Parsing Text With Markup Into Nodes

I'm trying to figure out how I can take user input from a textarea and insert it into a table as HTML. An example:

<script type="text/javascript">
<!--
function echo_input() {
var input = document.getElementById("user_input").value;
var cell = document.getElementById("display_area");
cell.childNodes[0].nodeValue = input;
}
//-->
</script>

<form>
<textarea id="user_input" name="user_input" onkeyup="echo_input();"></textarea>
</form>

<table>
<tr>
<td id="display_area">test</td>
</tr>
</table>

The childNodes[0] that the script keeps editing is the text node that starts out as "test". The problem is that I want the text that I put in the table to be parsed if any markup is in it (so that a <br /> will become a break, and not literally printed out as "<br />"), something that the DOM automatically escapes. It seems a very cumbersome means of doing this is to write a javascript parser, which would go through the text and create different types of nodes as it stumbles upon markup.