On the topic of COBOL/DB2 ...
It has been my experience that 2 things can blow up a major systems development project: 1) a solution that won't work no matter how much money you throw at it and 2) poor execution.
IMHO, CKFR eliminated risk 1 with their selection of COBOL/DB2. When you get down to it, CKFR's bread and butter isn't some slick server- or web-based application with a cute front-end. This project isn't sexy, it's high-volume, high-reliability, mission critical transaction processing, which means Keep It Simple on the technology. From what I have read, CKFR seems to have done that.
On risk 2, the choice of a time-tested technology should help here too. There will be plenty of tools in the market for testing, integrated project management, performance tuning, etc. Warning: This is where this post gets really boring ... I hope that CKFR employs some type of repeatable project methodology, because I have seen hundreds of millions of dollars wasted on projects that can't organize themselves out of paper bags. I'd love to see this project's scope statement, project staffing model and workplans.
KD |
