Given that there is really no quantitative analysis of the optimal amount of dissolved carbonate in water used for meadmaking (at least none that I've ever seen published), you should take blanket statements like the one in Steve's paper with a grain of salt (tangential pun there). That said, as Steve was AHA's Meadmaker of the Year back in '08 he has had more than a little success with his batches. ;D Still, I'd rather something like this be phrased more along the lines of "I have consistent success with water containing 200 to 500 ppm hardness as carbonates," or something similar. Calling that level of salts "optimal" without substantiation is a stretch.
There is a lot of meadmaking "lore" out in the community, some of it quite well founded, and some of it based on very loosely tracked anecdotal experience. A portion of that latter info is quite frankly, simply wrong. I'm not saying that Steve's position is one of those - I'm merely suggesting that you need to be careful to find the basis of the substantiation of any claims or conclusions cited anywhere in the online meadmaking literature. In the absence of such data, then take the recommendations that you read for what they are - the conclusions of one meadmaker.
That said, I can state that I've had excellent success with water in the 300 ppm range, but I've also had good success (including a couple of meads that earned medals and mazers at IMF and Mazer Cups past), using my current source of relatively soft (hardness less than 100 ppm) water. I think the range is wider than Steve believes it to be.