This is a follow-up to A Year of Google Maps & Apple Maps (discussed here). The last story was tagged with design, but this was less design-focused, so I decided against doing that here as this article is more about feature extraction.
I’m amazed by the level of detail in this post! Incredibly interesting comparisons and analysis.
I’m amused by the level of Cupertino Kremlinology. But yes, this (like its predecessors) is a high-quality writeup by someone with some depth in both digital cartography and business models.
Suggested “practices” tag, because the “AI” is apparently (at least in part) 5000 new employees in the new Hyderabad office. The more you know!
google employs similar numbers to edits its maps. When I was still at TomToms map unit (former TeleAtlas) 6 years ago google had already more people manually editing its map than TomTom had employees across all business units worldwide. Map making is suprisingly complex and requires a lot of manual work.
I worked on the US Census Bureau’s TIGER maps over a decade ago. Fixing some of their legacy maps (like, some of them just digitized paper maps) to modern commercial maps while preserving historical data was a great summer job! We churned through it pretty quickly with software help. Flyover counties in North Dakota could be cleaned up in an hour or two, while Queens, NY took two weeks (and the beefiest machine in the house, a Sun Blade 2500).
All that sounds pretty damning for Apple maps.