[Openstreetmap] OSM License and debian

Lars Aronsson lars at aronsson.se
Fri May 27 19:06:26 BST 2005


Alex Willmer wrote:
> I really think we need to make the decision on licensing for openstreetmap
> data. I believe the following summarizes the 2 choices:
> 
> 1. Creative Commons, Software or bespoke license
> 2. Viral (ie GPL/Sharealike) or attribution (ie BSD/CC BY) or dual licensing

Sorry, I don't understand the differences here, and this means I 
will not be able to explain it to my friends.  I've been a C/UNIX 
programmer for 19 years, and I'm very familiar with GPL, and all 
the discussions that projects like Wikipedia have had over GFDL or 
CC/SA licensing.  My personal opinion is that it would make little 
practical difference if all of Wikipedia was in the public domain. 
Most Wikipedia article texts are so mixed up with contributions 
from various anonymous and unknown writers, that it is hard to 
determine who has the rights to take a copyright violation case to 
court.  This means a republisher could likely get away with 
disrespecting Wikipedia's GFDL licensing just by claiming that the 
authors had double licensed the same text.  In the end it comes 
down to the amount of effort the community is prepared and able to 
spend on protection against such disrespectful republishing. The 
same will probably be true of data from Open Street Map.

Perhaps this is a provocative thought, but I think the question 
will come and we should have an answer prepared: Why not just put 
all the data in the public domain?  (Or a CC license that requires 
neither share-alike nor attribution.)

I'm new to GPS, but I think that the geocaching and outdoor sports 
communities already have large repositories of waypoint data.  
Just google for .WPT files.  I think Open Street Map could recruit 
many new members among these experienced GPS owners, and it could 
be a definite advantage to use the same conditions for sharing 
data that they are already used to.  So how do they do it?  I 
don't know.  Do you?

I'm not familiar with any other projects that collects and shares 
track log data. I think it would be a good idea to have this as a 
separate (sub-) project, because they can be used in so many more 
ways than the current OSM implementation does.  For example, at 
one particular street crossing, historic track logs could tell us 
that Monday thru Friday at 8 am, most people turn right here.

In fact, if such a repostory already existed, then OSM could get a 
much faster start my reusing old data.  It would enable OSM to 
focus on how to dig through old data, rather than to design and 
debug the upload servlet.


-- 
  Lars Aronsson (lars at aronsson.se)
  Aronsson Datateknik - http://aronsson.se




More information about the talk mailing list