Hi Daniel,
In the older versions of the devices, it is difficult sometimes to get
content that is locked away in CRM/CMS/Wiki/Help/Bug systems due to
the complexities relating to the way this information has to be
viewed. This is generally caused by minor workarounds but sometimes
major changes to the way the devices crawl is required.
Javascript, AJAX, ActiveX and Flash are very difficult to crawl
through, so the devices generally don't. As a result, some systems
cannot be crawled at all. To combat this, the connector frameworks
have been developed, a number of 'proxy' style systems were also set
up. Basically how this works, is that it abstracts away the 'crawl'
from standard HTTP requests and builds very specific methods of
extracting the data from these systems.
That way, all your previously unreachable content (which can't be
indexed) can now be indexed if the framework / system that you are
using is supported! It makes obtaining content and searching content
even easier and much more powerful.
Not to mention the fact that One Box modules can be implemented in
order to further the flexibility offered by these systems.
Regards,
Steve.