[OSM-dev] Server-side data validation

Paweł Paprota ppawel at fastmail.fm
Sat Jul 14 07:01:45 BST 2012

Hi Eugene,

> (...)
> So using Wikipedia as an example, OSM actually already embodies the
> "wikiness" of Wikipedia.

I agree and obviously it works well. However, what I meant with
"citation needed" for example is not only at the syntax level (FIXME tag
as you mention) but first and foremost at the _semantic_ level.

Meaning that Wikipedia has no way to automatically validate the content.
At the end of the day it is all a bunch of text, images, category
structure - how to check this against the objective facts? No (easy)
way, it all needs manual peer review. And of course that it at the heart
of Wikipedia and it works great.

However, I think OSM can go one step further. The difference to
Wikipedia is that geo data is structured and at some level it is
possible to do automated QA.

Example: OSM could have reference data for biggest lakes in the world
and use this data to QA its own data. Wikipedia has no way to enforce
the facts here - someone can write on biggest lake's page that it is
actually the smallest one. In OSM it is possible to precisely calculate
how large is the lake in OSM and compare that to the reference data.
Then _use_ this data - visualize, validate, notify etc.


More information about the dev mailing list