improve language on Guard as proposed by David Chadwick's point 5 from feedback http://lists.w3.org/Archives/Public/public-xg-webid/2011Dec/0140.html
--- a/spec/index-respec.html Mon Jan 09 14:06:18 2012 +0100
+++ b/spec/index-respec.html Mon Jan 09 15:09:26 2012 +0100
@@ -415,9 +415,9 @@
</dd>
<dt><tdef>Guard</tdef><dt>
-<dd>A guard is an agent, usually on the <tref>Server</tref> that can look at a request from the <tref>Client</tref> and decide if it needs Authentication by looking at the Access control Rules.
-If it needs Authentication it can request it, and it can use the <tref>WebID Verifier</tref> to complete identity checks.
-Finally it can grant or deny access.
+<dd>A Guard is an agent, usually on the <tref>Server</tref> that can look at a request from the <tref>Client</tref> and decide if it needs to be authorised by looking at the access control rules for that resource.
+If the request requires authorisation, the Guard can first demand authentication of the <tref>Client</tref> and use the <tref>WebID Verifier</tref> to check any claimed identies that would allow it to come to an authorization decision.
+Finally the Guard can grant or deny access according to how the verfied identities satisfy the Access Control Rules.
</dd>
<dt><tdef>Verification Agent</tdef> or <tdef>WebID Verifier</tdef></dt>