[Imports] NYC building + address import (was Re: Buildings & Address in Washington, DC, USA.)

[Serge Wroclawski]

On Thu, Jun 5, 2014 at 8:55 PM, Alex Barth <alex at mapbox.com> wrote:

> What hits _me_ the wrong way is how you present an import we are working
> hard on doing right in a pretty unbalanced light. This "we" includes you as
> you've invested a ton of time into this project too which I'm thankful for.

I did put a lot (not a ton, but a lot) of work into the NYC import
>
> I'd like to see a completely different attitude in interaction on this list.
> How can we make this work?

The issue I have with the approach that has been taken by MapBox is twofold:

1. Responsiveness
2. Cleanup
3. Honesty

The issue of responsiveness is straightforward. When a community
member finds a problem with how something is mapped and we go through
the speicifc steps outlined in the import process, and the individual
community members creating the problem are notified, I think there's a
reasonable expectation that they'll stop. Maybe they'd respond to OSM
messages, or respond to notes that they created, or respond to github.
My experience is consistently that with your mapper staff that they
simply don't respond to any of these. The only thing they've responded
to is DWG intervention (ie blocks).

That's a really huge hammer to have to bring down, but the alternative
is that there's bad data in OSM.

The second issue is cleanup, which ties very much into the first one.
There would be no big problem with waiting days and needing to contact
three or four people before getting a response, if the data didn't
stay bad. But instead, we see data that was put in badly and has
stayed bad. It's really a mess, which could have been fixed if the
attitude had just been to go a bit slower and when someone brings up
an issue, to take it seriously and not ignore it until days later
(importing with the problem in the meantime).


The biggest issue for me, though, is honesty.

I feel like you, Alex, are playing a game with the community. The game
is that you don't lie, but you skillfully omit. Reading the DC
proposal was therefore an exercise of "What is he not saying?", and
that's not in the spirit of collaboration.

Consider this... I still haven't seen an affirmative statement that
you're going to use paid mappers, yet the subtext is that this is what
will happen. If you're going to use paid remote mappers, just say so.
Just say "This is our plan." and let the community provide honest
feedback in response. I don't have a problem with paid remote mappers,
but I do have a problem when I feel that someone is trying to pull the
wool over my eyes.

The import process is supposed to be about collaboration and
consensus, and when important details that the community cares about
are omitted, it gives the appearance of a negotiation, rather than a
collaboration.

That is at the core of my frustration with the NYC import, and now the
DC import.

- Serge