Jeff Atwood discusses in his latest blog post the article written by Tom DeMarco, one of the most respected people in software engineering arena. It is a short article and I recommend reading it. I especially like the following paragraphs:

….This  leads us  to  the odd conclusion  that  strict  control  is  something  that matters  a  lot  on  relatively  useless  projects  and much  less  on  useful  projects.  It  suggests  that  the more you focus on control, the more  likely you’re working  on  a  project  that’s  striving  to  deliver something of relatively minor value.

and

So, how do you manage a project without  controlling  it? Well, you manage  the people  and  control  the  time  and money. You say to your team leads, for example, “I have a finish date in mind, and I’m not even  going  to  share  it with  you. When  I come  in  one  day  and  tell  you  the  project will end  in one week, you have  to be ready  to  package  up  and  deliver  what you’ve got as  the fnal product. Your  job is  to  go  about  the  project  incrementally, adding pieces to the whole in the order of their relative value, and doing integration and  documentation  and  acceptance  testing incrementally as you go.”