[OSM-talk] urban surveying speed - can it be estimated as population per hour?
david at frankieandshadow.com
Thu Jun 7 13:03:40 BST 2007
> > Given the current level of activity, and rate of growth, the
> whole project seems to be
> > almost feasible ;)
> it almost seems that way! but let's not get carried away - it
> wouldn't do to sell short the big map data providers ;-)
> anyways, thinking forward, do you reckon:
> (i) the rate of road addition will go pretty linearly up to 100%
> completion of road network and then stop (presumably, then, with
> ongoing addition of more detailed attributes and points of interest); or
> (ii) we'll start hitting a level of completeness where it's
> difficult to discern what's completed and what's not, and the
> last mile(s) will be by far the hardest? at what % completeness
> would we start seeing that?
> or do you see some other scenario? what ways could we mitigate
> any tail off as completion is neared?
I agree entirely. I think the biggest problem we have is that there is no
guidance as to completeness, but between us we probably do know that
information, we just don't have a way to represent it.
For example, I left two ends of a street disconnected recently. I now see
they've been (correctly) joined up, but the reason I left it was because I
hadn't been along it so didn't know whether there were any turnings off the
missing section (or rather, I did know that there is at least one which I
saw in the distance, but had run out of time to complete that day). Now the
street is joined up it looks like it is complete, but I know it isn't. But a
visitor would assume it was because the density of information makes it
When a skeleton of streets is added, it's pretty obvious there are gaps, but
when detail is added, but not systematically, it is very hard to tell.
So, I still think one of the most important things we could add to the
mapping is boundaries and areas of unmapped ground, the here-be-dragons
concept. We then push back these frontiers of the map as we add more.
(BTW now I can identify the person who made the change above, is there a way
to get in touch with them if they aren't on the mailing list and don't have
a wiki user account, as is apparently the case here? - I don't want to tread
on their toes if they are going to fill in the gaps I left, especially as it
is a train ride away for me, but I did want it to be completed
Secondly, things change. I've been wondering how to cope with this in
Cambridge. There are some new developments which I know will need mapping,
but things that have arisen since I visited an area are more problematic.
Maybe the number of people involved will rise to a point where there is
someone closer to changes than me, but it is not quite like wikipedia, since
it needs a site visit to know that something needs changing.
Unless, maybe, some local authorities in the UK at least might be prepared
to co-operate. Each council has an officer who is responsible for naming new
streets and notifying the Post Office new addresses in need of post codes.
If we could get on their notification list, then we would at least know
there is a street needing attention and can arrange to visit it at some
point. I think I'll try making contact locally.
More information about the talk