The financial industry has some reliability requirements that are daunting. The stock market doesn't stop and wait for you if your servers can't keep up or go down - your customers expect your service to be up and running. Disaster planning is critical - that's one of the lessons of 9/11. I worked for a place where the disaster recovery plan was a manual failover that was allowed to take up to 12 hours. I am now aware of a commercially available technology which will do data center failover in 5 minutes, completely automatically. What's more, it will spread the load between data centers, so that the resources in a data center are not idle. I like the idea of running a service 24/7. I grew up in the aerospace business where reliability was a religion (if you are in an airplane 39,000 feet over the Pacific Ocean and your options are Seattle, Anchorage, Honolulu, or Tokyo, and your computer says that something is wrong, abort, retry, or fail - that won't give you a warm fuzzy feeling). While it isn't cost effective to achieve aviation standards of reliability, it is cost effective to get to a higher level of reliability than most installations have. Morgan Stanley ought to understand that reliability requirement and fund it. I want to be a part of that effort.