I suggest folks read some good books on conversations and negotiations.
Bargaining For Advantage:
Getting To Yes:
Getting Past No (billed as a negotiations book, but really more of a conversations book):
I strongly recommend reading Influence before you read these - much of what is in the books above will make more sense once you've read Influence.
When you read these, keep in mind: Change is hard. Don't expect to read these and become good communicators quickly. It may take a few years of stumbling and practice.
I see a mixture of comments agreeing and disagreeing with the original submission. For those who disagree: Most of what the author is saying is in agreement with what the books say:
If your goal is to change someone, you will either fail, or will succeed at the cost of the relationship (and relationships at work do matter).
Another important related point: If you cannot summarize why the other person is acting this way without using phrases like "stubborn", "irrational" or similar negatives, then it means you have no idea about the other person's concerns and motives, and are being lazy. It is easier to label, and much harder to probe effectively. Additionally, people often act stubborn because they realize you are not really interested in their perspective. Internally their thought process (which is very rational) is "This person does not really want to hear me out, so I'm not going to invoke too many neurons engaging with him and will just dig in my heels." - which is why a lot of books focus a lot on listening skills (which includes skills to signal that you are listening - you may in reality be listening just fine but the other person does not know it - so you signal it by summarizing their stance).
A lot of the comments here are invoking false dichotomies. Since HN has a comment limit, I'll address some here:
>I don't believe you can have a successful software team with individuals who can't take a code review well.
This is tangential. You can give feedback in a code review poorly, or efficiently. Both ways allow for you to point out problems with the other's code. One way will not be taken well. The other way has a higher chance of being taken well. A big step forward is to realize you can have your cake and eat it too.
>I started to try and reason with people with carefully crafted questions to guide them towards my goal.
Leading questions is a bad idea (all the communications books say it). Learn how to state your concerns. It is OK to ask questions if genuinely curious. But if you want to point something out, learn how to state it in a non-defensive manner.
(3 separate comments below):
>If Kara's emotions and defensiveness can't handle a clearly articulated, rational, objective argument against design decisions, then for the sake of the product and the company, she probably needs to find another job. Avoiding discussions doesn't work for me.
>Learned to let go and he has his parts of the code base and I have mine.
>And this is how you end up with a terrible, in-cohesive product.
Again, false dichotomies. The solution is not to be quiet and let it go. The solution is to learn how to talk about the issues effectively. One of the books calls this "The Fool's Choice" - thinking that either you have to be quiet and not air your concerns (to save relationships), or that you have to air them and damage the relationship.
>It's either you convince them, or perhaps they convince you. Logic wins.
Logic alone rarely wins. One key point in one of the books: Don't pretend that emotions should not be part of the decision making process. The reality is that emotions are already part of the decision making process. If you get angry that someone cannot take your feedback well, emotions are present.
>It's safe to assume Kara wrote this article.
It is safe to assume that the author of this comment is unwilling to question his views on the topic.
That's what assumptions get you.
>I have seen more technical damage done by nice and competent people deferring to bullies in the workplace than by legitimate disagreements expressed passionately.
Another false dichotomy. What the submission describes is normal among non-bullies.
>The flaw here is that you assume that "Kara" will learn from her mistakes. Not always the case.
It is a similar flaw to assume that merely telling her what mistakes she made will make her learn from them. Definitely often not the case.