http://beyondrelational.com/live/groups/tsqlchallenge/forum/default.aspxenCommunityServer 2008.5 SP1 (Build: 31106.3070)http://beyondrelational.com/live/groups/tsqlchallenge/forum/p/5290/10527.aspx#10527Fri, 26 Nov 2010 11:39:04 GMT6e5011fa-7db5-4df3-bb79-9085c1d333b3:10527Sergejack0http://beyondrelational.com/live/groups/tsqlchallenge/forum/p/5290/10527.aspx#10527http://beyondrelational.com/live/groups/tsqlchallenge/forum/commentrss.aspx?PostID=10527<p>Every challenge ends with a performance test which determine the winner.<br />Among the winning solutions, I often see solutions whoose readability/maintability gets really poor because of a huge usage of granular tunning to gain like 0.001% performance.</p> <p>I think many people will agree that practice is a bad one in the real world requirements.</p> <p>I suggest, that amongst the best performing solutions (for instance : the fastest one + every solutions which doesn&#39;t last more than 120% of the fastest solution duration), people* should vote for the best solution.</p> <p>*it could be a few number of qualified persons or everybody or wathever sounds good.</p><div style="clear:both;"></div>