[OSM-talk] Remove opt-out from Automated Edits code of conduct

[John Whelan]

 > That doesn't mean that we should stop someone changing e.g. 
"highway=pirmary" to "highway=primary", and if someone says "I want my 
highway=pirmary leaving as they are" I'm sure the rest of their local 
OSM community will point out the error of their ways.

That hasn't always been my experience, but there again it was about 
highway=traffic_signals or highway=traffic_lights rather than 
highway=primary

Cheerio John


> Andy Townsend <mailto:ajt1047 at gmail.com>
> 21 February 2021 16:54
>
>
>
> ... largely by restricting the use of dumb automated tools and by 
> actually _having_ an "Automated Edits code of conduct".
>
> That doesn't mean that we should stop someone changing e.g. 
> "highway=pirmary" to "highway=primary", and if someone says "I want my 
> highway=pirmary leaving as they are" I'm sure the rest of their local 
> OSM community will point out the error of their ways.
>
> It does mean that tagfiddling between e.g. "contact:phone" and "phone" 
> (or vice versa) when both are in use is probably unhelpful.
>
> Best Regards,
>
> Andy
>
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
> Sören Reinecke via talk <mailto:talk at openstreetmap.org>
> 21 February 2021 16:33
> I agree. Part of OSMs future is using automatic imports and AI 
> specificially as well. Quality problems can be solved when they 
> occurr. We managed to maintain a map with higher quality data than 
> Google. Why also not repeating the story with automated edits and AI?
>
> I feel that some OSM members are rather old-fashioned about automated 
> edits and AI than modern and looking forward in using modern 
> technology for the good of humankind as we previously did and do with 
> OSM. But that opt out stands as a wall and preventing that.
>
> Another fact: Not only a single person edits an object but multiple 
> users do. So complying with wishes from individual users lead to 
> inequality of data since the tag the user touched is blocked for 
> changes made by automated edits (applies to all tags world wide the 
> user touched and get altered by the automated edit). It leads to 
> inequality because all other objects surrounding the object the user 
> touched having the same tags in question get altered. This is like 
> having a B in a field of many As' and when it comes to e.g. replacing 
> Key:diaper with Key:changing_tables and an user does not want this to 
> happen on Key:diaper tags they added it leads to that apps like 
> Babykarte dropping support for Key:diaper going to ignore the tag the 
> user added. This is because how a developer is going to thing: Why 
> should I support Key:diaper because only this one user things that the 
> automated edit which made my live and the lives of others easier would 
> had their work destroyed? A developer is definitely going to say, I 
> will simply ignore Key:diaper completely and only support the newer 
> Key:changing_table . This decision leads of course to not displaying 
> changing table information on this single object because of lack of 
> support for Key:diaper.
>
> Do you really want single users to prevent/slowing down the future as 
> the German party CDU/CSU does? German politics as exercised by CDU/CSU 
> work in the way of slowing down future as long as "old people" not 
> open to changes happening to their surroundings are pleased. There is 
> a reason why young people like me do not vote for CDU/CSU. We should 
> not act like Berlin, we should be open but also critical to new 
> technology, learning to use it and then make it to help us by setting 
> up our own guidelines and standards or adapting existing ones (perhaps 
> with modifications to it).
>
> Cheer you all
>
> Sören Reinecke alias Valor Naram
>
>
> -------- Original Message --------
> Subject: [OSM-talk] Remove opt-out from Automated Edits code of conduct
> From: Mateusz Konieczny via talk
> To: OSM Talk
> CC:
>
>
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
> Mateusz Konieczny via talk <mailto:talk at openstreetmap.org>
> 21 February 2021 13:56
> I propose to remove
>
> "Respect "opt out" requests, i.e. if someone contacts you and asks
> you to stop making automated edits to things that they have edited,
> you must comply with that wish, and you must modify your software
> or procedure to leave those objects untouched in the future."
>
> and
>
> "Information on how to "opt out""
>
> from https://wiki.openstreetmap.org/wiki/Automated_Edits_code_of_conduct
>
> Either automated edit makes sense and there is no good reason
> to exclude objects that were ever edited by specific person
> or automated edit makes no sense and would be rejected
> during review process.
>
> ---------------------------------------------------------------------------------------------
>
> Disclaimers:
>
> I made some automated edits, I never encountered
> anyone requesting this, I am not aware about anyone requesting this,
> I am not aware about any case where it would made sense.
>
> I make fun of that requirement in my bot documentation.
> ---------------------------------------------------------------------------------------------
>
> Note that it is part of a more general problem:
>
> Current Automated Edits code of conduct is in quite ridiculous situation:
> basically all automated edits violate it completely or partially.
> Look at https://wiki.openstreetmap.org/wiki/Category:Automated_edits_log
> - half of pages are for my edits, and I am definitely making far less
> than half of automated edits in OSM.
>
> It has an interesting side effect that you can revert almost any 
> automated edit,
> because it is done in violation of code of conduct.
>
> It would be useful to remove some of onerous and pointless requirements
> and keep just what actually matters.
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk

-- 
Sent from Postbox <https://www.postbox-inc.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20210221/d558b575/attachment-0001.htm>