RETEX: My Urban Recycling Trekking

To be continued, maybe:

• diary entry (upcoming): choosing tags for this trekking
• diary entry (upcoming): existential questions about my encounter with Panoramax

Context

Since April 2025, I’ve been discovering OSM and trying to contribute wherever I can. Needless to say, I’m learning something new every day about OSM mapping (thanks to the Wiki, forum, and fellow contributors).

While exploring my neighborhood, I added a voluntary drop-off point (for recyclable — though not always — waste) to OSM, located 800m from my home. I then discovered that the local intermunicipal authority has an app listing a large portion of the collection points. I wondered whether it was legally and technically feasible to retrieve those, and whether it would align with the OSM ethos.

The forum quickly (and kindly) set me straight (as I said, after only 2 months with OSM, I’m learning a lot every day):

• Extracting data from a website’s database is obviously illegal unless there’s an explicit license that allows it.
• Even if there were a friendly license (which is not the case here), it would still need to be ODbL-compatible to import the data into OSM.

This led me to pursue two parallel paths:

1. Finding a sustainable solution via an open data source, which means:

   • Trying to convince the intermunicipality to publish its data, ideally on www.data.gouv.fr
   • Convincing them to make this publication sustainable (i.e., generate an export every time there are updates)
   • Assessing how, in theory, this data could be used (completeness, attribute matching, duplicate detection, handling conflicts between open data and on-the-ground data — which can either be better or simply incorrectly entered)
   • Finding out how and by whom the data could be imported
   • Seeing if all of this could happen before 2035

2. Starting manual mapping myself through field visits (the foundation of OSM’s truth)

As it happens, since retiring, I’ve enjoyed walking on cool routes (max per day: 10–30 km, 600 m elevation gain, no technical sections), and I know quite a few trails in my region.

And that’s how my Urban Recycling Trekking was born.

Purpose and Organization

Quantity

Initially, I estimated 284 PAVs (voluntary drop-off points) within the intermunicipal area, which seemed manageable by the end of the year. In practice, I was completely wrong — it’s more like 650 locations, each with 2 to 8 distinct containers. That brings us closer to 2,000 or 3,000 nodes to enter.

They’re spread across about 60 municipalities, with some hosting nearly 250.

Locomotion

I briefly considered doing a car grid with side images (I don’t have a 360° camera to mount on my roof), which might allow manual OSM input from the photos. But I quickly ruled that out:

• It’s not very “fun” and wouldn’t really be worthy of a journal post
• Many PAVs are positioned with their backs to the street, so a roadside photo often doesn’t reveal the container types, let alone their reference numbers

I also ruled out using risky transport for someone my age 😉 (skateboards, rollerblades, scooters), and — more debatably — even biking.

So I’ve settled on walking-based mapping loops through urban areas, driving to the starting point in a fully electric vehicle.

What to Record

The goal is to record voluntary drop-off points (PAVs), mostly semi-buried or surface containers. Since their locations are listed in the GPSEO app, I can pre-plan efficient routes and aim for decent coverage.

Additionally, before even starting, someone asked me to contribute photos to Panoramax of the various sites (see my [future] journal entry Existential Questions About My Encounter with Panoramax).

So I need to walk past each site, spot it, photograph it, and later upload data to both OSM and Panoramax.

I also considered combining this with other data collection:

• StreetComplete seemed like a great option, but a first test showed it to be very time- and battery-intensive. In urban areas (its main use case), I spent all my time tagging street surfaces, pedestrian crossings, and bus stop benches. I know I could filter out some quests, but I chose to exclude StreetComplete from this project (though I still use it for less targeted outings).
• Since I already use OsmTracker to log PAVs, I thought I could also record other objects for others to map later from my traces. This wouldn’t cost much extra time or battery. But I dropped the idea because mapping without photos might be too imprecise, and I didn’t want to photograph every single object. Maybe I was wrong about this (and I might change my mind later).

In practice, I’ve made one exception: fire hydrants I encounter (thus non-exhaustive, since I only note those during PAV-focused walks). I tend to walk with my head down, so I can’t miss them (they’re red and under a meter tall). So for these: OsmTracker tag + photos (for my own later mapping). I don’t upload these photos to Panoramax (but I keep them in case it turns out useful later).

---

These RETEX (feedback) journal entries reflect my beginner’s choices, doubts, and questions. These are my own opinions, not Wiki entries. Many of these choices have been discussed on the French forum, but not all. I remain completely open to feedback and have no intention of offering recommendations here.

---

(translation from ChatGPT)