At Shopify, we build into the monolith unless there’s a strong reason to build it as a new service.

It makes more sense for us to extract things than to make everything microservice.

Storefront makes sense to be on its own service, so we are making it so.

Projections that I read from urbanists about autonomous cars is that it will make transit worst, not better.

We’ll have empty cars driving around the city to pick some up. Causing transit. It’s the same effect Uber is causing to cities. More transit and people taking less mass transit.

The scenario that I read from urbanists is that people will continue to own an autonomous car, and the car will drive you to work, drive back home to pick up your partner, drive her/him to work, drive back to pick up the kids, then drive them to school, and on and on. So more empty cars on the road.

While I see autonomous cars as a cool tech, I don’t see urbanists thinking it will solve urban planning.

An autonomous car should have a better reaction time than a human. If we have 100% autonomous cars, and a traffic light turns green, we should be able to avoid (or reduce) the ripple effect where cars down the line don't even start moving until a while later. They can all move at roughly the same time (and none of these cars will be reading their phones).

Likewise they will react more predictably. We won't have an idiot cutting lanes, or tapping his brakes since he can't maintain a constant speed. Apparently most traffic issues are caused by a few bad actors having an over-sized effect.

So I think even with more cars, we can get better flow, which is what traffic is all about. But it will be difficult or impossible with humans in the mix. I'm pretty sure many drivers will try to bully the safer driving autonomous cars if they can.

2004 and 2005 had a gigantic spike in deflorestation that is pulling the average up. So the data is a bit skewed.