開放街圖標誌 OpenStreetMap 開放街圖

I am using Osmose to find and fix polygon errors in Ireland. I fixed a problem in a golf course in Youghal, Co. Cork. I could have stopped there, and moved on, however the JOSM validator found some errors, so I kept working on that little area. This led me to spot some not great mapping around a school. Some houses tagged with name=Residential Dwelling, that kind of thing. I cleaned up the area. Removed unneeded name tags, fixed tagging of roads etc. I found the mapper, and changeset where this was added, and was able to leave a changeset comment on it. I’m not the first to notice, but maybe this mapper might return to OSM.

So I feel like I have improved the map in this area. I would not have known to look there if I hadn’t use a QA tool like Osmose, or the JOSM Validator. However if I was just looking at closed polygons, then I would have stopped as soon as I fixed that polygon, and moved on.

At State of the Map 2016, Frederik Ramm talked about Mechanical Edits, and one problem he finds with mechanical edits is that sometimes people just mechanically and automatically fix the problem that the QA tool has found, rather than looking at the area, or other contributions from the same mapper, and try to fix the larger problem, and fix other problems. There is often talk about more automatic and manual imports, with some suggesting that humans are expensive, let’s not use them too often. But you need to have humans there, to find and recognise other mistakes which the tool can miss, as my example shows.

電子郵件圖示 藍天圖示 Facebook 圖示 LinkedIn 圖示 乳齒象圖示 Telegram 圖示 X 圖示

討論

BushmanK2017年01月 9日 02時16分 發表的評論

Objects like this osm.org/way/331415576 are still tagged in a wrong manner - it should be just building=garage without any name or something.

amapanda ᚛ᚐᚋᚐᚅᚇᚐ᚜ 🏳️‍⚧️2017年01月 9日 09時01分 發表的評論

Objects like this osm.org/way/331415576 are still tagged in a wrong manner

It’s been fixed now. I missed that one. It just goes to show that we need a community of humans looking at the map. 😄

gileri2017年01月 9日 21時22分 發表的評論

“However if I was just looking at closed polygons, then I would have stopped as soon as I fixed that polygon, and moved on.”

Yes, there would have been errors in that area that would have stayed untouched, but by checking the area and fixing those errors you (most probably) left broken polygons on the map during that time.

Both approaches (“subject-driven” and “area-driven”) to fixing the map are valid; as the map can be always enhanced and human time is not infinite you’ll always have to make a choice of what you want to add/fix and leave something behind.

That’s the curse of the mapper :)

Hjart2017年01月13日 20時53分 發表的評論

Personally I find the “area-driven” approach, which Rorym so nicely describes here, much more satisfying than the “subject-driven”. It has many times caused me discover issues (sometimes serious) that most likely would never have been discovered otherwise. I have in many many cases seen “subject-driven” fellow contributors leave behind issues in the immediate vicinity which were much more serious than the one they fixed. This is not to say that the “subject-driven” approach is not valuable, but if you do have some extra time to spare, looking at the area around the initial subject can definitely be worth it.

登入 來留下評論