It’s just data
Whenever a site pings weblogs.com (the old way, with just the URL of an HTML home page), some backend script downloads the home page to verify that it has actually changed. As part of this process, couldn't that backend script use RSS autodiscovery to find the URL of the associated RSS feed? That information is there (in many cases), and it's just being ignored at the moment. This would require zero additional network traffic.
Posted by Mark atWhile were at it we could developed shared processing. That way when one server died perhaps the service itself might not.
The real obstacle is no one wants to devote time and energy (well maybe a couple people) to something like this when a) the services already exist in an embedded form and b) there is no revenue stream.
Its too small of an area to see a real strong group of people devote themselves to working on it for free in meaningful sense.
I think.