$\mathbb{R}^2$ is the set of all ordered pairs of real numbers. A rigid motion on $\mathbb{R}^2$ is defined to be a transformation from $\mathbb{R}^2$ to $\mathbb{R}^2$ that preserves distances. It can be shown that a transformation from $\mathbb{R}^2$ to $\mathbb{R}^2$ is a rigid motion if and only if it can be gotten either by applying a rotation about the origin then a translation or by applying an inversion about the x-axis then a rotation about the origin then a translation. In this answer, I define a rigid motion to be noninverting when it can be gotten by applying a rotation about the origin then a translation and inverting when it can be gotten by applying an inversion about the x-axis then a rotation about the origin then a translation.
I don't know why people consider a glide reflection special so I will make a guess. I'm guessing that sometimes people make the assumption that all inverting rigid motions are reflections, and when they discover that that's not the case, they find it interesting, and that's what made the glide reflection special.
In the past, I independently thought all by myself about how it's interesting that not all inverting rigid motions are reflections. The inverting rigid motions that are not reflections are glide reflections. Maybe some people consider glide reflections special for that reason.