"Engineering is a team sport" - me

You need to be able to work with others to produce things, unless you always work alone. Social skills are a very important part of that.

Clearly you need both; I've also worked with really sociable and nice people who were a great laugh but couldn't program their way out of a paper bag if the survival of the human species depended on it.

If by valuable you mean they keep you employed despite not doing anything, sure. They are valuable for career progress and such, but when faced with real technical difficulties, you are not going to talk yourself out of it. Or maybe you can, delegate them to some poor bastard. What a wonderful world we are living in.

Social/engineering skill are two sides of a rectangle, and the area is what matters.

In my experience, people are more interested in being around cool people than competence. (I'm not making a value judgment here)

Are you implying that being able to have an intimate relationship with another human being disqualifies someone from being a "good software engineer"?

Some people downvoting you may be lacking skills to detect that this, even if partly true, is a joke. For me it was good.

There's a skill to being honest in a productive way. Being honest in a way that communicates your thoughts and gets people engaged rather than defensive is really difficult but really valuable. But if you don't yet have that skill being completely honest will hinder you.

Honestly? That's not how development works at all, in my experience. Best developers aren't isolated on "most interesting problems", but rather, are getting constantly sucked into war rooms and emergency calls. And when they are working on "interesting problems", they are constantly working, verbally and in writing, to convince reluctant management and other engineers to agree to their new, "interesting" approach to the problem in question. Those are massively social behaviors. "Being honest" is necessary but not sufficient. Understanding people's motivations, finding reasons for them to believe your idea is beneficial for them and their own interests, that's what gets you what you want.

Here's an example. I am trying to do important, interesting work, but I'm often interrupted, because I work in DevOps and wind up drawn into other people's problems. So on the whiteboard next to my desk, I wrote my two "Wildly Important Goals" (from the book Four Disciplines of Execution), and underneath that, the sentence "If something does not apply to these goals, why am I doing it?" This became quite the controversy, but it works! It has significantly reduced my interruptions. That's social skill in action.

What is aquascaping?

Especially since most people really can't read or write code for more than 4-6 hours a day. If you're working more than eight hours, you're either in crunch (and consider leaving for a new job), or your day is being filled with useless meetings and other non-work.