Unless the blog author builds software that never crashes -- even when it get millions of hits on its first day, then he should cut them some slack. I'll put money on the fact that Google didn't have the same kind of traffic on their first day that Cuil did. And Google has had 10 years to get it write. Yet he expects perfection on the first day.
I don't think it is as much of a matter of writing software that never crashes, but rather a matter of protecting your site from crashing on the launch.
I'm interested in their reasoning for going with a "grand opening" and subjecting the site being hammered instead of inviting groups of early adopters/beta testers and incrementally stress testing the system.