http://ayende.comAyende @ RahienCopyright (C) Ayende Rahien 2004 - 2021 (c) 202660"The product sometimes gives an error, for no deterministic reason" It will be a deterministic reason - you just haven't found it. Possibly that's what he meant by suggesting it be taken off the list: Wait until you know the exact cause, until then you can't even be sure it is a bug. That's not to say I'm not on your side here :-) A product that gives very unexpected behavior to a top notch dev such as yourself needs more work, for sure. There's a limit to how much energy you should have to expend to figure out why the product is behaving, to you, in a non-predictable way. At some point, the behaviors are /effectively/ random and non-deterministic, because they are beyond the point where only /reasonable/ energy is required to be able to discover the causes that make them predictable. My number one coding motto is "follow the rule of least surprise". They obviously haven't, which means your critique seems entirely valid. /Mats http://ayende.com/2661/re-ssis-15-faults-rebuttal#comment2http://ayende.com/2661/re-ssis-15-faults-rebuttal#comment2Mon, 30 Jul 2007 08:06:36 GMTSteve your article is right on, not just SSIS, but SQL 2005 has complicated too many routine tasks that were striaghtforward in sql 2000, Hopefully sql 2008 will fix them. If you are going to be at DevConnection in Las Vegas will be pleased to bu you a drink http://ayende.com/2661/re-ssis-15-faults-rebuttal#comment1http://ayende.com/2661/re-ssis-15-faults-rebuttal#comment1Sun, 29 Jul 2007 13:12:35 GMT