I'll try to sum up my understanding.
Currently your view is the existing import plan for Canada building
imports does not meet the import guidelines and cannot be amended to
meet them. I was hoping that something might have happened in the last
six months on the preprocessing of data side which would have meant we
could have got a consensus to move forward but no movement on this has
apparently happened.
Second no consensus exists for importing all the buildings or how it
should be done. My personal view is originally I didn't think this was
achievable for all Canada. There were simply too many players and
getting complete unanimous consensus was always going to be a problem
but it was found the most difficult part of the Ottawa buildings import
was the import mailing list and the OSM guidelines for imports so based
on input from outside the original Ottawa group the decision was made to
go for a Canada wide import plan to minimise the requirements for
multiple people to submit import plans. The technical side of the
import by comparison was fairly simple and much more interesting. So it
was worth trying a country wide import plan.
The Ottawa import was thrashed out over coffee and the local mappers are
happy with the data quality even though some non-local mappers were not.
I still have reservations about the ability of some of the smaller
groups of mappers to get an import plan through all the loops.
So I think we are at the point where ideally we expect local groups to
come up a decision on data quality and submit individual import plans.
All three sources, Stat Can, Microsoft and NR Can LiDAR data meet the
licensing OpenStreetMap licensing requirements but your plan will have
to specify which license is being used for the import and how it meets
the OSM guidelines.
I note some imports in Africa are done by mappers who live outside
Africa so I'm unclear if you need to live in the area to import or not
or how far away you can be and still be counted as a local mapper.
Since in Nate's opinion the Canada wide import plan does not meet the
guidelines please do not use it as a template.
Have fun
Cheerio John
Nate Wessel wrote on 2019-09-27 12:03 PM:
John,
You are once again purposely mischaracterizing my position on this and
I do not appreciate it one bit. Your import did not follow the import
guidelines and was not approved by the broader community. It was not
sent out to the mailing list, it was barely documented, it was not
posted on the import page on the wiki... I could go on. I was not the
only person who had a problem with it - I was just the first to say
something. The buildings imported in Toronto were very low quality IMO
and the changes were happening very, very quickly and without notice.
But we do not need to rehash old fights.
I would like to see buildings (re)imported in Toronto (the rest of
Canada is not really my business), but I would like to see it done
right. I can elaborate what I mean by that, but so can the archives of
this mailing list. If people are interested in engaging in a serious
discussion about moving forward with a building import for Toronto, I
am happy to engage constructively with that.
Respectfully,
Nate Wessel, PhD
Planner, Cartographer, Transport Nerd
NateWessel.com <http://www.natewessel.com>
On 2019-09-27 11:44 a.m., john whelan wrote:
From memory we have imported Ottawa's buildings under the correct
license (Stat Can so the federal government's open data license) and
the quality was deemed acceptable by the local mappers.
Then we opened up a second import plan to import buildings and a fair
number were imported. This included an task manager set up with
tiles to assist the mapping.
The data sources were different as each municipality created their
own source.
My understanding is some mappers thought the data should be
preprocessed and two or three were going to come up with a plan to
preprocess the data.
About that time an American mapper, Nate, who was living in Toronto
took exception to 1,000,000 buildings being imported in Western
Canada and requested the DWG to remove them without waiting for
discussion on talk-ca to address his concerns.
We do have three sources of correctly licensed data, the Stat Can
data sets, the Microsoft data sets, and the NR Canada LiDAR data.
I do know that a number of departments and agencies would like to use
buildings and although they can use the open data sources using OSM
would be more convenient.
I'm not sure what if anything is happening at the moment.
Are we expecting local groups to draw up their own import plan as
Ottawa did since we seem to be unable to get a consensus across Canada?
My gut feeling is with three sources of data we'll see new mappers
importing in buildings without going through an import process. Are
we content to let that happen?
Have whoever it was who was going to come up with a preprocessing
plan done so? Has it been accepted by the rest of us?
I note that Pierre has noted there is now a validation process in
JOSM for correcting buildings that are not quite 90 degrees on the
corners. Would an acceptable approach be to import then return to
check the angles on the corners and correct them?
Does Nate still have unaddressed concerns about buildings being
imported in Western Canada?
Can we get a consensus about what to do next?
Thanks John
*
*
*
*
_______________________________________________
Talk-ca mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/talk-ca
_______________________________________________
Talk-ca mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/talk-ca
--
Sent from Postbox <https://www.postbox-inc.com>
_______________________________________________
Talk-ca mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/talk-ca
