[OSM-talk] Nominatim on the main page / Serges blog post
simon at poole.ch
Sun Feb 18 22:34:57 UTC 2018
Maybe it would be a good idea for everybody to take a deep breath and
count to 10?
For those relatively new to the project you should know that Serge has
been going on about these "issues" for years, some since I first had the
pleasure of meeting him. That doesn't mean that his points are more or
less valid, but the general sense of urgency due to a click-bait title
stuck above them is rather misplaced.
As to the points themselves, they are a mix of things from latent issues
that could easily turn serious (lack of man-power in some core areas),
over Serges pet peeves (nominatim doesn't do US junctions), to
trade-offs that he might disagree with, but are nonetheless not without
merit (for example layers, that now and then would be nice to have, but
are rather at odds with OSMs more holistic approach to modelling).
For now I just want to touch on the money side of things and employing
staff. The amounts available for speculative financial investment
naturally have nothing to do with operating a self-sustaining
(geo-services-) business. And if anybody doubts the difficulties of
doing the later, we just had the likely best ever funded geo-startup
that was literally able to throw money out of the window, miserably fail
when it was supposed to stand on its own feet and actually turn
geo-coding, map tiles etc, "over a free tier" in to a sustainable business.
Now the OSMF might have a slight advantage due to brand recognition, but
that doesn't change the fact that there is not very much money available
out there for such services and that such a pivot would come at the cost
of ruining the many legit small business that are small but sustainable,
and of which many contribute in one way or another to OSM software
development and operations.
All that said, OSM operations and core software development are
dependent on a very very small number of people, some of which have been
with us since the project started, we really need more hands full stop.
If paying for staff is the solution, difficult to say, what is clear is
that we would need a substantially larger continuous stream of donations
(think a pineapple fund sized donation every quarter) for any meaningful
number of operational staff.
In any case I'll be asking for help in a follow up posting on a
specific ops issue and expect to see lots of people volunteering :-).
PS: The limitation in Nominatims POI search is discussed roughly every
half year, well documented and understood, it would undoubtedly be nice
if it was fixed and if it is only to avoid the continuous surprise
re-discovery of it.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 488 bytes
Desc: OpenPGP digital signature
More information about the talk