I ummed and erred about posting this, as I wouldn't (and don't) want to detract anyone from reaching out to Lifeline. This is strictly an n=1 sample.
I disturbingly had a less than great experience with Lifeline the one time I needed some help. Specifically the operator told me she didn't understand my issues or why I was calling. The fact that I had expectations of being able to access an understanding counsellor and had saved them as a last point-of-call when other options were exhausted made this particularly unhelpful. Thought I'd post this as potentially my own expectations made this situation worse (and if I'd read this or other online accounts I might have had lower expectations).
Still, I got through that so maybe it was the right thing to tell me.
> People don't really read things online, they scan. some things jump out, and only if you're very intrigued will you actually read the individual words. In fact, very few of you will even get to this line that I'm writing. The chances that you'll read each line decreases exponentially.
I only read that first sentence of that paragraph.
> Note that I've made all of my main points in the first sentences of what I've written. Did you even read everything, or did you just scan through and pick up on a few words? The attention paid to each line decreases exponentially as you continue.
I went back and read the first sentence multiple times and didn't see a point. There's a principle at work? That's the point? If it was then I'm disappointed. I thought there might be some kind of mention of what it actually was.
"Strong opinions, weakly held" sounds right to me.
I couldn't care less how confidently somebody holds an opinion.
I think the word "strength" here should signify a battle-hardened opinion that has won out over others in a practical or theoretical sense. Strongly-supported...?
That's not to say that that it's wrong to be confident in the opinions you hold, but if you want to have strongly-supported opinions you can't protect them from being tested against other opinions. Let them win on their own merit...
I'm not seeing much of an argument against factory_girl (in unit tests) other than: it's not as fast as setting up your model by hand. And no description of the alternative?
The alternative is to stop building so many dependencies between your models so you don't need a tool like factory_girl at all! No fixtures, no factories, make only and exactly what you need for each test. If that test setup is getting untenable, that's test pain! Yay! Now refactor your code and fix the pain properly by reducing coupling.
The problem with factory_girl is that it lets you hide this pain instead of fix it, letting you get on with building massively coupled systems that turn into maintenance nightmares.
As a side note your unit tests should never hit the database or any external service. Your unit test suite should finish in at most a few seconds. If it's slower than that you have problems.
Good question. The thing I most loved was the progressive refinement of the technique. I love seeing things transformed from rough to fine and seeing the rationale for each step.
