[Imports] NSW, AU Public Schools Import

[Frederik Ramm]

Hi,

On 25.04.2018 09:30, Andrew Harvey wrote:
> I'm planning to import NSW, AU Schools data.

Quick thoughts on your message to talk-au:

* I find access=private a bit strange. By your logic ("can be thrown out
any time") every restaurant and shop would also be access=private. I'd
leave it out, I do not think it adds any information.

* capacity - I suggest to leave that out. The current enrolment figures
don't say anything about the capacity; the school could even be over
capacity. (NB capacity:pupils and capacity:persons are both in more
frequent use with schools than just capacity, though they probably all
stem from imports.)

* Mapping source "ref"s with imported data is suggested by many
importers but it's not generally well liked. By trying to keep the link
with the "original" data, you give the signal that this is somehow not a
native OSM object but a copy of something kept elsewhere. If a mapper
splits up a school campus for some reason, what are they supposed to do
with the various "refs" - keep? remove from one half? remove from both?
call you to discuss? This uncertainty might lead to mappers not daring
to edit the object. Do you want to discourage people from editing the
imported data?

* contact - I tend to use "phone=x" and not "contact:phone=x" but I
guess that's a matter of taste.

* school:gender - I'd seek more input on that, it is very rarely used.
Unsure if it is the best tag, maybe a more general way of saying "who is
this school for" can be found.

* source tag - on the changeset.

* You write "I'm aware there are a number of "schools" in the upstream
data we might not consider schools for OSM". I'd try to weed them out
somehow, or at least try to find tags that allow the consumer to weed
them out. I figure that people might do things like "judge the quality
of accommodation by how far it is from nearby schools" or so, and if it
turns out that the nearby school is just a summer excursion destination
- not so great.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"