It should be possible to use the Echo format directly for creating archives, when used with full-content posts. Echo format archives would also make an excellent means of blog BackUp.
Prior art :
Using RSS 1.0/RDF the blog content can be stored as <content:encoded>, and all the existing resources (title, data etc) maintained directly.
The RSS 2.0 approach appears to be to strip away <title> etc. and put the full content in the <description> element - the <guid> appears to have been replaced by a Radio-specific (namespace qualified) element <id>.
This approach to archiving allows "Blog Browsers" to be built - use of a consistent language would mean that all newsreaders in effect become "Blog Browsers".
For Echo there may be searching/indexing advantages in using Echo-specific language constraints rather than just a generic XML database, though this might cause problems with elements from other namespaces (unless this can be covered by the SyntaxExtensionMechanism).
Using the RDF model mapping (using the RdfOwlVocabulary), the data could be maintained in an RDF data store, though again there may be problems with elements from other namespaces without mappings (again unless this can be covered by the SyntaxExtensionMechanism).