Example of task with a mis-aligned bridge, as seen from aerial imagery, and detected by data analysis.

Data analysis don't always catch mis-alignments, like this one.

I tried to find fault/misaligned imagery with extensive observations in this area, but Bing (though cloudy in some parts) is very well aligned with Strava within this boxed imagery. I cannot find any significant anomaly. Good works on Bing (in this boxed area), unlike Digital Globe which have many distorted/stretched images found.

Boundary: Generally from upper Quezon City/Kalookan (on the North); On the East up to Batasang Pambansa in Quezon City, left of Marikina River up down to Taguig (lower righ corner), but up to NAIA runway only on the South.

This will save a lot of time on workflow (in fact I don't recommend aligning and interpreting partial GPS data, nor individually interpret Strava for localized alignment). I HIGHLY RECOMMEND that we should just simply align all OSM data exclusively on Bing within the said boundary, as well as align all other imagery (Mapbox, Digital Globe, ESRI on Bing) for the latest map feature update in the area.

The effort on this particular project should focus on establishing accurate offsets on IODb for Mapbox, Digital Globe, ESRI referenced to Bing (on its default Zero,Zero offset) for ease of tracing and realigning streets and adding new roads.

Determining offsets, and recording them on IODb is a secondary goal @ralleon

Take advantage of AI with Strava's iD Slide, which looks very helpful for major roads, but must be used with caution:

• Check intersections
• Only use on roads with very high heatmap values

Use the following for the Custom URL Utilsplugin2: https://strava.github.io/iD/#background=Bing&map=19.00/{#lon}/{#lat}

Grab's OpenStreetMap data team has been in touch, and is co-ordinating their effort to identifying missing roads, and improving road geometry in Metro Manila. They've also adopted the use of the Imagery Offset Database to maintain mapping consistency across the mapping community.