Particularly in Africa, there are huge number of small round buildings. I believe the best way to model them is with a single node, tagged building=hut; optionally with a radius or width. However, the dominant OSM practice is to use a whole way with lots of nodes. So, I must go along with the community. Unfortunately, I encounter a lot of this while working, for example on #MapLesotho
That’s 7 huts, 130 nodes, all the wrong size, some overlapping - likely the result of very bad copy and paste. (There are actually 8 huts there; they missed one.)
My fingers have developed the rhythm for cleaning this sort of thing up in JOSM. First set the simplify-way.max-error to something appropriate (0.3), then for each hut:
- Click to select buliding.
- ‘G’ to unglue it from the surrounding buildings, if needed.
- Drag to center it correctly.
- Ctrl-Alt-drag to scale it down to the correct size.
- Shift-Y to reduce nodes.
- ‘O’ to re-round the circle.
Within a few seconds I have this:
8 huts, 86 nodes. It would be probably hard to automate this further without computer vision (because of the manual visual alignment), but using the standard JOSM position of left hand on keyboard, right on mouse, it can go very quickly with the above steps. One possible optimization might be to combine Shift-Y and O into a single keypress i can reach with moving my left hand.
Needless to say, if you are setting up a mapathon for people to map Africa, PLEASE teach them how NOT to make the bad big huts like the above, so I don’t have to fix so many thousands of them.
Discussion
Comment from BushmanK on 20 December 2015 at 02:41
Honestly, I don’t think it’s caused by the lack of skills (which could be fixed by teaching). It seems to be caused by ignorance, which is a kind of deeper problem. Many contributors of OSM have never used OSM data themselves, they don’t understand, how and what for it’s supposed to be used. For them, mapping is a game (more or less entertaining), which involves some sort of score (time, number of objects, covered territory or whatever). I’m struggling with the same thing in discussions of tagging schemes - people in general don’t care about meaningless tags, misuse of tags and so on. They just like the process. So, changing this situation requires something more that just saying “don’t do that”.
Comment from bdiscoe on 20 December 2015 at 02:45
It’s almost always “too many” poorly placed nodes, but just now I found a rare example of “too few”:
I’m guessing this newbie apparently considered nodes and ways so precious that they attempt to cover multiple huts with a single minimalist blob :) and yet they still manage to create bad overlaps and self-intersections. I want to applaud their urge to model efficiently… but I’m still stuck slogging thru cleaning this up :)
Comment from BushmanK on 20 December 2015 at 04:36
Or they just don’t know how to use circle tool.
Comment from MKnight on 20 December 2015 at 11:21
Comment from SomeoneElse on 20 December 2015 at 19:48
The bit I don’t understand is “… So, I must go along with the community”. I don’t understand what benefit there is to using an artificially circular way as opposed to a node (at least if way_area isn’t going to be used to determine size - which if the examples on this page are anything to go by, it couldn’t be reliably). Surely (while there are still things to be map) the benefits of mapping throughput would outweigh the prettiness of overnoded ways?
Comment from bdiscoe on 20 December 2015 at 20:00
@MKnight Indeed Shift-Y, I’ve corrected it above. Apparently my fingers know it perfectly, but my higher brain not always :)
@SomeoneElse The only “benefit” to going along with community practice is that people are happier, I don’t have to respond to dozens of messages on HOT tasking manager telling me my 1-node huts are “wrong” and “please read the instructions”. Most of OSM is based on rough consensus and there’s a limit to how much energy I want to put into arguing about tagging or other practices; it’s “pick your battles” and otherwise just go along.
Comment from malenki on 24 December 2015 at 10:05
A faster way of correcting the huts (utilsplugin2 required):
Instead of always drawing a new hut I often just copied and then pasted one since a lot of them have the same size. Thus you would need even one step less – three instead of six.
For way vs node: What would you do if there was a rectangular hut instead of a circle shaped? Still use a node with radius, width or length? I find it easier to map the length of the way instead of describing it.
Additionally: It is surely more rewarding for (new) mappers to see their added buildings show up on the map instead of it remaining blank. (afaik building on nodes aren’t rendered)
Comment from DeBigC on 26 December 2015 at 13:45
Just to be clear here… #MapLesotho is a crowdsource, trying to build a base-map of a country out of nothing (there was nothing in 2013). Because 800 people have been involved the interpretations of the hot task instructions differ. There are thousands of errors like the ones shown. The majority of mapping in Lesotho (as shown by Martin Dittus) is done by a very consistent core of <10% of mappers… who are the backbone and consistent influences. There is a phase of work after this one which will address the problems,.. in due course. Geofabrick tools will help, of course a certain level of human intervention is needed too, and validation is what allows us find the problems like the ones above.
Absolutely every single effort to assist #MapLesotho is warmly welcomed. If mappers wish to map other than the task instructions on HOTOSM that’s cool. That principle applies to whether they want to map single noded buildings and allow others draw the building shape, or wish to start harveasting Mapillary to tag buildings. What is needed by the “use cases” developed are details of precise distances between buildings, and the promimity of these buildings to roads, rivers and powerlines.
These suggestions are seriously good ones, and thanks again for them, please help out with validating the three large tasks that remain. Please show up the problem and what you think is the solution. Main point: don’t be afraid to say what you think is efficient, but understand that others may have different priorities.
Comment from dieterdreist on 11 January 2016 at 14:59
nice post. I agree with Malenki, please don’t advocate for nodes and parametric tags, use (reasonably noded) osm ways.