My first grease monkey script ^{[citation needed]}

Here's the source of the script if you are interested:

// ==UserScript==
// @name            No Wikipedia Citation Needed
// @namespace       http://marcolson.net
// @description     Gets rid of [citation needed] in wikipedia
// @include         http://*.wikipedia.org/wiki/*
// ==/UserScript==

var anodes, node, supnode, parentnode;

anodes = document.evaluate("//a[@title='Wikipedia:Citing sources']", document, null, XPathResult.UNORDERED_NODE_SNAPSHOT_TYPE, null);

for (var i = 0; i < anodes.snapshotLength; i++) {

    node = anodes.snapshotItem(i);

    supnode = node.parentNode;

    parentnode = supnode.parentNode;

    parentnode.removeChild(supnode);

}

document.normalize();

Basically, we do 3 things:

1. Get a list of all of the nodes that have a citation title
   
2. Traverse that list and remove the parents (the superscript node) from the document
   
3. "Normalize" the document to combine the fragmented text nodes
   

The first step is to get a list of all of the anchor nodes that contain a "Citation Needed" tag. If we look at the source of the page, we notice that they all have a common title ("Wikipedia:Citing sources") so we look for all anchor nodes with that title with our nifty xpath expression: //a[@title='Wikipedia:Citing sources']

anodes = document.evaluate("//a[@title='Wikipedia:Citing sources']", document, null, XPathResult.UNORDERED_NODE_SNAPSHOT_TYPE, null);

The next step is to traverse that list, get the parent of the parent. The firefox dom works in such a way that to remove a node, you need to know the parent and do a removeChild() operation. We also don't want to leave the superscript node in place, so that's why we get the parent of the parent--it's cleaner this way.

for (var i = 0; i < anodes.snapshotLength; i++) {
    node = anodes.snapshotItem(i);
    supnode = node.parentNode;
    parentnode = supnode.parentNode;
    parentnode.removeChild(supnode);
}

The final step is just a bit of housekeeping. We want to clean up the dom, so we combine any fragmented text sections.

document.normalize();

And that's it! ^{[citation needed]}