[HOT] Highway=residential in Africa

Laura O'Grady laura at lauraogrady.ca
Wed Jun 15 19:35:38 UTC 2016

Hi all,


Just jumping in here with some ideas based on my own experience in patient/public engagement. Bear with me as I work through my idea from health care in an attempt to align it with HOT efforts. 


I worked on a research study that engaged people with diabetes on their use of online message forums for exchanging health information. We were interested in how patients tagged the content of their posts, i.e., what terminology was used by people with diabetics as opposed to what IDC-10 ( <http://apps.who.int/classifications/icd10/browse/2016/en0> http://apps.who.int/classifications/icd10/browse/2016/en0 or MeSH ( <https://www.nlm.nih.gov/mesh/> https://www.nlm.nih.gov/mesh/) would use. In other words we wanted to know the layperson taxonomy as opposed to the professionally derived system. 


>From my experience with OSM I know that the origins of tagging system is tied to the beginning of OSM, it has evolved over time based on user contribution and any new tag suggestions are voted upon. I also know there are regional disparities in the tagging system (terms that are common in one geographic region are sometimes not as applicable in others). 


Since I’m new to HOT I’m not entirely sure about how much local knowledge is integral to the map mapping effort in a crisis scenario. I do know from reading some of the recent publications that locals often provide valuable information about colloquial names for locations, roads, etc. I understand in this instance a specific wiki was created with instructions for mappers. I wonder if there is any value in asked the community what tag(s) are best used in this context? Or is this not relevant/useful since the target audience for the maps will be those that needs tags that are more in line with overall OSM taxonomy and traditions?


Thanks for any clarification.




Laura O’Grady

laura at lauraogrady.ca <mailto:laura at lauraogrady.ca> 




From: Michael [mailto:ohrosm at gmail.com] 
Sent: June-15-16 6:23 AM
To: hot at openstreetmap.org
Subject: Re: [HOT] Highway=residential in Africa



as I have struggled with this myself in the past I would like to comment from my point of view:

2016-06-14 23:51 GMT+02:00 Alex Bogedain <alex.bogedain at gmail.com <mailto:alex.bogedain at gmail.com> >:

In terms of the Unclassified/Tracks/Paths debate, in these rural areas especially, I believe tracks should almost always win out.  


I am afraid that I disagree, even though I mapped a lot of roads as tracks in the past. You need to make sure not to be mislead by the word 'track' but how 'highway=track' is defined in OSM. To me this is about ways not for standard traffic but more agricultural/forestry uses. But as a non-native speaker I probably have the small advantage of not directly associating a specific definition with 'track'.


I say this because if these indeterminable lines pass by and connect houses/huts or fields, then the road type is clearly not unclassified, as it has a classification of connecting residential or agricultural areas. 


Here the word 'unclassified' is pretty misleading as well. For some reason 'highway=unclassified' does not mean that the road is not classified at all but does not have any special classification. In other words it is something like "other road" or any road that is not tertiary or above. 


Further, it is indistinguishable for us at a distance to determine between tracks and paths because one day a family could get a motorbike and start traveling what could have been considered a path with a motorized vehicle, or anyone for that matter could drive down the path/track on a motorbike to access the houses or fields.  From my understanding, out in rural areas, especially Africa, there are no controls on the modes of transportation as transportation is limited to what any person is able to acquire within their means.  Going forward into the future this means that the modes will be changing as more families get motorbikes or 4 x 4 vehicles, thus, classifying a road as a track covers all the bases for now and moving forward.  The local population will be the ones to make the distinction and things will change as time progresses.


'highway=track' is mainly about what the road is mainly used for, less what vehicle can be used on it. In Germany there is such a thing as 'highway=track, surface=paved' - but other than 'highway=unclassified' those tracks usually have sign limiting their use to agricultural use only. This is not really intuitive but that's just the way things evolved in OSM.


To wrap up my thoughts on this, when viewing OSM (not logged in for editing) residential and unclassified roads appear the same, but tracks appear differently as how unmaintained roads would appear on nearly any map around the world.  In terms of the PMI and other humanitarian initiatives, it would be disingenuous to classify these narrow roads as unclassified as the users of the map should be ready for a track, or possibly something untraversable for their vehicles.  If we are to map in terms of an areas actual use of roads, to my understanding unclassified would not cover these types of roads and would not describe them accurately at all for actual drive-ability.


Well, as a general rule within OSM the tagging should not based on how current maps render things but rather on the actual situation. In other words, for example mapping a road as  track to get a dashed line is not a good idea. If there is a need to display this differently on the map, the rendering should be changed, not the tagging. 'Tagging for the renderer' in fact is something that is not really appreciated in OSM. 




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/hot/attachments/20160615/85730fc8/attachment.html>

More information about the HOT mailing list