Use Redundancy Wisely

Backups are often neglected because they don't get used enough. If you find a frequent use for a backup, then you'll frequently be making sure the backup is effective. Distribution of data is a fantastic example. If your data distribution strategy pulls the data from a backup, then you will very quickly learn about any failures of the backup. Don't be tempted to cut the redundant middleman (the backup) out just because it's flimsy. That's like saving money by buying cheap food. It will eventually catch up with you in the form of health issues (or, data loss).

There are other areas where redundancy is a good idea and not used in a way that virtually guarantees its viability. It takes some consideration to use it wisely. Let's take canned goods. Having a lot of cans of food is good protection from natural disasters that could disrupt the flow of groceries to the stores and restaurants around us. However, if those cans get damaged or lost, they won't do much good. One solution is to eat canned food every now and then, and use the oldest cans you have. 

The are racks that are slightly tilted so that placing cans on them sideways allows the cans to roll to one side. You can buy new cans to put in on the highest side, and take cans to use from the lowest side. But what if you only want canned food in emergencies? I have an idea to handle that situation too.

A periodic disruption of a regular pattern is very healthy. That's the idea behind the Sabbath, and other forms of a "day of rest," even including vacations. A monthly "canned food day" might be a good idea.  It would provide that periodic disruption, and also give you a chance to ensure that your "backup food plan" is in working condition.  If you really don't want to eat canned food, you can just go through the motions on that day. This presents the last problem I want to address here, which is the difficulty with "practice."

Our brains are like ant colonies, constantly cutting corners to find the fastest route. When "going through the motions" serves no immediate purpose, motivation to go through with it will dwindle. I can't think of an attractive solution to that right now, so I'll leave it as an exercise for you.

To wrap it all up, remember that:

1. Redundancy can provide security, so it's a good idea.
2. Redundancy is susceptible to decay to the point of being useless.
3. Incorporating redundant systems into routine and useful procedures protects them from that decay.
4. The flimsiness of something that doesn't seem that important, like a routine secondary use of redundancy, can be deceptive (See #2).

A husband and wife who are both evolutionary biologists have been interviewed together a few times and I can't remember their names.  The husband, at one point, said something I found profound, and important.  Maybe it was someone else who said it, but the gist of it is that we create a lot of misery for ourselves by failing to pay attention to the negative consequences of the changes we make to our routines.  If you've designed a routine which uses something that is redundant specifically because you wanted to make sure the redundancy is maintained, then that routine is important on a level no one will ever see except in the nightmare scenario where the routine is thrown away, the redundancy falls into disrepair, and then the process for which it is a redundancy fails.  Any routine that serves this secondary (and, I argue, more important) function of ensuring healthy redundancy, is important.  Point out that importance to people, and marvel at it.  It's the kind of thing evolution figures out WAY before we do, and, as such, I think it's one of the beautiful elements we get to discover and share.