So I had two helpful commenters on my journal offer two pieces of advice from my last entry. In addition, there’s also one other avenue I’m going to try.

1. User Traces

1. There is this thing called “User Traces”. When enough users add traces, it gives you a rough mean of where the roads are in relation to the imagery.

What I was able to find was a couple lines on First, which is expected for a small town.

2. Strava

Strava is a social network for runners and the like. I’m not terribly familiar with them, but apparently, they have offered permission for their heat map of users to be overlaid on ID and JOSM. Possibly. Don’t quote me. However, like anything web, these alliances are short and fleeting. Here is a quick summary of the two tickets which offer the current state of affairs:

https://github.com/openstreetmap/iD/issues/4896 https://josm.openstreetmap.de/ticket/16100

… See full entry

I am especially not good at dealing with image offsets.

My challenge is finding a close-enough to truth for certain locations. In absence of proper markers, what I will be using for reference consists now of the following:

1. The roundabout.
2. The lanes of the Trans-Canada highway as they meet Roberts.

For Bing imagery, this gives me an offset of 1.61, 1.53

I’ve realized a few weeks ago that I’ve been tracing building outlines to non-corrected imagery.

At this stage, a lot of the road information seems correct except for the service roads that are between the main residential streets in the core of ladysmith.

I’ve been attempting to correct my prior traces slowly, while using a correction factor of 6.84, 1.19.

I’ve made the intersection which I’ve marked on the map (2nd avenue and white) my source of truth. This is assuming the location of the roads is correct.