[Talk-us] OSM Data Quality

Martijn van Exel m at rtijn.org
Fri May 31 21:21:51 UTC 2013

As already noted, quality is in the eye of the beholder. That said, there
are some objective quality indicators such as positional accuracy,
completeness, resolution. I summarized this in a paper a few years ago from
another source, where I also introduced the notion of 'crowd quality' in an
academic attempt to capture specific quality considerations for
crowdsourced geospatial data:

Not much of an academic, I later picked this up in a more pragmatic manner
to create the notion of data temperature I presented at SOTM US 2011:

Someone else mentioned we need more mappers. There is truth to that but we
also need to care about building out the community: how do we reduce the
churn rate, or in other words how to keep mappers involved and motivated to
continue mapping? how do we nurture the power mappers, those 5% who create
80% or more of the map data - especially in light of the large amounts of
new mappers coming in? and finally how do we make local communities work?
Latter is super important because great local data (transit, businesses,
addresses) is key to the usefulness (hey, another way of thinking about
quality!) of OSM. Great local data is something you only get if folks who
know a place, folks with different interests and from different walks of
life, work on the map together. Currently that happens in too few places. I
think one of the most important keys to making good OSM data great lies in
figuring out how to build strong local communities. In Europe, we have that
down.It all started with that. Get together and map. Have fun, figure it
out together. While traveling in Germany recently, I did not have to go
online once to find my way, my hotel, restaurant, bus stop etc. The map is
*that good*. Sure, there are more mappers per sqm there. But it is just as
much about people getting together, motivating each other, collaborating on
more complex mapping tasks (stuff like transit relations[1]). We have a
long way to go still in the US, and we may need a different approach than

I think I just wrote half of one of my SOTM US talk. Thanks Frederic ;)


[1] http://www.overpass-api.de/api/sketch-line?network=VBB&ref=M1&operator=

On Fri, May 31, 2013 at 2:28 PM, Richard Welty <rwelty at averillpark.net>wrote:

> On 5/31/13 3:15 PM, Frederic Julien wrote:
>> Dear all,
>> I'm working on a presentation and interested to hear your thoughts. What
>> are the top 2-3 changes that could improve OSM data quality? That could be
>> processes, tools, methods, training, peer review, attributes, etc.
>>  at one level, i agree with Clifford Snow's comment that first you need
> to define data quality.
> at another level, i think that we can talk about the following:
> 1) consistency in tagging. editor improvements, better documentation,
> better
>     training materials can all help with this
> 2) improved processes and controls for data import (this is work that is
> happening
>     on the US import committee). there are a lot of imports of the past
> that suffer
>     from Quality Control issues, and lots of imports that never should
> have been
>     done because of problems with the data quality.
> 3) in the US (and you did ask on talk-us), identifying and dealing with
> the shaky
>     Tiger data from the 2007 tiger import. some of this has been done, but
> it's an
>     ongoing effort and is one of those things that is easier to say than
> it is to do
> richard
> ______________________________**_________________
> Talk-us mailing list
> Talk-us at openstreetmap.org
> http://lists.openstreetmap.**org/listinfo/talk-us<http://lists.openstreetmap.org/listinfo/talk-us>

Martijn van Exel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-us/attachments/20130531/477cb8fc/attachment.html>

More information about the Talk-us mailing list