You can get a raw text version of a page using an URL like that one:
http://larpwiki.org/cgi-bin/wiki.pl?action=browse&raw=1&id=BackUp

I do nightly (well, for every night I leave my box running) backups with the following perl script. Use at own risk. 

Tilmann is probably doing some server side backups as well.

--RalfHuels, 2004-04-26

 #!/usr/bin/perl -w
 # Warranty void
 # RalfHuels, 25.4.2004
 
 chdir '/home/spambin/wiki.org'; 
 
 $seiten=`wget -q -O - http://larpwiki.org/cgi-bin/wiki.pl?action=index`;
 
 $seiten=~s/^.*\/h2><p>//s;
 $seiten=~s/<\/p><div.*$//s;
 $seiten=~s/<br \/>/\n/sg;
 
 @seiten=split /\n/,$seiten;
 
 foreach $line (@seiten){
  $line =~ s/^<a .*>(.*)<\/a>/$1/;
  $line =~ s/ /_/g;
  print "$line\n";
  `wget -O "$line.txt" "http://larpwiki.org/cgi-bin/wiki.pl?action=browse&raw=1&id=$line"`;
   sleep 10;
 }

 `tar jcv --remove-files -f wiki\`date +%F\`.tar.bz2 *.txt`