yes, and also it makes easy to manipulate the dump using Hadoop, and retrieving some values for all the article using jq http://stedolan.github.io/jq/, could you do the same with XML? Btw You can find a little sample of the dump here: https://dl.dropboxusercontent.com/u/4663256/tmp/json-wikiped...
XML? No clue how to manipulate it properly. I could figure it out of course, but now I don't have to. JSON + Python just works™.