> but otherwise focus more on git’s behaviour when you use it and why sometimes git behaves in unexpected ways
This is my criticism of the approach that the investigation in this series of posts has taken; it would probably better be called "Git from the outside in", because that's what it is—it concerns itself mostly with the shape of Git commands, how things are expressed in the UI, and a some persnickety not-alrogether-very-important implementation details about how things work in the current release (that could change at some point and very well might change) and not so much "how Git works".
Previously:
> The biggest problem with people's attempts to wrap their heads around Git, its awful UI notwithstanding, seems to be an insistence on thinking of it foremost in terms of the experience of running any given command. I.e., trying to understand it from the outside in.
(NB: this is not just a facile criticism of the title; I believe this series resembles how a lot of people actually try to work out an understanding of Git, which is a problem, because it's fundamentally the wrong way to approach it and indeed is the mindset that's responsible for so many people left still confused.)
I fully agree, the problem is that people approach git the same way they approach all other software tools, which is the wrong approach obviously and people are wrong for doing it. It’s the people who are wrong and not the tool. It’s like people complaining about the new version of Windows, I always tell them “have you even read the series of articles about the theoretical structure of the underlying OS? Before you do that, how can you even complain the user interface is obtuse!?”
This is my criticism of the approach that the investigation in this series of posts has taken; it would probably better be called "Git from the outside in", because that's what it is—it concerns itself mostly with the shape of Git commands, how things are expressed in the UI, and a some persnickety not-alrogether-very-important implementation details about how things work in the current release (that could change at some point and very well might change) and not so much "how Git works".
Previously:
> The biggest problem with people's attempts to wrap their heads around Git, its awful UI notwithstanding, seems to be an insistence on thinking of it foremost in terms of the experience of running any given command. I.e., trying to understand it from the outside in.
<https://news.ycombinator.com/item?id=39791274>
(NB: this is not just a facile criticism of the title; I believe this series resembles how a lot of people actually try to work out an understanding of Git, which is a problem, because it's fundamentally the wrong way to approach it and indeed is the mindset that's responsible for so many people left still confused.)