[Jackrabbit Wiki] Update of "QuestionsAndAnswers" by EricSolberg

[Apache Wiki <wi...@apache.org>]

Dear Wiki user,

You have subscribed to a wiki page or wiki category on "Jackrabbit Wiki" for change notification.

The following page has been changed by EricSolberg:
http://wiki.apache.org/jackrabbit/QuestionsAndAnswers

------------------------------------------------------------------------------
  
  
  ----
+ 
+ === Why no node SPI? ===
+ 
+ Question:
+ I read in the FAQ that the interface was not intended to provide a SPI for attaching to other storage.  Specifically, it says that you could probably achieve this for a read-only link to an existing relational database table, but it says you shouldn't/couldn't do it for read/write relational data.
+ 
+ I am curious why this closed architecture?  I am considering this for a major project bt need the ability to map to relational schemas and other existing content stores, including other document management systems, and could even envision using that mechanism to federate multiple Jackrabit repositories into a central repository.  I thought I might just need to extend VersionImpl or one of the other Node types and override some of the methods.
+ 
+ I would even envision implementing node types that map to transient runtime data such as performance counters etc.
+ 
+ Can you elaborate as to what architectural/design issues make this not possible.  Or, is it just a philosophical point of view or ?  This is one of the powerful capabilities that Slide offers.  I understand that putting this abstraction over the RDBMS would have performance issues, but assuming my use case isn't sensitive to that, what are the reasons it won't work?
+ 
+ Thank you very much.
+ 
  
  === Where can I find examples of using the JCR API? ===