Hacker News Reader

Comment by baddox | Hacker News Reader

> The five years he has spent on the quest is fairly typical.

> The Uber driver did the run from the newspaper's office in King's Cross to Big Ben, in Westminster, in 22 minutes; the cabbie did it in 18, by taking a slightly longer route he knew to be quicker.

The apparent extent of London cabbies is undoubtedly impressive, but I'm not sure the benefits outweigh the benefits of the Uber driver not having to spend five years studying the roads.

View on HN

The Knowledge's purpose is not just to know the fastest travel between A and B in ideal conditions (although a 20% improvement in that situation would be nothing to sneeze at), it's to know it in any condition and to be able to find B however it's specified and to be able to explain landmarks to tourists. In fact the original article on the comparison notes at least part of that:

> The Uber leg does not start well. After the app is downloaded, it stubbornly refuses to acknowledge that the Guardian exists. Only with the help of an old-fashioned desktop computer is registration completed and a car booked for a four-mile journey from the Guardian's offices in King's Cross to Big Ben in Westminster.

and the difference in fare between Uber and the black cab was 8%.

> and the difference in fare between Uber and the black cab was 8%

When not on surge pricing, UberX is generally 50-60% discount on the price of a black cab in my experience. Being able to get home for £7 after a night out is amazing.

Thanks for this. Have to decide between black cab and uber later this week and didn't know which would be cheaper. Do you know how accurate Uber's fare estimates are (do they update during surge pricing)?

You study the roads by driving them everyday. It makes you familiar with traffic patterns, POI, and small details that a GPS simply won't give you, (ie. when primary schools let out avoid this route).

I've met some Uber drivers who are using the service as mileage builders until they are ready to take the tests.

Google or Apple maps could easily suggest you "avoid this route when primary schools out" because they have average traffic speed data and can see that around those areas at those times traffic moves slowly. They don't have to know the underlying cause to be effective.

They could,but they don't. Also it would be good if they could predict traffic, like if there is a game in town or a planned protest, then don't take certain routes, even though they might be completely empty right now. A good cabbie will know that, sat nav - not a chance.

Google also doesn't know when there's roadwork, and will happily suggest that you take a street that's going to add 15-25 minutes to your journey, but only this weekend.

I once had it give me three routes: head down 210th until you're halfway to your destination, then take a right and drive to 220th, then keep going; head down to 220th until you're halfway to your destination, then take a left and drive to 210th, then keep going; or head down to 220th and drive the whole way.

My wife, who grew up in the area, said 'that's stupid', drove down 210th the whole way, and got us there faster than we would have otherwise (because we weren't taking any left turns on busy routes the whole way).

Been driving into the city for 22 years, not as a cabbie but a frustrated commuter.

There is no knowledge that can predict a system as complicated as London, as you will see by the number of taxis wedged in all the traffic jams. The utility of the cabbie is mostly mysticism and religious thought if you ask me.

However, the all seeing eye of Google Maps for example which aggregates just about every bit of info there is is all powerful. "drive home" - 42 mins, 8 mins of traffic, pick a route, gets worse on the way home, "faster route found!"

Yeah, this isn't the USA with it's perpendicular grid of streets, named 1st street, 2nd street etc

It's also a really small sample size. You'd could have gotten a similar result by comparing two Uber drivers, or two taxi drivers.

Also, I wonder whether the taxi drove in a bus lane (which are off limits to Uber drivers and regular people), and whether that impacted the journey time.

Exactly, 2 different taxis could just as easily have arrived 4 minutes apart, and the bus lanes are a significant advantage either way. A 4 minute difference doesn't sound too bad.

The benefits of a registered taxi in London are obviously a lot more significant than getting to Parliament from the station four minutes quicker than Uber. I'm sure you understand this and you're just commenting on the speed of passenger delivery.

Reply on news.ycombinator.com