[Openstreetmap] Process and scaling up

[Hugh Barnard]

Hi Folks

I have been musing about this:

> my thinking was that people would upload to a web site, which would
> validate the file against the schema for gpx, do a number of basic
> checks on the usability (i.e: lack of noise and fubar'ed points) of
> the file and then import it to the database.
>

Roughly, we find some volunteers or occasional helpers etc.  The whole
process for
contributing and annotating needs to be pretty accessible:

X takes a GPS bike ride, gets some points, uploads and fills in a TIGER
coded annotation form,
for example. This could duplicate/validate/contradict, partially or wholly
some other tracing or update it..There are tens of Xs and Ys all
enthusiastically doing this (happy days!)..The 'unit' of tracing varies from
one street to tens of streets on each side of the city..

This is just a thought experiment about scaling it up to London wide and
scaling up the volunteers (including people that would help but are less
into computers/mapping etc.).

I don't have any good answers but I have a sneaking feeling that all the
'raw' tracings should sit by themselves intact somewhere so that we can
recombine them in different ways/try different processing experiments. For
example, an algorithm could drive (well, cycle, let's keep it green) through
and create an empty  'slot' for street information each time it felt it had
turned a corner..

Best regards Hugh