Comments on: Breaking rocks http://lostconsciousness.wordpress.com/2008/10/12/breaking-rocks/ Random thoughts on technology, machines, science and people Wed, 28 Oct 2009 00:01:36 +0000 http://wordpress.com/ hourly 1 By: Jeremy http://lostconsciousness.wordpress.com/2008/10/12/breaking-rocks/#comment-99 Jeremy Sun, 12 Oct 2008 22:35:35 +0000 http://lostconsciousness.wordpress.com/?p=80#comment-99 Phew - well that's a cheery start to the week! Phew – well that’s a cheery start to the week!

]]> By: Steph Gray http://lostconsciousness.wordpress.com/2008/10/12/breaking-rocks/#comment-98 Steph Gray Sun, 12 Oct 2008 19:50:43 +0000 http://lostconsciousness.wordpress.com/?p=80#comment-98 Of course, if you just have one rock, that's one rock to store, clean, insure, keep updated with WinRock(TM) software, protect from theft etc etc, so the risks are still pretty significant. In theory, much better to work out what you want the rock for, and get someone to lease you those rock characteristics as and when you need them. In principle, outsourcing is the optimal approach, just like in principle it's possible to manage by incentives and targets. It's just that I've hardly ever seen either of them work optimally, and outsourcing is particularly, spectacularly unsuccessful in the case of IT (based on my simple end-user perspective). Given the fuzzy world we live in (the complexity you describe), our innate inability to predict or describe future needs properly, human nature and the imperative to creatively exploit incentives, outsourcing is almost always riskier and probably costlier than insourcing. I enjoyed Joel Spolsky's latest column in Inc.com on the gaming of incentive schemes by employees - I think the same issues apply to outsourcing by and large: <a href="http://www.inc.com/magazine/20081001/how-hard-could-it-be-sins-of-commissions.html" rel="nofollow">http://www.inc.com/magazine/20081001/how-hard-could-it-be-sins-of-commissions.html</a> Of course, if you just have one rock, that’s one rock to store, clean, insure, keep updated with WinRock(TM) software, protect from theft etc etc, so the risks are still pretty significant. In theory, much better to work out what you want the rock for, and get someone to lease you those rock characteristics as and when you need them.

In principle, outsourcing is the optimal approach, just like in principle it’s possible to manage by incentives and targets. It’s just that I’ve hardly ever seen either of them work optimally, and outsourcing is particularly, spectacularly unsuccessful in the case of IT (based on my simple end-user perspective). Given the fuzzy world we live in (the complexity you describe), our innate inability to predict or describe future needs properly, human nature and the imperative to creatively exploit incentives, outsourcing is almost always riskier and probably costlier than insourcing.

I enjoyed Joel Spolsky’s latest column in Inc.com on the gaming of incentive schemes by employees – I think the same issues apply to outsourcing by and large:

http://www.inc.com/magazine/20081001/how-hard-could-it-be-sins-of-commissions.html

]]>