[Talk-ca] Nunavut has a very pour data coverage. Why not Integrate BNDT/CanVec dataset from Department of Natural Resources Canada?

[Andrew]

On Fri, 2014-03-28 at 19:08 +0100, ivo at gmx.com wrote:
> Mainly I am interested in the Quttinirpaaq national park. I think it is region 120 and 340. 
> 
> http://en.wikipedia.org/wiki/Quttinirpaaq_National_Park
> 
> The Queen Maud Gulf Bird Sanctuary would be also nice. It is region 66N and 66O. Center of it is Karak Lake N67.23719 W100.26031.
> 
> Why would it take years? The complete Canada topo map from Ibycus is "only" 3,5GB big. 
> 
> Ivo
> 

Ivo:
	A couple of reasons that it takes me soooo long, is:

	A lot of the time at, least down south is that there is existing data
already mapped. I'll open a canvec tile and then download the existing
data. Now for example say some one has mapped a railway line I have to
remove the rail way line from the canvec data set and then make sure the
segments are connected. The I'll run Josm validator and remove all the
duplicate nodes. Then I'll review the combined data and connect lakes
and rivers so they are "one body" of water. Since canvec chops lakes and
rivers in little pieces. Now granted I don't do this with the forest
data, had to decide what was a reasonable time / effort to make the data
look "pretty". Then upload the data.

	So basically it all takes time and it is a hobby so I do a little now a
little later and well you get the idea.

	If I do 4 tile segments a day which I'm guessing is an hours work

Tile 120 has 
Blocks B, C,  D, E, F, G
B has 03, 04, 05, 06, 11 12, 13, 14, 15
I'm looking at 15 tile pieces for 120B would take say 4 days to import.
Then onto the 120C, with a quick look at 120 C I would guess a week or 2

So If, I just blindly imported data I could fly through the tiles and
end up with a pile of spaghetti.

	Andrew