Working Group Minutes/DWG 2020 01 29
DWG Meeting on Wednesday 29 January 2020 at 20:00 UTC
DWG members present
- Michał Brozozowski (RicoElectrico - joined later)
- Dennis Raylin Chen (Supaplex - joined later)
- Anton Khorev
- Vladimir Marshinin (mavl)
- Toby Murray (ToeBee)
- Tom Pfeifer (Polarbear-osm)
- Frederik Ramm (woodpeck) (chairing, minutes)
- Guillaume Rischard (Stereo - joined later)
- Andy Townsend (SomeoneElse)
- Marc Zoutendijk
- +3 more lurking
We frequently receive requests from property owners to "remove trails" that run across their property. When asked for any material that would corroborate the ownership, complainants often fall silent, which can be interpreted as them overstating their rights - perhaps they just own an adjacent plot and don't want others to walk past.
Most of these claims are related to trails being shown on AllTrails. We are aware that AllTrails show their own tracks as an overlay over an OSM base map, so we are only responsible if the trail in question is not from AllTrails' own database.
We discuss whether we should generally demand evidence for ownership before we act. We agree that while this is an option in dubious cases, we can also accept convincing claims not supported by hard evidence, or contact the original mapper as necessitated by the situation. We will never delete a trail that exists on the ground - at most we will mark it as access=private.
Organised mapping efforts not naming the ultimate controlling entity
We're seeing paid mapping where the mapper is a freelancer and takes orders from a client, or where the mapper is employed by a services company and that in turn is hired by a client. (For example, while Apple do their own mapping, they also contract with companies like Kaart, and we will then only be told that a project is run by Kaart but not that it is ultimately on Apple's orders.)
We agree that the Organised Editing Guidelines requirement to be transparant about who you work for when doing organised mapping is not restricted to actual emplyoment; it means that the community should have a chance to know who calls the shots - so if you work for company A and company B hires company A to do mapping, then the community should be informed that this mapping is done for company B. There is no room for "shell companies" in organised editing.
We will occasionally remind people of this situation but not go further at this point; we bank on involved parties wanting to do the right thing. Also we are aware that the individual mapper might not even know who ultimately pays for a certain activity.
A while ago, the OSMF board contacted us asking for our view of the relations between the board and working groups - do we want more, less, different things, etc.; the current situation is that the board has expressed a desire to have a chat about this. We realize that the new board has enough stuff on their plate and there is no urgent need to discuss anything from our side. We'll signal availability but otherwise not worry (todo: Andy).
various OTRS, website, and DMCA issues
We discuss the three ways in which requests reach us (OTRS tickets, website issues, emails generated by DMCA form) and how workflows can be improved. We decide to help improve the DMCA form so as to attract fewer meaningless submissions (todo: Tom, note: completed as of publishing the minutes, https://github.com/openstreetmap/chef/pull/263). We discuss some bugs and features of OTRS, and some details of the OTRS workflow.
We agree that we need to re-start putting out activity reports (todo: Frederik). Also, participation of DWG members in mailing lists and forums is a good idea to ensure we keep good community relations.
Setting up DWG Hall Monitor instance
We decide that we want to set up our own version of Ethan's "Hall Monitor" to watch objects and users (todo: Toby to take a look at Ethan's code).
Erroneous automatic "brand" tagging through the iD editor
The iD editor tries to "guess" brand names based on amenity names, and occasionally gets things wrong, for which we receive complaints (e.g. "Gino's", "HSBC"). We decide to flag these issues in a ticket to the "name suggestion index" (todo: Andy; note: completed as of publishing the minutes, https://github.com/osmlab/name-suggestion-index/issues/3521) but at the same time we will not wait for things to be fixed in the editor, and simply revert bad edits.
End of meeting
The meeting ends at 22:05 UTC. The next meeting will be in the second half of March.