Stackoverflow, analysis of users, "question difficulty", 12 languages

I think that the premise that you can judge a language's difficulty based upon the ratio of unanswered question on StackOverflow is fundamentally flawed.

Intuitively we know that C is not a simple language to master and a quick check on unanswered questions from (say) JavaScript indicates that the great majority are about DOM manipulation and not the language itself. Thus I would suggest that an analysis that rated JavaScript as more difficult than C is using the wrong metrics.

I would expect that languages with relatively low usage but also relatively expert proponents (with some enthusiasm for the topic - Scala??) were more likely to be of an interesting and technical nature and get a prompt and effective response.

In other, perhaps high volume, areas we might expect a proportion of questions to remain unanswered if they are uninteresting or frequently asked.

[EDIT] Perfect example popped up on HN a few minutes later see http://stackoverflow.com/questions/7828865/javascript-matrix... . This is a UI problem - how to restrict user entry to a valid matrix and to only process the input for valid entries. This is not by any real stretch a JavaScript problem.

Given the nature of the question, then a full answer would be lengthy to prepare (I would use jQuery and some client side code to manage the data entry area) and it might thus go unanswered.