[OSM-legal-talk] ODbL: How obscure/inaccessible can published algorithms be?

[Frederik Ramm]

Hi,

80n wrote:
> I think you've lost the thread.  Now, you are arguing that you can't 
> spot a derivative database.

My original question was aiming at whether or not there are ways to 
weasel yourself out of the requirement release derivative databases or 
the algorithms leading to them.

I think we have now established that whenever you do something with OSM 
data that involves a derivative database, but just to make things 
simpler for you and not as an absolutely necessary component, then 
nobody can prove that you are using a derivative database, and nobody 
has a legal right to challenge you for an explanation - it's your 
business secret.

People can write to you and ask; and if you don't reply (or reply that 
you don't use a derivative database) all they can do is sue you for 
breach of license and hope that the judicial process finds out the truth.

It is not such a big difference from what we have today. Even with 
Easter Eggs and all, I can never prove beyond doubt that someone is 
using OSM, I can only collect evidence and then ask a court to clear 
things up.

But it is probably easier to collect that type of evidence than to 
collect evidence for someone having a secret derived database.

Bye
Frederik