[OSM-talk] urban surveying speed - can it be estimated aspopulation per hour?

[David Earl]

> -----Original Message-----
> From: talk-bounces at openstreetmap.org
> [mailto:talk-bounces at openstreetmap.org]On Behalf Of Christoph Eckert
> Sent: 06 June 2007 22:42
> To: talk at openstreetmap.org
> Subject: Re: [OSM-talk] urban surveying speed - can it be estimated
> aspopulation per hour?
>
>
> > What productivity do you think can be achieved when surveying an
> > urban area? ie, if you know the population of a place can you then
> >  estimate how many hours will be required to completely map it?
>
> I doubt there's a linear correlation between the amount of inhabitants
> and the time necessary for mapping. It much more depends on the
> topography, the area, the overall length of streets, cycleways etc.

Actually, in the work I've done I've been suprised to find a very close
correlation, to the extent that I can now predict how long I need to allow
to map somewhere.

Not counting the time to get to and from the start of the survey, I reckon
an urban area (village upwards) takes me roughly 45 minutes per 1,000
population, and the same again on JOSM, i.e. one and a half hours per 1,000
in all. Time doing the survey is almost always the same as building the
graph and uploading the results.

Cambridge took me 150 hours in all and is somewhat over 100,000. Ely, which
I've almost finished now, is 15,000 people and has taken somewhat over 20
hours so far, 10 hours surveying and 10 hours data processing.

This is on a bike, using a dictaphone, and mapping to a consistent level of
detail which includes street names and features like schools, post boxes,
pubs etc.

David