Is it possible that Sam Collins failed at failure? What if he and his team could have re-purposed their modular utilities/design for HTML5 support? There's still a great need for very high level utilities to aid domain experts in building specialized web services.