I have used (and continue using) many tools and solutions that are not popular. Some of them make me happy, some do not. All of them bring practical advantages.
Not every good tool will become popular, and many popular tools are not good.
I think the corollary that people, when looking also at the data or at constraints, are acting irrationally or its sophistry or whatever is bunk, though. Plenty of people do actually engage in some amount of evaluation, and it's intellectually lazy to disregard that because they may have also made an aesthetic or otherwise personal decision on some level as well.
I use a bunch of niche software, and it's difficult or impossible to completely separate my aesthetic preferences from the concrete reasons I also have for doing so -- my aesthetic preferences and my joy in using certain things actually stem from those concrete benefits in some cases!
> Not every good tool will become popular, and many popular tools are not good.
Yes. (In my opinion, there are also specific kind of features that it has or lacks, that I might think is good even if it is uncommon or bad even if it is common, but there are also features that are good as well as common, or that are bad as well as uncommon.) (However, there is also the consideration of: something might be good for some purpose but bad for others, and might be used for many of the purposes that it is bad for even though there is clearly something better.)