I can't speak to what others will find interesting. It certainly
sounds interesting to me, if only for the dogfooding aspect.
The biggest issue standing in the way of introducing a change like
this into the core will be backwards compatibility. There is plenty of
code out there using the existing syndication framework, and we've
only just recently completed a deprecation cycle around changes to the
syndication framework. This is probably the reason that you can't find
many tickets against the syndication framework at the moment; that
rewrite addressed most of the big problems with syndication at the
time. Unfortunately, that rewrite happened just before CBVs were
introduced.
However, that shouldn't stop you from looking at this project as a
standalone project. Worst case, you can always keep it as an external
app. However, once you've got some working code, we'll have a better
idea of how complex a migration task would be involved, and whether a
seamless migration to a dogfooded syndication framework is even
possible.
Yours,
Russ Magee %-)