they didn't even both to look at the machine-generated preview to check that a link appeared. In other words, like the first spammer, the second spammer is also stupid.
This sort of wiki spam appears very hard to deal with. I don't have any ideas at the moment, but I'll give it some thought over my holiday. When you have humans working in some sweatshop in China, spamming wikis, not many of the automated responses will work.
For now, I get an email when I see spam, and then it's a race to fix it. This wiki is designed not to give page rank to links on old versions of pages, so the spammer gets no benefit from their behaviour. It's just an annoyance for me to clean up.
http://sandbox.merjis.com/index/diff?version=703
Same Chinese spammer. Same manual method of alteration.
I've blocked China now.