You are mistaking two effects. Indeed, glass does not "flow" (it's not a "very, very viscous liquid"); but most assuredly it does sag under gravity - just as a heavy steel beam supported at the ends would.

This problem is much simpler to solve for mirrors than lenses, as mirrors can be supported under their entire surface (see [0] for how it's done in larger amateur telescopes), while lenses - only along the edge.

[0] https://www.cloudynights.com/topic/562906-18-point-flotation...