[Imports] CLC meadow cleanup

[Ilpo Järvinen]

On Sat, 22 Mar 2014, Pieren wrote:

> On Sat, Mar 22, 2014 at 8:32 AM, Paul Norman <penorman at mac.com> wrote:
> 
> > If the area I looked at is representative, I am contemplating proposing
> > a mechanical edit to remove the bad data. What are peoples thoughts on
> > this?
> 
> I think France is the 2nd or 3rd biggest community of contributors in
> OSM. I'm surprised that you start the discussion here. I feel like an
> American receiving a request from one guy in the middle of the pampa
> to delete all tiger data in US because this guy was able to detect the
> bad quality of that import. What I mean is that you should ask first
> the local community if the idea is good or not especially when this
> community is big and mature. We always strongly encourage people to
> replace to original CLC landuse polygons with better ones (hem... is
> it not the principle of OSM ?). If you find a place where the landuse
> is not that good, it is simply because there is no contributor there.

?? He was fixing it and you start ranting? Do you also rant to those 
from FR like this if they propose removal of CLC? Wasn't particularly 
"encouraging" words you used here ;).

So far I've not come across mappers who find it easier to "replace" CLC 
than draw from scratch... but maybe it's different in France? I don't 
speak French so it's hard to understand if mappers there are giving 
positive or negative feedback about it, except that I now have a non-FR 
guy who was fixing/contributing in FR and immediately gave negative 
feedback about CLC which caused you to go up in arms (whether any 
opposition similarly silenced by few in FR community too, I don't
know. :().

At least here in Finland it's only negative feedback from the mappers, so 
far here only those who have imported it speak any positive about it even 
them, like you, acknowledge that it is tends to be "not that good" when 
there's bad place that hasn't been fixed by real contributors and that it 
should be replaced with "better ones".

But I knew this all along, like I said in my other post to this thread, 
there will be resistance on removing CLC even if it is found/known to be 
incorrect and bad quality because of some magic reason (that you btw 
failed to tell use) :(.

> Is it a good reason to remove all of them ? If I find some place in US
> with very bad Tiger data, may I propose in my turn a mechanical edit
> to remove all Tiger data ? Because if you remove CLC meadow, you
> should remove all CLC landuses. And you should not stop in France but
> go ahead for all European countries where CLC was imported because it
> was the same source/process,

CLC is garbage everywhere else too. People answering to this thread 
certainly spoke mainly about other than France and Paul didn't seem to 
target France in particular either so I really don't understand why you 
think he would have had only France as "target" instead of CLC in general.

-- 
 i.