[OSM-talk] talk Digest, Vol 44, Issue 6
dave_ebling at yahoo.co.uk
Wed Apr 2 14:02:34 BST 2008
> well, the attribution they want is so close to that
> cc-by-sa as to have no real-world difference
>>whenever we display NZ we would have to display
>>copyright" somewhere totally undermines the purpose
>not necessarily - the settled upon option appears to
>be a link on the
>map page pointing to the attribution page, where all
>data sources are
>listed. the link doesn't change, nether does the
>text on the front
>page, or the attribution page
>the CC-by-SA licence - to make sure people know that
>the data is freely available.
>the two (copyright and cc-by-sa) don't have to
>contradict each other
True, I acknowledge that, but requiring two seperate
attributions for OSM data is going to be confusing to
people who use the data. So far we have managed to get
by on just one attribution.
>well, the license is the overriding point here,
>and it says anyone can do whatever they want, as
>long as the source of the data is acknowledged.
>anyone can change, add, remove, merge, whatever
>and they (linz) and we (osm) won't mind.
I am not arguing about the license. I don't consider
myself competent to say whether the licenses are
compatible. What I am bothered about is having to
change the attribution on all OSM derived data
>we're not proposing to undermine anything - anyone
>can still take the
>osm dataset and do with it as they wish, so long as
>they follow the
>(very brief) rules laid down in cc-by-sa. adding in
>the linz data does
>not change anything in that regard
It changes the attribution which all data derived from
OSM (where NZ is involved) must display to one which
is considerably longer. At the moment you can just put
"(c)openstreetmap CC-by-SA" in the bottom corner of a
map, right? Won't all derivations and derivations of
derivations have two licences and attributions applied
to it, even if they are compatible?
The fact that we can provide a set-up on the OSM home
page that meets LINZ's requirements is one thing.
Whether everyone who ever uses the data in future
wants to have to display LINZ's copyright is another
matter, and the one that concenrs me more.
Imagine if we import data for many counries in the
world, each with an extra attribution. Now imagine if
I print a map and put it on a leaflet, incorporate the
data or a map into some software, etc etc. There may
not be easy attribution schemes that meet all the
possible uses of OSM data.
If we carry on down this path and keep adding
attribution requirements, we will end up with a map
that meets this description: "maps you think of as
free actually have legal or technical restrictions on
their use, holding back people from using them in
creative, productive or unexpected ways." Does that
sound familiar? It's things like this that make me
wish that OSM was public domain not CC-by-SA.
Unfortunately I know this will never happen.
>agreed, it is a big step - it does need to be
>discussed and analysed
>in great depth and of course, we can go back,
>if the data is labelled as being
>sourced from linz
Except that all the other data in NZ will end up being
linked to the Linz data set, surely? - for example, if
someone adds a footpath that links to two roads that
came from the Linz data set, what happens to the ends
of that footpath? What if it only has two or three
nodes? And if someone corrects the name of a road in
the Linz dataset, or adds a bus stop to a node, etc
etc. As soon as the dataset is imported, it will begin
to be merged with OSM data. Removing it again will
mean deleting peoples' hard work. So I believe we
should be in no rush whatsoever to go ahead, even if
we have agreement from LINZ with the proposed
Sent from Yahoo! Mail.
A Smarter Inbox http://uk.docs.yahoo.com/nowyoucan.html
More information about the talk