Unclear management of the resource

This internal data can be modified by Javascript if needed, separate from modifying the DOM

There could be an entire subset of stuff inside <head> element to be changed: namely, <title> element for the page (it is important for bookmarking and interface), metadata (e.g. related resources, tags, description, and all that meta information used by native applications - see wiki.whatwg.org/wiki/MetaExtensions and microformats.org/wiki/existing-rel-values). Your model requires all these aspects to be updated by Javascript. I agree that current JS framework don't do that, but as a matter of fact a native application should be able to do it. It shouldn't be that difficult using model attributes - but you have to suggest it.

you could export this data for form processing if you wish.

Please make this clearer.

This internal data can also be push updated by a server or connected directly to a local database for caching or persistence - the browser manages this now, instead of the web developer/Javascript.

These are two very different cases that should be taken into account separately.

push update relies upon server/client data exchange. Consider elaborating a user interface to delay/pause push updates, or you'll end up with the same criticism moved against <meta refresh>
data cache could be somehow useful. As a matter of fact you are suggesting that only data, and not the whole "resource"* should be cached. It represents economy in terms of used memory, but it requires page reparsing every time a resource URI is invoked by the user agent.

*let's call "resource" a template structure filled in with non-initial fetched data. This will allow us to distinguish it from "page", which is the object initially sent to the client.

mozumder / HTML6

Unclear management of the resource #10