I'm not as up-to-date as I need to be, but isn't the idea of truly semantic markup and RDF and microformats that Instapaper (and many, many other sites) won't have to have site-specific parsers? Now, I recognize that a) Instapaper can't do anything about that, and b) the vast, vast majority of sites are unlikely to change. So I guess the main question is: if you have really good semantic markup, do you need this, or will Instapaper already handle your stuff correctly?
You can always Instapaper your own site to check but, no, you most likely won't need it. There will always be edge cases and poorly coded/broken markup on many sites though. That's what this is for.
He should add support for the <article> tag (if he hasn't already).
Yay, the instapaper future is full of ads.
They’ll go through admin approval before going live, to prevent abuse, and then they’ll improve the text parser for everyone.
"If the instapaper_body class name is present, the automatic selection and automatic stripping processes are disabled, leaving full control to the site’s author."