Exceptionally Uncaught

So yeah, I wrote a post for the TOWER blog. From people who've read it, I've gotten comments like "what company do you work for again?", "[in a] possibly career-limiting move..." and "Simon dropped something of a megaton bomb...".

The post grew out of a conversation I was having with various people at work where I asked that exact question. And I still haven't heard a good answer.

There's the old compliance cherry - "it's the only way you can prove to a judge that this is all the relevant records". Except it's not - you can equally prove to a judge that your search algorithm will find all the records. And either way you'd still have the problem of proving that data hadn't been removed before you did your search.

Gordon talks about how it's always going to be faster to find something that's been properly filed. But we're talking degrees here. What do I care if it takes 10 times as long to find something in a mess then if it had been correctly filed, if we're talking about seconds?

Dean argues that letting Google keep track of your mess is really just getting a smart computer to organise your data. So, what do I care how it finds my stuff quickly? Would you honestly keep up with your personally filing if you knew a magic gnome would come along every so often and do it for you?

Gordon's points about context and control are interesting, but I still want better answers! I tried asking Google, but the top hit was my re-post of the original question...