Sorry, for the past couple weeks the entire index hasn't been parsed regularly. I just checked your URLs and it's not a parse problem, but just that they're not in the set that's being parsed.
I'm working on redoing part of the pipeline to reduce the overall latency from the time something's crawled (including support for crawling things specifically for the Social Graph API) to the time it's available via the API. During that process I've scaled back which content is parsed.
Things have been going well the past few days... I expect to get back to including the full index this week, and hopefully have some new stuff available for you guys to play with diagnostic-wise to see how things are being parsed behind the scenes. (exposing the parser by itself via a new API endpoint)
- Brad