Platonides schrieb:
So you want to mess even more the parser, isn't? Those clouds are very nice. I think instead of asking the parser to do the whole job and inserting JS, it would better be added by a JS (will only show to JS browsers anyway). It is even possible now... fulling the Monobook.js of almost-unused code. What about doing a MediaWiki:articlejs/<PAGENAME> include (if exists) for each article?
Well my graphic makes most sense if it's inside the article; silhouette animations on the other hand might be better in a popup-window when you click its "thumbnail" graphic. Especially when the animation is simply much larger than a usual picture.
But again on the other hand you might also make such interactive graphics nearly as broad as the article text, i.e. they are placed in the middle and dividing the text. The disadvantage of them is that they fairly increase the size of the HTML code.
Of course it would be also somehow nice if you might be able to store interactive graphics as one file in the Wikimedia Commons as if they were Flash animations; on the other hand in most cases you will have to localize them, therefore you'd get a new file for each language Wikipedia is available which is not very sensible-like, because in many cases an interactive graphic would be used only once from a single article...