You're doing the exact thing Steve Yegge and Ola Bini warn against. You're assuming that you need a "massive technology infrastructure" and then deciding how to

You're doing the exact thing Steve Yegge and Ola Bini warn against. You're assuming that you need a "massive technology infrastructure" and then deciding how to go about building one. Well of course it's going to be massive; you're planning to make it massive. Let's try another thought experiment. Assume you had to run Amazon.com but were only allowed to write 50k lines of code. What would you do? What percent of Amazon's problems could you solve with this code?