[OSM-legal-talk] Deconstructing the "loss of data" claim

[John Wilbanks]

Hi everyone. My name is John Wilbanks. I am the VP for Science Commons 
at Creative Commons, and I'm the one who wrote the Protocol for 
Implementing Open Access to Data.

I've been lurking here for a couple of weeks. I don't like showing up 
and posting without getting a sense of the community. But I think it's 
important to join this debate and get your responses...I'll stay here on 
this list, I'll take questions, and I'll take flames. I'm glad you're 
having this debate. It's a good debate. And I'm stoking the fires of it 
with this post - but I think you deserve my honest sense on this, and 
not a mumbling political post.

I am speaking here as an individual who endured 18 months of research 
into open data as part of CC, but not on behalf of CC the organization.

 > > The issue was quite simple. We need to have a license that better
 > > protects the OSM data

I'm going to be a little provocative here and say that your data is 
already unprotected, and you cannot slap a license on it and protect it. 
This sounds to me like someone deciding to put a license on the United 
States Constitution. You're welcome to try that - but that license will 
not change the underlying legal status of the Constitution, which is the 
public domain.

That means I'm free to ignore any kind of share-alike you apply to your 
data. I've got a download of the OSM data dump. I can repost it, right 
now, as public domain. You can perhaps try to sue me - though I'm pretty 
sure I would win. But you absolutely couldn't sue anyone who came along 
and downloaded my copy and then reposted that same data as public 
domain. There's no copyright on your data, and that means that only the 
people who sign your deal are bound to it, not anyone who gets it from 
the people who sign your deal. Unlike the copyright on a song, which 
travels along with the song no matter what, the share-alike is 
restricted to the parties of the contract. It's called privity - see 
http://en.wikipedia.org/wiki/Privity_of_contract

That's why the ODC was something that CC didn't support. It holds out a 
promise of power that is illusory. Privity + no IPRs = easy to repost as 
public domain.

 > Do we? What's the threat? How has it been assessed?

The threat comes from people who are worried about data "Capture" - but 
it is wrong to think that the law is a magic wand here. GPL style 
approaches only work in the presence of copyrights. They simply don't 
have the power in data. The real answer is: if you put your data online, 
it's essentially in the public domain.

Bad people know this and will exploit it. That's what you don't want to 
allow. But simply posting the data gives the bad people a lot of power. 
The nice people are going to obey your rules and be constricted by them, 
but not the bad people.

I would encourage you to think not about the threat, about those who 
will take the data and not recontribute, or who will sell the data, than 
about the good people. Given that the data is already in the public 
domain, whatever license you choose, how will you instead *reward* those 
who think it's worth being a good citizen? That can be done with moral 
statements of normative behaviors and a trademark for OSM - only those 
who behave get to use it and advertise themselves as OSM compliant.

It's a far healthier strategy in many ways than relentlessly trying to 
stamp out perceived violators. One can imagine a data owner beginning to 
empathize with the RIAA, who see violations everywhere and in so doing, 
loses focus of the opportunities for good outcomes.

 > > OSM never started out as a PD project so why would we think that it
 > > would be better to recommend it go PD now?

Because that's the legal status of the project. My copy of your data is 
in the public domain. It doesn't matter what the wiki license says - 
that license only protects copyrighted things. You can choose to use 
contracts to entangle users, like the ODC does, but it will only block 
the good guys. The bad guys can work around that in less time than it 
takes to download your data.

jtw
ps - Jordan Hatcher, who drafted both the ODC and the PDDL, is a 
remarkable attorney and far more knowledgeable on the details of the 
contracts than I am. I'm proud to call him a colleague.