[OSM-talk] oneway=1

[Arieh Skliarouk]

On 10/3/07, graham <graham at theseamans.net> wrote:
> Arieh Skliarouk wrote:
> > 99% of OSM users are primarily
> > interested in getting around by car.
>
> Or are you just assuming that osm users are identical to tomtom users?

Ok. Here we are at something.

I see OSM as the "opensource" version of proprietary offerings. OSM
users are not less, than users of proprietary systems.

Ideally OSM users would include more users, as there are much more
allowed usage scenarios.

Ah, and as most users are interested in getting from point A to point
B by car, this is the usage scenario we should develop most. Run poll
if you think I am wrong. (I would do it myself, but I don't have
access to the OSM website code).

> > What a waste... time of the bike users can be much better spent on
> > markup and postprocessing of the GPS-traces.

> Personally, I would much rather spend my time on a bike than in an
> office doing markup and postprocessing, which are jobs for a computer
> not a human being ;-)

Excellent point! You just expressed ideal contribution procedure for
OSM user: drive around, not necessary related to mapping, occasionally
answer questions on PDA when it stumbles on something. This is what I
am keep repeating here again and again: deduction, deduction,
deduction, deduction.
like Steve Ballmer here:
http://video.google.com/videoplay?docid=6304687408656696643

:-)

> I also don't see the point in choosing our priorities to match those of
> tomtom etc: isn't one of the reasons for a community project that what
> the community creates may be quite different from the commercial
> alternatives?

I don't care about proprietary stuff and their priorities.

I want something I can run on Linux, post map on on my website or send
to friend, fix mistakes, add and share my points of interest. Care to
match that with proprietary stuff?

> > Just think, couple taxi drivers [compensated correspondingly] with
> > GPS-tracer turned on and SD big enough in two days will cover area
> > that will take bike tracer year to trace.

> I don't think so. You will end with lot of points at taxi stations, a
> lot of big roads

Excellent!
Now use an algorithm to convert the points to nameless continuous
links, eliminate duplicates, deduct road direction, add questionable
tag "public transport only", upload to server, propagate to
pda-clients so the users will add tags once in vicinity.
Voila! You just saved yourself (and many many users in many many
cities around the world) lots and lots of resources (time, fuel, food)
and [potential] accidents:
http://wiki.openstreetmap.org/index.php/Mapping_accidents

> but almost no small streets. A bike surveyor takes notes about
> street names and often aims at completion. Much better.

I, as OSM user, prefer to have 95% of the city outlined in a week, so
I can fill necessary details as I go, instead of waiting for some bike
surveyor to map 100% in two years.

Oh, and one small question: where do I get bike surveyor for my city
(population 10.000.000) please?

-- 
Arieh