[Imports-us] Portland Metro Area Building Import

[Rafa Gutierrez]

Serge,
Thanks for the comments and questions. I'm seconding Darrell's statements
here:

- Metro is on board with their data in OSM so I'm confident this will be a
manageable issue to work around if/when needed.

- This is an iterative approach in that we import 1:1 buildings first and
then import 1:M buildings on a more case-by-case approach where local
participants will have the most impact.

- Data deltas from the source: the script already accounts for buildings
already imported to OSM so when we run this again in a year or two, only
new buildings will be considered for import. Buildings removed should be
handled directly in OSM since these are fewer and likely to be spotted by
the community before it even makes it to the source data. I would add that
since we're in close contact with the providers, they've expressed interest
in making future updates seamless. They're really great to work with!


On Wed, Nov 19, 2014 at 7:53 AM, Serge Wroclawski <emacsen at gmail.com> wrote:

> On Wed, Nov 19, 2014 at 10:45 AM, Darrell Fuhriman <darrell at garnix.org>
> wrote:
> >>
> >> 1. The main thing that gives me some pause is the data being available
> >> under the ODbL.
> >
> > While the irony of this is not lost on me, I’m not sure that it’s our
> responsibility to worry about theoretical changes.
>
> While I understand this position (and frustration underlying it), my
> question wasn't entirely theoretical, so let me rephrase it in a more
> pragmatic way:
>
> If the data owners are setting the license for you, could they do so
> with a more permissive license that's future-proof?
>
> Otherwise, license changes are inevitable and we'd just be kicking the
> can down to the poor saps who'll have to deal with this down the line.
> Based on our level of committment to OSM, I'm afraid we're those saps!
>
> >> How will you differentiate between what is a Phase 2 section and what
> >> is Phase 1? Is there a process defined for this?
> >
> > Well, it’s simple. Phase one is the one building, one address situation.
>
> So it's about a dataset of 1:1 building correlations and then another
> dataset of 1:multiple and then handling it that way? So there will be
> some "holes" in the intermediate?
>
> I'm not suggesting that this is wrong, just want to explicitly
> understand what's being proposed.
>
> > Phase  2 is everything else: dealing with multiple address, multiple
> building situations. The exact plan for this is not yet determined, but it
> has been suggested that adding a number of address points inside of the
> building is the right way to start.
>
> I agree with this view.
>
> >  This may be sufficient for single buildings that have more than one
> address. For areas where an address point is potentially associated with
> several possible buildings, the tentative plan is to have ground-truthing
> expeditions to assign the address points appropriately.
>
> That would be awesome.
>
> >> You mention organizations which will help. What is the mechanical
> >> process you suggest for helping mappers perform this?
> >
> > I’m not sure I understand the question being asked here.
>
> It's at times like this when I truly regret living in Washington, DC
> for 16 years. I sometimes forget how to speak normal English.
>
> You have dataset A released today. In a year or two you'll have
> dataset B. There will be a delta between the two that will hopefully
> capture new buildings, removed buildings, new addresses, etc.
>
> I'm not even asking how to handle this delta from a community
> standpoint (though that'd sure be nice) but rather just "Have you
> tried to perform this delta and looked at the results?"
>
> - Serge
>



-- 
- Rafa Gutierrez <http://scavengeo.com/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/imports-us/attachments/20141119/2e432bf3/attachment.html>