Give it a twiki page, it'll screenscrape for the twiki source and then will apply a few simple regexps to get it into moderately tolerable mediawiki markup, which you can then cut and paste into mediawiki.