There's a lot of social issues I care about - unions, voting rights, equal rights in general, the right to repair... and GNU/Linux. Wrapped up with my care about GNU/Linux is (A) the view that open-source development is fantastic, and (B) that security and privacy are growing and relevant issues in today's day. Furthermore, (C) that we should know what our software is doing - we shouldn't have to just "trust" corporations.
I care about these issues quite a bit... but I didn't just start caring out of thin air. In fact, back in the day, I didn't really care about these things. Why? Because, as far as I was aware, caring about privacy and security entailed a hassle, it entailed having to do a lot of figuring stuff out.
'Linux' always seemed like this wonky, unstable, alien thing. And the first time I made the leap to use it, it was because I was working on a project that required 'bash scripting', so I decided I might as well try it out. I looked around the web, found how to do get Ubuntu running (Ubuntu's website is actually pretty useful, ie this tutorial on making a bootable Ubuntu USB stick), and got it running. I was highly impressed on my first impression - and still am, Ubuntu is intrinsically far better than Windows.
So I stuck with it - really there was no reason to go back. For almost any kind of project, there is excellent software native to Linux which works well. If not, well, I kept Windows dual booted. I rarely use it.
For the first six or so months, I used Ubuntu just like a regular old desktop, and using the terminal to install things (which is much, much easier and better than just downloading things off of the web... although you can also do that with Linux). It boots faster than Windows. It doesn't auto-update like Windows. It's less vulnerable than Windows. There are a variety of other features which generally make it much nicer. That is to say, I didn't feel a reason to switch back.
As I became increasingly familiar with Linux, I became more and more inclined to use 'free and open-source software' (FOSS). This is largely because such projects are very good, and are native to Linux. This comes with a side-effect though - you start to become aware that secure and privacy-respecting alternatives exist, and are just as usable. For example, Chromium is effectively the same thing as Chrome, except it respects your privacy. Of course, it's available regardless of your operating system, but you became more aware of these discrepancies.
At first, privacy is a beneficial side-effect, not the main concern. But soon,having enjoyed the privilege of privacy-respecting software (for the most part), you start to become more skeptical of spying done. An abstract aim for 'privacy' is not what will keep most people into Linux - but the effectiveness of Linux will cultivate a concern for privacy. Soon you'll realize that all of the supposed 'paralyzing choice' in Linux doesn't really exist, it's all FUD - most applications dominate their field, and thousands of contributors work on the project. Often times, there are not thousands of little projects out there, so that you have 'radical choice' - software freedom is not about 'choice' in that sense, it is about privacy, the right to know what software is doing to you (and you know by proxy of no one in the community raising the alarm - YOU don't have to really do much).
There are other reasons why people evangelize for Linux - it is more effective for workflows, it is free, it is open-source (I guess I hit on that a bit), but none of these reasons are enough to marshal a market of consumers. These are legitimate concerns though. Instead of selling Linux on these points, we should sell these points on Linux. Many distros are pretty awe-inspiring nowadays, and very sleek. They are capable of selling themselves on their own merit. We need to market Linux. Of course, we can hammer on the other points. But on the basics, the fundamentals, the 'feeling' of using a computer - Linux wins here. And that's where the battle is lost or won.
Now this isn't to say that discussing things like the above issues (privacy, open-source, potential for better terminal workflows) would not contribute to Linux doing better. But they can't carry the weight. Right now, Google Chrome is the most popular web browser, Amazon is the most popular online market, social media is thriving, and Windows is the most popular desktop operating system - clearly the above issues are not a big enough problem in-and-of-themselves. We need to show there is a choice (well, the web-stuff is less obvious...), an ecosystem where meeting the above goals is feasible, and possible right now.
As 'Linux for Everyone' states: 'Looming over all of this is a stark, simple truth: beyond our bubble, nobody knows squat about any of these things. They’re not even thinking about it.'; 'Our target audience spends zero minutes per week even thinking about desktop Linux. Until that’s fixed, no other factor can even come into play.' [1]