Blog
Long-form posts that combine the model’s outputs with urban-design context. Each post uses the same metrics that drive the rest of the site, with the numbers and the methodology spelled out so you can disagree with both.
Measuring what nobody mapped
The atlas now describes how you get around inside a park. For 91% of them the honest answer is that we do not know, and for slope there is no answer at all: the open elevation models read flat ground in Downsview as a 6% grade and flatten the Christie Pits bowl to one metre.
Colin Smillie
What people actually look for in a park
Visitors open a park page to find out what is there, not to read a score. Answering that honestly meant widening the amenity vocabulary with OpenStreetMap and reconciling the map itself, adding 515 missing parks including Biidaasige.
Colin Smillie
Why Trinity Bellwoods works
A neighbourhood park as Jane Jacobs would have liked it: dense mid-rise frontage, permeable streets, real shade despite a thin canopy polygon. What the metrics actually say.
Colin Smillie
Toronto's ravine paradox
A third of the Toronto Park Catalogue is ravine green space. Most of it scores low on every Jacobs-style measure, and that may be the right answer.
Colin Smillie
The hidden success of parkettes
Tiny, tightly-framed pocket parks routinely outscore much larger destinations on the model's terms. What that tells us about Toronto's mid-rise neighbourhoods.
Colin Smillie
A note on tone
These posts read parks through a quantitative lens, but they are not pretending to be the last word on any park. The model picks up things that map well to data; it misses what data doesn’t catch. They try to say what the data shows clearly, and to be honest about what it can’t.
For methodological details see the methodology page.