The "OSM Standard tile layer" looks wrong (white lines, abusive comments etc.)

[stevea]

Thanks, Paul. That’s succinct.

[ZeLonewolf (Brian M Sperlongano)]

@Geonick

I merely suggest that the DWG’s activities, which it carries out anyway, be made available in a central feed.

That’s not always appropriate, for a variety of reasons. Much of the peacemaking has to happen behind closed doors to be effective. They do produce a quarterly report somewhere. My favorite line item was the time that a college student asked them to block his account so that he could focus on his exams.

[Geonick (Stefan Keller)]

@ZeLonewolf

That’s not always appropriate, for a variety of reasons.
(…)
My favorite line item was the time that a college student asked them to block his account so that he could focus on his exams.

Nice example. You can leave out exceptions in the log/feed. In addition this is a nice example of clear non-vandalism.

I am not looking for excuses or arguments for a “mission impossible”. I am looking for solutions.

@ZeLonewolf

They do produce a quarterly report somewhere.

This could be the way to go; i.e. the DWG keeps a log anyway. But a “redaction feed” would be a feed, as the name suggests.

[dieterdreist (Martin Koppenhöfer)]

Bringing it back to this vandalism, it could be redacted because it’s defaming someone. That would be if Andy complained, but he could redact it himself.

I think you should do it, him complaining or not, it is obvious we don’t want this in the history

[ivanbranco]

@ZeLonewolf

They do produce a quarterly report somewhere.

https://osmfoundation.org/wiki/Data_Working_Group

The last one is from 2023 Q2 tho. It would be interesting to make the new ones public as well.

[Rijk]

I still keep finding new examples, this time in CyclOSM, all of which have been resolved today after I reported them. I’ll report again when I find more.

[SomeoneElse (Andy Townsend)]

Someone has suggested https://github.com/osm-fr/infrastructure/issues as a way of reporting these issues with the French infrastructure.

I would suggest that a “remove all items from cache from the problem period” is a better approach than “wait until each example is reported”.

[InsertUser]

On the French forum they’ve said:

Fond CyclOSM

Il a été grandement impacté à partir du zoom 5.
La base de données (répliquée) qui sert à le produire a désormais été mise à jour.

Toutes les tuiles dans les caches ont été marquées comme devant être régénérées.

It would appear that they did this by resetting all tiles to a false render date, which confused me a bit because they were still showing as ‘clean’.

I think requesting them might be the thing that triggers a re-render (due to them being ‘old’) rather than the report. Deleting the tiles from that time period seems like it would have been a better approach as this seems like it will linger for ages, but I’m not sure how easy this is now that they’ve got fake dates on them.

[SomeoneElse (Andy Townsend)]

@InsertUser

but I’m not sure how easy this is now that they’ve got fake dates on them.

Assuming that “normal” OSM tools are used it should be pretty straightforward since the “make things older” process makes thing a known amount older.