Grabbing a random link that roughly talks about the topic:
http://www.nytimes.com/2015/02/24/m...-gap-between-moviegoers-and-academy.html?_r=0
There's been a lot of words / analysis done on the idea of "popular vs artistry" - and while different awards definitely cater to different groups (see your Cannes vs MTV Movie Awards) - I guess I think of the Emmy / Oscars / Grammy as the three "mainstream" awards - the awards ceremonies that aren't tagged to just popular culture (ala MTV) or just a given festival (Cannes) or even just pure artistry; I've thought those three have been considered to take all of those into account. In those mainstream awards; I wouldn't want a movie to be inherently discounted because it was popular or because no one saw it.
Aside: The LotR one is a bit more personal because a roommate's (when I was in college) mother happened to be a voter at the time, and she had given us some inside scoops on the politics involved for LotR and Best Picture. (On top of it, one of the girls on our floor happened to be dating one of the main LotR actors at the time, so we got to listen to a hilariously drunk celebratory phone call / party when RotK got 11 awards. Small world.)
It sounds like the Hugos are the Sci-Fi literary equivalent of the Academy Awards / Emmy / Grammy; which is where I was going with it. If the Hugos truly were just becoming a self-serving echo chamber (which people have accused the Oscars of being, especially w/r/t Selma and American Sniper); then I can get the argument of "hey, we need to shake things up". But, as it is being confirmed by posts, that argument seems to be a shell for the true motivation to push what books "should" be.
TL;DR - I can see an argument for making sure what purportedly is the "mainstream" set of awards aren't becoming insular and self-serving - but it sounds like that is purely an excuse being used to cover real motivations in this scenario.
I think one of the more fascinating subplots of this is that we may have to re-evaluate the entire idea of representative subsets of data when it comes to fandom. I imagine in the past, a network might get, say, 50 letters for a given show, and they could generally assume it was at least somewhat representative of overall fan reaction (knowing that they're getting those who are passionate enough to write letters one way or the other). But with the power of modern communication & social media; I believe it's really easy to tilt something disproportionately - such that those representative samples aren't really indicative of overall sentiment. I feel like Gamergate more or less abused those older notions when they first started; sending a thousand emails to a company complaining about XYZ thing may have initially panicked the company into thinking there was a mass disenchantment towards XYZ thing, as opposed to just a thousand coordinated folks. I think they did some twitter analysis and found that there weren't actually that many people involved in Gamergate, it was just a lot of the same people talking over and over, or making shell accounts. Could be wrong though.