[Talk-GB] Fwd: Re: Open data (Was: Parliamentary debate mentions OSM)

Richard Fairhurst richard at systemed.net
Wed Mar 30 07:14:01 UTC 2016

Rob Nickerson wrote:
> Is it OK to leave it to the data users to merge the open data with OSM 
> or is that burden too large for them to bother (at which point the 
> pressure of OSM in the UK reduces)?
> The reason I ask is because I don't have the answers. Hoping some 
> of the data users on the list may be able to suggest a point where 
> the burden would become too large.

It's absolutely fine to let the data users merge the open data. Not least
because you don't know, and shouldn't second-guess, _how_ they want to merge

I use lots of UK open data on cycle.travel: built-up area outlines from OS
OpenData, AADT traffic flows from dft.gov.uk. It's not at all difficult to
integrate either of them into my OSM-based database. But I do it in the way
that makes sense for my uses. If the data were integrated into OSM, chances
are it would be done differently in some way (for example, the extrapolation
from AADT survey points to B roads would be different) and therefore less
suitable for my needs.

People sometimes underestimate how lovely government open data[1] is to work
with. It's consistently attributed, it comes in a file format everyone
understands, it's layered, and best of all, it has consistent coverage. OSM
is none of those. I love OSM dearly, obviously, but all my data-processing
headaches are OSM-based.

Concentrate on what makes OSM good and unique, not in redistributing a
poorer-quality, harder-to-use mirror of other datasets.

Signed, a data user.


[1] except TIGER. And that was the one that got imported. Gaaaaaaaaaaaaaaah

View this message in context: http://gis.19327.n5.nabble.com/Open-data-Was-Parliamentary-debate-mentions-OSM-tp5870653p5870931.html
Sent from the Great Britain mailing list archive at Nabble.com.

More information about the Talk-GB mailing list