Well, I wasn't talking about the insecticide.. 
In most SOHO scenarios (where a DNS-323 is most likely to be used), the kind of constant availability that RAID was designed for is simply unnecessary. Secondly, recovering from some failure of the array (even the simplest) is often not as easy as you might think.
SOHO - Small Office Home Office - I'm self employed and I run a two man operation from what was once my garage - it doesn't get any more SOHO than that.
We work within the confines of an SLA (service level agreement) and if we don't deliver it costs us money at the end of the month, worse yet, failing to meet SLA also raises the possibility of loosing contracts - I run RAID on my main server and the data is backed up to my DNS-323 (which also runs RAID) and if my primary server fails I can run directly from the DNS-323, until such time as I can get the main server back online.
I would suggest that any SOHO scenario where RAID is not used is simply because the individual is unaware of the advantages to be had - small businesses need redundancy just as much as large ones do.
As for recovering from even the simplest of array failures, I would assume that to be a disk failure - in the case of a DNS-323 you determine which disk has failed (either from the front panel LEDS, the status email that the unit sends or the status screen), power the unit off, remove the failed disk and replace it with a new one, power the unit on, log into the web admin page and follow the prompts - is really IS that simple, and I have had a disk fail, so I speak from experience.