[OSM-talk] Terms of use for OSM tiles and API?

[Frankie Roberto]

2009/9/29 mdeen <mdeen at xs4all.nl>


> As GNU puts it quite rightly: 'you should think of "free" as in "free
> speech," not as in "free beer"'.
> OSM does not hand out free beers (unlimited access to rendered tiles) but
> gives free speech (you can use and modify the data freely).
>
> But I agree it should be made clear that while OSM does give out some free
> beers, it is limited and when the tap is broken, there is no guarantee it
> will be on when the pub opens again.
>

I don't think anyone is suggesting that OSM should make any guarantees as to
the availability/speed of its tileservers or XML API.

However, some kind of policy spelling out that, for instance, embedding
tiles on another site is okay, even if it's commercial, so long as it
doesn't use up a ton of bandwidth, would be a good idea.

With the XML API, we could simply spell out that it's mainly intended for
interactive use, and perhaps even suggest a limit to the number of requests
(above which people should implement their own server from the planet
dumps).


> > But, the root cause is something unfortunately out of OSM's control -
> the
> > cost of server space, server power, and bandwidth needs to come down,
> it's
> > a significant hurdle to not-for-profit hobbyists wishing to develop
> > powerful websites.
>
> There is no such thing as free beer. If they want to depend on it, they
> should support the item they rely on i.e.: offer server space or give
> monetary support).
>

Suggesting that people who use the API or tileservers for commercial
sites/projects should donate something is a good idea.

I'm asking all this because

a) we have an export feature which uses our tileserver, but doesn't make
clear how you're allowed to use the embed code.

b) I've met dozens of people, from big commercial clients to developer
contacts, who have wanted to use the tileserver or API, but who haven't been
sure about what's allow, and haven't got the time or expertise to set up
their own servers.

I do think that, for OSM, Wikipedia, DBpedia, Freebase, GeoNames, etc are
more appropriate examples than the Linux kernel. Whilst the Linux kernel is
pretty stable, and gets included and compiled into bigger projects, OSM and
Wikipedia contain fast-changing data, and so it's more natural to have the
place where they're consumed and the place where they're created much closer
together. So whilst there are plenty of data mirrors of those services out
there, most people tend to use the APIs provided by the services themselves,
unless the have particularly big requirements.

Frankie

-- 
Frankie Roberto
Experience Designer, Rattle
0114 2706977
http://www.rattlecentral.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20090929/1009e1ac/attachment.html>