[OSM-talk] Mechanical Edit?
steveaOSM at softworkers.com
Sun Jul 25 21:26:26 UTC 2021
Let's be deliberate and careful here.
Our wiki are useful guidelines and really DO act as a sort of de facto documentation. Both of you are correct that wiki are an important "pillar of truth" in OSM.
It is easy to point a finger at a large-scale edit and say "I see something fishy here." Those edits likely happen every few seconds, if not MORE often than that. (I estimate here, but I don't think I'm too far off). It is easy to get a bit overly-aggressive on what documentation (wiki) says and perform what are in some technically-true sense "automated edits."
There is a word "prudence" (being "prudent") meaning "acting with or showing care and thought for the future." It appears that both of you (Mike and Casper) are acting with prudence here. It is somewhat remarkable that it has come to a public forum like OSM-talk, as perhaps it should. Sausage is made, sausage-making sometimes has its curtain pulled back and a wider audience sees "how" sausage is made. Here, anybody who likes has a chance to say how sausage is made.
I've already said we want to draw a careful line between "this looks fishy" and "I'm merely correcting documented errors in the map" which does likely does NOT step over the boundary of "mechanical edit." Just because one uses a "power tool" (like OT) to determine these data — and these might and often do range over a very wide area — doesn't necessarily make it "all wrong."
Where do we want to draw this line? We've identified the navigational aids on how to talk about this, we've got a solid example of where "this" happens (hey, looks like a wide-area and/or mechanical edit) and "that" happens (hey, I'm simply correcting what are documented errors in the map) and hm, seems like a clash. Is it? I don't think so, but I don't know how to tease these apart.
> On Jul 25, 2021, at 2:12 PM, Mike Thompson <miketho16 at gmail.com> wrote:
> On Sun, Jul 25, 2021 at 3:02 PM Casper Kersten <casperkersten1 at gmail.com> wrote:
> 3) I would be ok with the changesets here if they had been discussed with the community and documented ahead of time, although during the discussion I would have pushed for smaller bounding boxes, which is entirely possible with the type of changes we are talking about.
> The documentation is already on the Wiki. These tags are already marked as deprecated and there is a clear and unambiguous alternative for them. Rather than replacing these tags locally over many years and still missing most of them, I chose to update them systematically to save other mappers the effort of checking buildings individually for outdated tags and to save data users the effort of having to consider two or three tags that mean the exact same thing. You're very welcome.
> You must document *your* *plan* for making the changes, which you didn't do prior to making the edits (correct me if I am wrong, and provide a link). The fact that these tags are deprecated, and that is documented, can (and probably should) be cited by your plan, but that documentation is not a substitute.
More information about the talk