Update documentation

It's no longer true that all joins must be pre-declared.  You can use the
dot-syntax for ad-hoc joins

gmt-web/content/documentation.html

@@ -120,15 +120,14 @@ <h2><a name="benefits">Benefits</a></h2>
 <ul>
   <li>database queries don't repeat themselves again and again</li>
   <li>you never write insert/update/delete statements, or work out constraint order yourself</li>
-  <li>allows you to write methods which address an object individually, with some ways to avoid tons of individual database queries</li>
+  <li>allows you to write methods which address an object individually, with ways to avoid many individual database queries</li>
   <li>explicitly clearing the cache is less complex than explicitly managing the caching of data</li>
 </ul>
 
 <h2><a name="issues">Issues</a></h2>
 <ul>
   <li>the cache grows until you explicitly clear it, or allow the Context to prune the cache by setting object count limits explicitly</li>
   <li>there is CPU overhead checking the cache if you really are always going directly to the database</li>
-  <li>Joins must be pre-declared through the class definition system</li>
 </ul>
 
 <hr />

lib/UR/Manual/Overview.pod

@@ -174,8 +174,8 @@ order yourself
 
 =item *
 
-allows you to write methods which address an object individually, with some
-ways to avoid tons of individual database queries
+allows you to write methods which address an object individually, with
+ways to avoid many individual database queries
 
 =item *
 
@@ -198,10 +198,6 @@ the cache by setting object count limits explicitly
 there is CPU overhead checking the cache if you really are always going
 directly to the database
 
-=item *
-
-Joins must be pre-declared through the class definition system
-
 =back
 
 =head1 Class Definitions