[Imports] Fwd: City of Seattle imports

Wed Dec 5 04:46:10 UTC 2012

Whoops! Should have cc'd imports@

---------- Forwarded message ----------
From: Jeff Meyer <jeff at gwhat.org>
Date: Tue, Dec 4, 2012 at 5:40 PM
Subject: Re: [Imports] City of Seattle imports
To: Serge Wroclawski <emacsen at gmail.com>

Serge - apologies for the confusing formatting. I use Gmail's web client.
It's not the problem. I human formatted the message to look nice in Google,
forgetting that the listserv software doesn't do well when people put in
their own (poorly-chosen) bullets. The general format was:
[your comment]
\>\> [my reply]
\>\>\>\> [subcomment of mine to my own reply]

Many of your responses are based around assertions about what OSM and what
OSM is not, and what's important and what's not, but I'm not sure those are
relevant to the topic at hand. I'll focus on responses relevant for the
import of City of Seattle data. My goal is to help create better
documentation of best practices for imports. If there's a document like
this that I've missed, please let me know.

SW: Those scripts and datafiles need to be part of any discussion, as early
as possible.
JM: We identified these files and developed preliminary scripts a couple
days ago.
Here are a couple - these may not be the only sources considered for import:
https://data.seattle.gov/dataset/Master-Address-File/3vsa-a788
https://data.seattle.gov/dataset/Street-Network-Database/afip-2mzr
https://data.seattle.gov/dataset/2009-Building-Outlines/y7u8-vad7

SW: Manually in that you will be adding all features by hand, you, yourself?
JM: I'm not sure why you are asking if I'll be doing this myself. This
email thread contains several references to people who will be assisting.
Cliff Snow and others attending the meeting to discuss this import. We hope
to have a team of local mappers focus on imports where they have local
expertise.

SW: What long time, trusted, local OSM community member do you have working
with you on this?
JM: I'm not sure who decides how long is long enough or who is trusted.
Right now, Steve Coast is planning to attend our planning meeting & will
hopefully be reviewing our plans. That said, a good import process should
be well documented, stand on its own, and not be dependent on this type of
participation.

SW: Discussion is better than notice, and imports isn't the greater US
community
forum, talk-us is.
JM: Perhaps I miscommunicated what I meant by "notice" - I was hoping to
start an email discussion by giving notice of our intent to commence our
import planning with the e-mail to imports@

Thanks,
Jeff

"Be nice." - Dalton, Road House

On Tue, Dec 4, 2012 at 4:21 PM, Serge Wroclawski <emacsen at gmail.com> wrote:

> Jeff,
>
> I've read this mail over several times, and simply find it hard to
> parse. I don't know what email program you use, but standard email
> in-line reply is > for a quote and >> for a quote of a quote. The
> style you've adopted makes it hard for me to identify whose quoted and
> in reponse to what
>
> For more information on in-line quoting, see
> http://en.wikipedia.org/wiki/Posting_style#Quoted_line_prefix
>
> On Tue, Dec 4, 2012 at 6:53 PM, Jeff Meyer <jeff at gwhat.org> wrote:
>
> > source tags are not really appropriate for indivdual elements,
> >>> How do you identify when individual elements of the same way or node
> are
> >>> derived from different sources?
>
> One of the key important elements to understand about OSM is that OSM
> is not a collection of sources, but is in fact a single unified
> dataset.
>
> Except in a few exceptional cases (which I'll address later) we do not
> care that a features comes from one person or one source and not
> another.
>
> In fact, what's been shown is that "official" sources are a problem,
> as people are hesitant to edit them, where we want people to be not
> just be practically, but encouraged to edit. But when something is
> stamped with "official" or from an external dataset, they're less
> likely to.
>
> Now, there are times when you want to know what source something came
> from, and this is where changeset tags are far more useful.
>
> For example, let's say there's a road, and someone adds it from some
> local datasource. Someone else may look at the feature in imagery and
> change its geometry. In that case, we'd like to know what imagery they
> looked at.
>
> But in OSM all of these techniques are considered secondary. Our
> primary and always preferred method for data are people surveying. We
> would always prefer, and should always favor a person on the ground
> against other sources, and surveying doesn't have a "source".
>
>
> >>>>> I didn't see any discussion or dispute of this approach on the wiki
> >>>>> page, so I'm assuming it's ok.
>
> This is a bad assumption. It's heavily debated, and largely deprecated
> in favor of changeset tags.
>
> I realize I addressed this above, but please, stop adding >>>>> to
> your mails, this is odd and confusing.
>
> > more importantly, the specific
> > datasets should be referenced, as data.seattle.gov doesn't talk about
> > the version of the data, or even what dataset it is.
> >>> Good points. We will add these clarifications.
>
> > Where is the source of your transformation scripts?
> >>> From the email above: "some translation instructions Cliff has put
> >>> together." We will include the specific translation code either at a
> github
> >>> page or on the http://wiki.openstreetmap.org/wiki/Seattle page.
>
> > Where are the specific data files you're transforming?
> >>> They are at data.seattle.gov & we will provide links to the sources.
> We
> >>> will also consider posting separate snapshots of these source
> datafiles if
> >>> we can figure out where to host them.
>
> Those scripts and datafiles need to be part of any discussion, as
> early as possible.
>
> > How will you handle object conflation?
> >>> Manually and methodically.
>
> Manually in that you will be adding all features by hand, you, yourself?
>
> >>> Please identify any un-discussed concerns. I'd like to make sure we've
> >>> accounted for them all.
>
> Imports should generally not be done by those who do not have a long,
> long history with OSM. Bad imports are not simply the same as bad
> edits, but have long, multi-year consequences that communities
> sometimes never dig out of.
>
> What long time, trusted, local OSM community member do you have
> working with you on this?
>
> > but more generally, you seem to not have discussed this with the greater
> OSM
> > community, including the talk-us list.
>
> >>> I thought that imports@ was the "Greater OSM community."
>
> See above.
>
> >>> From the email above: "A group of Seattle OSM'ers (led by Cliff Snow)
> are
> >>> getting together to discuss the methods for importing and
> double-checking
> >>> the information.
> > http://www.meetup.com/OpenStreetMap-Seattle/events/93524692/
>
> >>> I will forward the prior notice to talk-us@
>
> Discussion is better than notice, and imports isn't the greater US
> community forum, talk-us is.
>
> >>> advise as to the checklist steps we are not following, or which steps
> should
> >>> be added to this checklist?
>
> See above.
>
> - Serge
>

-- 
Jeff Meyer
Global World History Atlas
www.gwhat.org
jeff at gwhat.org
206-676-2347

-- 
Jeff Meyer
Global World History Atlas
www.gwhat.org
jeff at gwhat.org
206-676-2347
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/imports/attachments/20121204/b3f57cff/attachment-0001.html>