The technology that makes the Web go doesn't have any built-in notion of a
"site" or a "home page", even though that's how people think about things.
This causes all sorts of practical problems; the well-known
/robots.txt mechanism for crawler control is kind of a kludge,
and works really badly when multiple sites are on the same server, for
example members.aol.com.
Another symptom is the fact that it's kind of hard to find the RSS feed for a
web site.
Well, we may be starting to address this issue over in the TAG.
At our last telecon we accepted this issue, as raised by Tim Berners-Lee, and today I posted the beginnings of a strawman proposal for introducing the notion of a "site" to the Web architecture. My note points to Tim's original proposal, but there have been lots of these over the years, for example by Roger Costello et al.
There's lots of room for work here, but this one might end up being a really low-hanging fruit.
