Documentation:
<http://wiki.qcodo.com> is live, and a great place to document anything you feel is lacking. The good stuff will probably be cherry-picked, and put into the official docs.
Issue Tracking:
The bugs.qcodo.com address isn't quite live, but the issue tracker itself is accessible via <http://zcodo.com/project/issues/zcodo> (QCodo and ZCodo will share the same bug tracker according to the merge announcement). It allows reports of feature requests, bug fixes, comments on them, etc. The only thing I think it's missing is source control integration.
Mailing List:
Personally, I find the forums much more organized and easier to follow, as well as being more open. I get all the updates I need via the Email Notification and RSS features.
PHP Version:
Unless there's a major security reason to force a newer version of PHP, I'd recommend staying as low as possible.
QForms without JS:
I'd love this. I suggest you post it as a feature request in the issue tracker. :)
PDO:
I'm unclear exactly what benefit this would have. Since it's not an abstraction layer, we still need each DB adapter in order to customize the behaviour of things like LIMIT clauses.
ADODB seems a much better choice, since it is an abstraction layer, and would allow us to write one ADODB adapter, and connect to multiple actual DB servers without changing code.
FrontController:
I've always been unclear on the benefits of a front controller. I have a master class my pages all extend, giving me the same functionality. So I'm not sure this is something that needs to be in the core. Having tutorials on how to set up sites this way is definitely good to have though.
Permission Framework:
The most up to date discussion about this is here:
<http://www.qcodo.com/forums/topic.php/3224/1/>
Note that it sounds like the versatile framework I am hoping to see for QCodo in general does not sound like what will be adopted for Erick's CMS, so please don't confuse the two.
Localization:
It's worth noting that the existing .po based localization is not adequate for all but the most basic systems, since there is no translations for database content. Ideally we would define a framework including DB tables that allow for full translations throughout an application.