[Imports] Value of imports in general

[Martin Koppenhoefer]

2014-02-07 12:16 GMT+01:00 Christian Quest <cquest at openstreetmap.fr>:

> If an import results has the same data structure and quality as any manual
> contribution, what's the problem ?
>


I hope this should have been quality as many _good_ manual contributions,
because we have a lot of not so optimal manual contributions by newbies
where the quality is tolerable for several reasons 1. beginners obviously
make errors, but they are usually limited in quantity and extension, and
probably they will be later corrected by the same mappers as they go on
learning how to map in osm
2. even if not optimal they might add valuable information that someone
else can later correct (given that this is only punctual information and
not loads of loads of data)
3. accepting everyone's contribution is psychologically important (lowers
the bar to contribute, makes people more confident for future edits, is
motivating)


the same quality we do accept from a newby would probably be rejected in an
import, as it does make a huge difference to tag one object differently to
how we are used to, and to make the same mistake hundreds or thousands of
times in an import (distorts taginfo stats, gives a bad example to other
mappers in the area, etc.)



>
> If it's not, it looks like a bad import to me and I agree that its value
> may be considered as negative (more work to fix it).
>


also if it isn't fixed very quickly it will teach newbies in the area
"wrong" tagging / mapping, it will draw a bad picture about OSM in the area
and might lower overall reputation, it will diverge tagging and puts the
whole project at risk.

cheers,
Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/imports/attachments/20140209/a2779b2d/attachment.html>