It is really impossible to overstate the impact that Michael had during his time in the Mac community. I don't think that it would be hyperbole to state that there is not a single functional Open Directory implementation running today that he did not somehow assist, either directly, via his training classes, his book, posts to far too many mailing lists to count, email, his contributions to AFP548.com, or iChat.
He was one of those guys who would get paid, and well, to fly somewhere and fix a failing Open Directory implementation, then give out that same advice to thousands of people for free.
I know that personally, over the years, there was no way I could have talked any company I worked for into writing the check for the hours of free advice he gave me directly, mostly because I just asked.
Beyond the technical work he did directly, he was a tireless advocate to Apple for the needs of sysadmins, and a rather large thorn in their side, because he tended to be right, pretty much constantly. He knew that Apple could, and should do things better, and he was absolutely fearless in demanding that, even if it might burn some bridges along the way.
I know that for many of us, "If Michael said that, it must be so" was how we looked at things. To achieve that level of respect and trust from one of the crustiest, most cantankerous groups in the computing world takes a lot of hard work and dedication, and Michael had that to spare.
What a lot of people didn't know was how many long hours he spent diving into the internals of things like Open Directory, and Kerberos, and Active Directory, just so he would know how to do things the right way. Doing things the right way was quite important to Michael.
I know that there are lots of smart people out there, and in time, someone else, or a few someone's will step up and share knowledge the way Michael did. But there's no one person who's going to replace him. On a personal note, he was a friend, and I know I'm going to really miss him.