T
thereiwas
Guest
At first I was upset by Jim's remark about alt.space but then I realized that I felt exactly the same way about newcomers in my own industry. I have been in the back-room IT business for 35 years and in that time I have learned, sometimes the hard way, how to prevent things going wrong. I have also learned how to do an accurate schedule.<br /><br />The younger people in my industry, a lot of them from India and China, are not stupid. But they lack this experience and it shows in their code ("Where are the comments? Where is the error-handling?") and in their missed schedules. Even when there are some "old hands" in a project, if upper management wouldn't know a GANNT chart if it fell on them it is difficult to get them to agree to reasonable schedules. I doubt this problem is unique to the software industry.<br /><br />One of the big rules of scheduling, which has been mentioned here before, is that you can't accurately estimate work you have never done before. The way I dealt with that when working at the bleeding edge was to insist that I have the time to build a working breadboard/demo/throw-away version first, with emphasis on careful research in the unknown areas. On completion of the breadboard I could demo it to management and give a reasonable estimate of how long it would take to build a product-quality version of the same thing. The breadboard might use completely different development tools and methods than the final product, but the algorithms were all the same. I had enough success with this technique that management learned to allow me my breadboard time.<br /><br />I see SpaceX as doing the same thing with Falcon-1. It is their learning vehicle and test platform for new ideas, which will be scaled up in F9. And SpaceX has some "old hands" working there. I hope they are listened to by management.