Plan B

The [911] dispatcher stays on the phone with the woman for 10 minutes and 21 seconds. She tells the caller to try to hide in the house.
[...]
And four times in total, she says there isn't anyone who can help.
[...]
According to police records, a few minutes later, the woman's ex-boyfriend [...] used a piece of metal to pry open her front door. He then attacked her. Eventually, state police arrested him, and he pleaded guilty to sexual assault and sodomy, among other charges.
Loss Of Timber Payments Cuts Deep In Oregon

While it's become popular in certain Republican and conservative circles to complain about "dependency" on government as a way of attacking Reaganesque "Welfare Queens" (and, as I see it, primarily Democratic/liberal constituencies), it seems to me that the real danger of being dependent on government is in stories like this. The unnamed woman in this story appears to have no other recourse than to call 911, and when they cannot help her, neither she nor the dispatcher can come up with any other workable alternatives.

One of the things that we often lose sight of is that the very real efficiencies that come with division of labor on a large scale comes with an accompanying cost - our specializations rob us of the ability to perform other tasks well, or maybe at all. I would be a terrible farmer, in no small part because never having needed to grow food or care for animals, it's simply not a skill that I possess. Whether I recognize it or not, there is a risk there. Therefore, it behooves me to be aware of that risk, even if I don't take any action to address it directly. If I don't have a fallback plan, it should be due to the (reasonably) informed decision that I don't need one - not because it never occurs to me to create one.

Because nothing is perfect, or even perfectible, not even government. And not just government. The fact that for the first three months that I owned it, my cellular phone constantly displayed a harmless, yet annoying, error message taught me that I should never trust it with my life. Not because the technology was unproven, but because so much of the infrastructure that supported it was in the hands of people who couldn't manage to convince the phone that it was, in fact, supposed to be doing exactly what it was doing. Expecting it to work was fine, but I needed to know what I would do if it didn't.

This is inefficient. And over a wide enough scale, impossible. Sometimes you're stuck with relying on an outside agency. But the flip side of efficiency is resiliency, and so perhaps we should start to reconsider the impulse to accept the greater inefficiency that bringing some tasks back "in-house" will bring, so that when Plan A, whether it's government or any other institution fails, there's a Plan B to back it up.