[Imports] Buildings & Address in Washington, DC, USA.
filbertk at gmail.com
Sat May 24 14:52:18 UTC 2014
On Sat, May 24, 2014 at 9:38 AM, Serge Wroclawski <emacsen at gmail.com> wrote:
> I have a number of concerns about the import proposal.
> Firstly, your goal is stated to add ~66,000 buildings and addresses to
> the OSM map. This builds on the work of the previous building import
> in DC which was never completed. There is nothing in this proposal
> about the existing buildings and addresses, about any conflation or
I would have a task for all the census block groups in the city.
1) validate / conflate / compare new dataset with osm data
2) if needed, import new buildings / addresses
3) validate again.
I don't think there is a need to delete the current osm buildings and
advise against that.
http://data.dc.gov/Metadata.aspx?id=59 is dated from 2011 and in some cases
OSM may be updated to be more current.
the GIS data has a CAPTUREYEAR field, so we can identify any that are newer
and import those in cases the GIS data has new buildings. Most will be
older and same as what's already in OSM.
This concerns me, as it means that part of the city will have new
> data, part of the city will have old data, and there is no plan to
> correct/update as part of this process.
> Secondly, your schedule is far too agressive. Giving the OSM community
> two days to review your plan is simply not reasonable. You must
> provide people the opportunity to review your proposal, ask questions,
> review your data, review your process, get answers, review those
> answers. I would say that two of meaningful discussion would be a
> miniumum that it would be possible to do this in, not 48 hours.
agree. It would be good to have the buildings added soonish, but there is
no deadline or need to rush.
> I haven't even had a chance to get to the data yet.
> Thirdly, this proposal doesn't mention who will be doing the import
> work. I suspect that based on the supporters that this means that the
> import will be done using paid staff. Is that correct?
I don't think that's a good idea, imho.
If it's not
> correct, what plans/process will you use for working with the local
> community to do this work? If it is correct, and you will be using
> paid staff, based on the tremendous problems that this caused NYC on a
> multitude of fronts, what steps will you be taking to address the
> problems of oversight, of integration, etc. that we had in NYC?
It looks like for NYC imports, they simply deleted existing buildings and
added the new ones. This includes areas that I surveyed on the ground,
ensuring POIs lined up with the buildings correctly, updating for
demolished / new buildings etc.
To be honest, that is quite discouraging and demotivating to continue
contributing to OSM. The same thing ought not happen for DC, if we don't
want to alienate the community.
I think the community can handle this. Main steps are to get the tasks
setup in tasking manager and have the scripts (which probably need
review) available, then the community can take it's time (as needed) and
care to properly integrate new data.
I look forward to having this data finally complete in OSM and happy to see
interest in this.
> This ties directly into your aggressive schedule. How do you know the
> work will only take until June 30th?
> Fourthly, I'd like to know if there is any contact/money/services
> exchanging hands that is in any way connected with this import. This
> isn't part of the guidelines, but based on the supporters explicitly
> listed, I think this is a question that the community deserves to know
> the answer to.
> - Serge Wroclawski
> Imports mailing list
> Imports at openstreetmap.org
filbertk at gmail.com
@filbertkm / @wikimediadc / @wikidata
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Imports