zlacker

I feel like I've spent half by career telling people this.

My general recommendation is to engineer for only 1 order of magnitude greater than you need today, but even that stops applying eventually.

>>simonj+(OP)
The response is easy "find me an otherwise successful company that actually collapsed from failing to scale things like a database and we'll use it as a case study."

Security and PI stuff is unfortunately something you do have to be super anal about these days but scalability? no.