It's a bit of a silly dichotomy. No linguist will deny that certain sentences are ungrammatical ("cat sees I"), so they are just wrong. And lots of philosophers believe that normativity, what is wrong and right, is an important part of language. But it appears to be something dynamic, language changes over time, yet I can't change language by pretending dog means cat today.
So prescriptivism is wrong in the sense that you can't claim to have the sole authority on something. But descriptivism is vacuous, in that it doesn't explain the pervasive tendency for people to consider things to lie on continuum of wrong and right.
I offered neither a definition nor a description. Indeed, descriptivism doesn't mean that, and that link is where I got the example from. I was correcting the implicit definition people use when they invoke descriptivism as a cop out from any argument on usage (although I certainly applaud ending any such argument). Descriptivism is a stance for linguists to take in their research, and has nothing to do with the rest of the world.
Coming back to the original point (and not in response to you in particular), using "begging the question" to mean "raising the question" is very common now, and trying to object to that does no one any favours.
That it's common is not sufficient--the majority is not necessarily right. I think pointing out a historical misunderstanding is worthwhile, especially in a forum with smart people or in an academic setting. Using this expression with a meaning based on a misunderstanding reveals that one wants to sound learned without actually being that, so I do think it's a favor to point it out. Of course it's an uphill battle and there's no hope, but I still think there are enough people who do care about where things come from and using them right.
