* Note that on font sizes 12 or less, zoom levels -3 and onward are essentially useless for editing and the text is simply not being rendered correctly inside paragraph as opposed to how it is being rendered inside SVG. This is the best I could do given the current paragraph textarea implementation. LATER: we may want to consider doing the entire thing in SVG, as that would then make all this significantly easier to match (albeit also more difficult, as we would need to simulate caret and all the other text formatting stuff--perhaps having an invisible textarea that is mapped onto an svg text may be a way to go?)