Comment by drunner
The example by the author is an unsafe one. You think putting unsafe sharp curves at the bottom of hills for cars is acceptable?
Your example is good, and the matching example of a bike lane is the opposite of what was shown in the article.
> You think putting unsafe sharp curves at the bottom of hills for cars is acceptable?
Yes, it is absolutely acceptable and it happens all the time in road design. It happens specifically because people (apparently in all vehicle types) will speed on the long straightaway down the hill into whatever comes next. Many, many suburban areas with hills will have sharp curves or T-intersections at the bottom of hills as a matter of course and it does work at slowing down the traffic going down the hill.
If you look at street view on the actual road, that is exactly this situation: a very straight road with a lot of visibility, leading toward a very long and complicated intersection where everyone needs to pay attention and go slowly. Nobody traveling a reasonable speed on a bike will be surprised by that turn since they can all see it coming for literally a mile.
I assume the alternatives to this specific merge are either:
* Merge the bikes into the car traffic and pray to god that they obey traffic signals at the upcoming intersection (we all know this isn't happening), while also accepting that they accelerate very slowly compared to cars.
* Set up a dedicated bike lane with dedicated signals (which is very expensive).
The engineers here clearly opted to instead merge the bike traffic with the pedestrian traffic through that intersection so that the existing pedestrian signals apply to them. You can see that the bike lane continues after this intersection, so they literally just did this to handle the intersection.
TFA on this point reads like "cyclist (or illegal moped user) doesn't want to be slowed down and doesn't care about anyone else."