When Apple started this, they employed some GIS based people. GIS is not cartography. THat was the biggest mistake.
To say Apple were trying to be clever is a joke, they were trying to be cheap.
Read through the blog pointed to earlier by dv8r and myself (taken from another thread on this forum for my part) :
http://blog.telemapics.com/?p=399
Hiring GIS people would have been a good thing actually.
----------
God damn it, I know I've said it a hundred times, but they have $100bn, why don't they use that bloody money to set up a proper Maps department? Not with 10 employees, but with 100.
Money doesn't buy time yet. Throwing money and people at a problem rarely results in something functional. Some things just require time.
Heck, throwing too much money and people at a problem can actually hinder it as you introduce complexity and conflict in the design. Keeping a small core team and giving them proper time tables is sometimes much more efficient. The problem is it seems Apple rushed Maps out, and they made quite a few poor choices in the actual design of their systems to integrate data from providers and to Quality Check the result (as pointed to in the blog, the mistakes seem to come from automatized testing rather than human testing).