[OSM-talk] The world’s best addressable map

Thu Oct 23 22:59:32 UTC 2014

2014-10-23 8:57 GMT+02:00 Marc Gemis <marc.gemis at gmail.com>:

> First question: how do you build "the world's best addressable map" ?
> Through imports or survey ? Is the timeframe that Martin puts forward fur
> surveyed addresses acceptable and feasible ?
>
> As for the imports:
>
> Wouldn't it be great that we would have import tools that can be used by
> many groups ? Just as we have great editors for "manual" mapping.
> Shouldn't the OSMF (or however) invest in a procedure + tools for import ?
>
> So when a group wants to perform an import, they only have make sure the
> license is compatible and the data has enough quality. The process of the
> import is already described, the tools ( task managers, data conversion,
> some quality checks etc.) are in place. They deliver the data and the
> import can start.
> I have the impression that right now each group has to take all those
> obstacles all by themselves.
>
> Maybe I have a totally wrong perception of the import process. please
> correct me if that's the case.
>
> regards
>
> m
>
>
I partially share your impression that each group has to take obstacles by
themselves. Since OSM is a social project I think it's good that people
with experience on (importing) addresses and buildings share their
experience. I see this happen on the @import list, user diaries etc. The
somewhat scattered knowledge around could however be shared more
centralized to make it a bit more accessible, so I created a page on the
wiki (http://wiki.openstreetmap.org/wiki/Importing_buildings_and_addresses)
in which anybody can share their knowledge (I will do so this November).
Furthermore, SOTM Buenos Aires hosts 2 presentations on addresses.

I don't see manual work for OSM as an obstacle. Any import (or survey :-) )
requires some manual work because the current map is far from blank, so the
almost full-automatic import you describe is IMHO not possible. Though
semi-automatic assistance by tools (like the wonderful Geofabrik inspector
address view) help a lot keeping the work fun.

As for tools: it would help when more tools for importing and updating
could be programmed. As an example: I'm not aware of any tool which makes
it possible to semi-automatically compare building outlines to the outlines
of an updated government database, highlighting major differences in
geometry.

On the role of the OSMF: I think it shouldn't be the OSMF investing in a
procedure + tools. It would already be great and an energizing factor when
(members of) OSMF would express support for having more address and
building data in OSM, be it either from surveys or from imports.

Cheers, Johan

>
> On Wed, Oct 22, 2014 at 8:28 PM, Clifford Snow <clifford at snowandsnow.us>
> wrote:
>
>> Steve Coast suggested changing the mission statement of OSM to be
>> something like “The world’s best addressable map” I thought
>> I'd separate out this into a new thread. As some backup, I attended my
>> first SOTM-US in 2012 where Steve proposed we build an addressable map. A
>> group of us from Seattle decided that we'd take that challenge and import
>> addresses for Seattle. That import was completed.
>>
>> Here are the address related comments:
>>
>> On Wed, Oct 22, 2014 at 4:37 AM, Martin Koppenhoefer <
>> dieterdreist at gmail.com> wrote:
>>
>>> 2014-10-22 12:15 GMT+02:00 Steve Coast <steve at asklater.com>:
>>>
>>>> Together, we could do this in 6-12 months and finish addressing in 1-3
>>>> years. At that point we wouldn’t have just made the world slightly better,
>>>> we would have put a big dent in the universe. Nobody would use a closed map
>>>> ever again, and it would be people like you that made it happen.
>>>>
>>>
>>>
>>>
>>> I agree with you that addressing is very important for a lot of
>>> commercial (and non-commercial) map users. What I don't understand is how a
>>> paid board would help us map more addresses. Unfortunately mapping
>>> addresses is typically less fun than going to the video arcade. Looking at
>>> the current figures we are not doing too bad. Currently there are 130
>>> Million buildings in OSM and 46 Million housenumbers. I don't know exactly
>>> how many buildings there are in the world, and how many of them don't have
>>> addresses, but I guess it will be at least 1 Billion addresses in the
>>> world, probably more. According to the stats page we have roughly 25.000
>>> active contributors a month (
>>> http://wiki.openstreetmap.org/wiki/Stats#Contributor_Stats ). To get an
>>> address on all currently mapped buildings in three years time, (84M to go),
>>> every active contributor would have to add 93 addresses a month -
>>> constantly. To get 1 Billion addresses mapped by 25.000 contributors in 3
>>> yrs, it would be 1111 housenumbers a month per active contributor. Are we
>>> planning to pay the mappers as well? The only solution seems to get more
>>> contributors mapping, and have them insert addresses.
>>
>>
>> On Wed, Oct 22, 2014 at 4:59 AM, Marc Gemis <marc.gemis at gmail.com> wrote:
>>
>>> It would be nice to know how many of the buildings and house numbers in
>>> OSM were imported versus surveyed / drawn by hand. I have a bad feeling
>>> about how feasible it is to crowd surf house numbers.
>>
>>
>> On Wed, Oct 22, 2014 at 5:04 AM, Oleksiy Muzalyev <
>> oleksiy.muzalyev at bluewin.ch> wrote:
>>
>>> It is not necessary to put down a number on each building. It is
>>> possible to use *addr:interpolation* (*odd, even*, or *all*).
>>>
>>> We put down a number on the first building, then on the last, connect
>>> them in JOSM, and add *addr:interpolation: all *. For example here:
>>> http://osm.org/go/0CFn0AZ_d--?m= . It is also very useful on a street
>>> with many small houses. And it is searchable. For example if there is
>>> number 15 and number 27 on the map for a street, and they are connected
>>> with *addr:interpolation: odd, *and if one searches number 21, the map
>>> will show the number 21 all right.
>>>
>>> Then, there is another approach. We first map addressable large
>>> building, where a lot of people live or work. Kind of of going after the
>>> low-hanging fruit.
>>
>>
>> On Wed, Oct 22, 2014 at 5:06 AM, Matthijs Melissen <
>> info at matthijsmelissen.nl> wrote:
>>
>>> On 22 October 2014 12:37, Martin Koppenhoefer <dieterdreist at gmail.com>
>>> wrote:
>>> > Currently there are 130 Million buildings
>>> > in OSM and 46 Million housenumbers.
>>>
>>> Do we know how many of these addresses come from imports? I wouldn't
>>> be surprised if over 90% of the housenumbers in OSM come from imports.
>>>
>>> The Dutch BAG import accounts for 8 million adresses, and the Czech
>>> RUIAN import accounts for 3 million addresses. Then there have also
>>> been large imports at least in Germany, Poland, and France, but for
>>> these countries I can't find exact numbers.
>>
>>
>>
>>
>>
>>
>> --
>> @osm_seattle
>> osm_seattle.snowandsnow.us
>> OpenStreetMap: Maps with a human touch
>>
>> _______________________________________________
>> talk mailing list
>> talk at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk
>>
>>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20141024/b8300564/attachment.html>