Visualizing Prediction Inaccuracy

February 26th, 2012
mbtaplot, transit
A major goal of mbtaplot is to visualize bus predictions by converting timing information to locations on a map. One thing I've struggled with is how to represent information getting out of date: the longer since we last got a position update the less accurate our estimate is. I think I've got a good solution now, adding a line back from the prediction to the last reported location. The longer the tail, the less sure we are about the position:






The main thing I like about mbtaplot over nextbus or mbtainfo is that because it gives me more information I can second guess the predictions. The goal with this new feature is to intuitively convey more information about the trustworthiness of the information.

(At some point I want to figure out how to get something like this on my phone; the current version uses too much javascript for that.)

Comment via: google plus, facebook, substack

Recent posts on blogs I like:

It isn’t always evil to ally with Steve Bannon

Scott Alexander wrote:

via Thing of Things October 19, 2026

Was the HuggingFace hacking incident scheming?

In July 2026, internal OpenAI agents hacked HuggingFace (HF) in order to figure out how their evaluation grader works. This involved a thousand agents coordinating on an improvised message board to carry out an elaborate sequence of exploits. A later grou…

via Victoria Krakovna October 18, 2026

There's no point at which turning your brain off will work

In early 2025, I started seeing people turn off their brain as they use LLMs1. They would have an LLM take an action (summarize text, write some code, etc.), and just assume that it worked2. This generally didn't work in early 2025 and the result was …

via Posts on September 18, 2026

more     (via openring)