[Imports] Import of addresses in Poland

[Johan C]

2014-02-04 Frederik Ramm <frederik at remote.org>:

> As you probably know, OSM doesn't value data over community. Where
> imports are concerned, we usually look at whether there's a local
> community to "digest" the data that has been imported, to fix it where
> it's wrong, to update it in the future.
>

I think a strong community is extremely beneficial to OSM. The attendants
of the workshop 'Road map to the future' at SOTM Birmingham wrote down
their following, diverse, dreams for OSM in 2020:

1. People: end-user

   - Useful car navigation in all first world countries
   - Be embraced by the open source software community
   - Have a business model that supports the OSM ecosystem and at the same
   time provides a good end-userexperience
   - Ease of use (data). My gran can create a map in 10 minutes
   - An easy and 'default' place for people to find options to a specific
   need
   - For the general public to understand the possibility of different map
   styles (and power it gives)
   - User selectable rendering
   - It should be open gl
   - To be able to select features much more dynamically than today. That
   the map can be a street map, orienteering map, cycle map or powerline map
   without having a specialized project making tiles for it
   - That people or organizations have the tools to make the maps they need
   using OSM
   - Enthusiastic embrace of multiple projections and warps. Warps to match
   OSM with historic maps that you don't want to distort, or diagrammatic
   distortions eg one-dimensional maps.


2. People: community

   - SOTM's everywhere
   - Local chapter growth
   - Community specific groups
   - Not a single map on osm.org (sign-up should make clear that OSM is
   much more than a single map like Gmaps)
   - Put a map of meetings on the frontpage, including conferences, SOTM,
   HOT, pub meetings.


3. Technique: editing/tools/quality

   - Focus on routing particularly during editing. Height, weight, turn
   restrictions (+ view)
   - Ease of use (editing). Mobile: place + label within 2 minutes. All
   maps have an 'edit me' buttonk
   - Easy for non-techies to add data
   - High quality data
   - Automatic collection of data from non-technical peoples devices to
   enhance quality (people not interested to be mappers)
   - Consistent tagging format or rules (described)
   - No federated tagging / worldwide consistency / no federational mappers
   - Polygons on as polygons. Treat them as their own type
   - Customized tools for interest groups (hike, tree, walking)
   - Ability to move on from poor initial tagging conventions
   - Niche/long tail mappers/users: a. Custom map display showing a1. Task
   based custom editor a2. Stored in global OSM DB (problems)


4. Competitive advantage

   - OSM should be the default map everywhere
   - OSM gps units
   - All imagery should come directly from DigitalGlobe in the highest
   quality and accuracy and cut out the "middle men" aka Micro$oft, Google,
   Govt etc.


What is the right way to get to Rome (that is in my opinion: lots of
end-users and a big 'alive and kicking' community in each country)? I would
prefer to have thousands of people per country helping to improve OSM.
Preferrably by ground truth editing. But in ten years of OSM no country has
created a community large enough to create a reliable database of addresses
(and POI's) by ground-truth editing. Is going on with addresses like in the
past ten years the best and only approach to travel to Rome?

I think OSM should take open data as a chance. And use as many local
community members as possible to import this data. But just having lots of
data in a database is not enough to get to Rome. Maybe Steve Coast will hit
the right chord and figures out a faster way to get to Rome with the combi
Telenav/Skobbler, and hopefully builds an easy-to-use app which encourages
end-users to improve the OSM database by ground truth editing...

Cheers, Johan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/imports/attachments/20140204/f8ac9516/attachment.html>