What with the .net 2.0 version of NHibernate in the works I thought it
might be interesting to point to this (insofar the developers haven't already seen this themselves)

http://msdn.microsoft.com/netframework/default.aspx?pull=/library/en-us/dndotnet/html/StringsinNET20.asp#stringsinnet20_topic9

The gist of it seems to be that for all "non-UI" strings we should be
using StringComparison.OrdinalIgnoreCase or StringComparison.Ordinal because it makes for more predictable results and much better performance.

IMHO NHibernate could benefit from this, especially performance wise.