Obsessive-compulsive downloaders

In fact, I believe that the exact opposite is true: the mirror-the-universe approach is a product of the absence of all three above-mentioned properties of modern-day data. Most people mirror because they see the web (and the whole internet) as something passing, much like TV, in the sense that if they won't keep a copy of some data that they stumbled upon while browsing, it will effectively be lost to them. This is more or less the same reason why people are more willing to by into a 'time limited' or 'while stocks last' marked opportunity, even if the actual cost is the same.

Another point is the belief that local resources are easier to access (which they are) and easier to search (which they are not). Precluding fundamental breakthroughs in communications, chances are that latency and bandwidth to your own house will be much better than to cross-country destinations. On the indexing issue however, I allways long for a Google daemon that indexes the local hard drives (local meaning household-wide or LAN-wide)...

In the end, there will allways be a storage vs. bandwidth tradeoff based on a combined metric of ease-of-use and cost.