The work flows for creating a map from digital data would be similar for a modern traditional cartographer, but I don't think the same can be said for most GIS map makers or data aggregators. In either case what has really changed are the data collection methods. They seem similar, a couple guys driving around collecting road data, but there are some subtle (yet important) distinctions. To begin with, contrary to what the film portrayed, for CSAA and ACSC at least, the field checker was often the actual cartographer working on the map being checked. They usually had a more intimate knowledge of the area to begin with and specific items they were checking when in the field rather than mass collecting road data. They (we) would also set up appointments to meet with city and county planners, engineers, fire department officers etc. to go over recent or future developments. This was usually a source of good future change information that data aggregators like Navteq rarely have access to (beyond simply collecting what the agencies make public anyway). The difference in the data collected shows up when Navteq has paper streets that don't exist (or won't exist for another year or more); has new developments that have road names the developers assigned on the plat sheets, but not what's on the actual road sign; or don't show future changes to local road alignments or bypasses as "under construction" or "future" because they didn't have direct access to the planners and are unaware of coming upgrades.
Another difference, one that has more to do with how we generate maps now, is that historically map updates were done one map at a time for the entire map. The aggregator approach is to collect in huge swaths to feed the master databases. The difference is that when you picked up a AAA road map with a revision date on it that meant the entire map was warranted to be as current as reasonably possible. With web maps the data date reflects that some portion of the current map view has been updated recently but not all of it. Again, not really a methodology difference but one of map type (web vs published).
Local agencies do indeed maintain their road data but experience working for over a decade in just this type of map production (and now as a city GIS specialist) has taught me that it is rarely to be taken at face value. In fact the reason most govt. agencies were so willing to work directly with us was that they received all of our correction notes in the end. That doesn't happen with Navteq or Google. And even when a city planning department is keeping accurate records on what's going on they may not being sharing it with their own GIS group. You might be surprised to find out how often various groups using GIS within a city don't willingly share their information.