There is nothing wrong, in principal, with consistently selling vol* to capture the roll - if you are aware of the risks and manage them. The total return to shorting the VIX is still significantly positive - it's not like accidents such as those on Monday wipe out all your accumulated profits if you've been doing this for 15 years. That means in practice you should probably not do so via leveraged ETNs (on which the pre-set leverage means you can and will be wiped out) but via futures where you can set the leverage at the appropriate level. That means you should scale your position so that an overnight doubling, or tripling, of vol is something you can survive.
(* Ok you can do slightly better: don't trade the very front contract which although it has the best roll also has the highest skew, maybe throw in a trend following filter on the underlying to try and get yourself out of the car accident before it happens [this probably wouldn't have worked on monday, but at least it means you won't get back in again until there is a clear downward trend in vol established], only sell when the future is in contango [backwardation is a sign of high stress, and also means you're not being paid for shorting, though again this won't get you out in advance, but will stop you getting back in too early], and sell less when the VIX is really low in absolute terms [though this is often when the roll is the highest]; but none of these will be a get out of jail free card that allows you to earn the premium from shorting vol whilst avoiding the horrendous drawdowns)
GAT