[Talk-ca] CanVec Questions

[Austin Henry]

- Sam Vekemans arranged a host of electrons thusly: -

[snip]

> I didn't include the accuracy_meters=* tag, but i could.  I didn't feel it
> necessary, since it's just a rectangle.  And when you go an map more detail
> (add in where the school parking lot is) and where the playground/sports
> areas area) The building just needs to be highlighted.
> 
> > out of where it should be, and the CanVec data still shows the old building
> > for Sardis Secondary School, which was demolished in 1995 and replaced with
> > a new building in a different location/shape.
> 
> I didn't include the latest-update data as a tag, but i could.
> 
> Austin, has experience with school tagging, as the area of the school would
> not include the 'building=yes' so then it all shows up better.

There have been a few instances like that in Colwood/Langford too, but
for the most part, the schools have no money, so haven't done any major
building for the last 10 years or so since the data seems to have been
collected.

In the area where I've been working, it has seemed that every school
area fron canvec was a building, rather than the whole school yard.  I
can't speak for other places, but I expect that it's the same everywhere
(consistence seemingly being a big deal for this kind of data :)

> > I feel comfortable doing the CanVec import for my home town because I know
> > the area well enough to omit poor data, but I don't know if I should go
> > ahead with CanVec conversion for other areas. Most of the data is good, it's
> > just the few outdated points that are questionable. Should I go ahead with
> > other areas and just blindly import, or is it better to wait for local
> > mappers?
> >
> 
> Yup.  ... I'd say just load your local-local area that you know is right.
> Maybe do a reverse select, and save that .osm file (which omits the area you
> just worked on) as a the-rest-of-data.osm file, and keep in handy (on your
> computer).
> Since your the main-mapper for the tile area, other local mappers would
> consult you as to what data to copy over.  (you can also list the details on
> the main Canada Import Chart if you like, but not necessary)

It's always a bit of a trade off, I figure.  On the one hand, if there's
nobody mapping in an area, errors could persist for an awfully long
time.  But if there's data (perhaps even blatantly wrong data), a
non-mapper might be motivated to fix "just one thing" (and get hooked --
I'd know, that happened to me).

Your call, really.  I'd invite others to comment, too.

cheers,
	Austin.
-- 
Build a man fire, he'll be warm for a day. 
Set a man on fire, he'll be warm for the rest of his life.