[OSM-legal-talk] Attribution

Mon Apr 28 22:57:47 UTC 2014

Hi Alex

I think it comes down to finding some data to support your argument about share-alike, given that there are so many companies which have no problem using OSM as it stands, and who could potentially lose out under a change.

There’s a lot of silent evidence out there; nobody is going to do a talk titled “why we have no problems with the license and everything is fine”[1]. Whereas, if a license change was going to benefit a company or organization rather than the whole ecosystem of companies, you could imagine it would create a lot of incentive to make a big fuss about how it should radically change, rather than them having to pay for a proprietary map to do whatever they wanted to do.

Besides, we already have large map datasets without attribution or share-alike. For example, TIGER. But there’s a reason nobody gives back  to that or builds anything around it: There is no sustaining share-alike feature. Instead, companies just take TIGER and use it commercially, closing off their fixes to anyone else. You could imagine someone doing that to OSM if the license was altered to remove share-alike, so I think we should keep it as is, for the benefit of everyone.

I think how you frame it as making it “more open” is very clever and insightful, but I’d frame is as “alive” or “dead”. Today, OSM is alive with people and community. If the license changed to help just one company, you could imagine it being dead like how the proprietary map user contributed systems work. Facebook vs. Google+. Alive vs. Dead.

Someone could prove me wrong by building an open community around public domain data, but it would probably be hard work and take a while. Perhaps there is one that exists somewhere, compared to all of the ones that exist around GPL or CCBYSA?

We should definitely spend more time on the geocoding interpretation stuff though.

Best

Steve

[1] - Although Eric Rodenbeck’s slide at SOTM US was great: “We don’t care about the license”

On Apr 28, 2014, at 4:35 PM, Alex Barth <alex at mapbox.com> wrote:
> Steve,
> 
> Agreed on a transparent process for tracking unattributed applications of OpenStreetMap.
> 
> Separate from attribution however, the issue with “share-alike” is that it's not open, and hurting our community. ODbL's share alike is simply shutting out OpenStreetMap from many use cases = adoption = incentives to contribute. While share-alike was intended to foster this project's growth it now is only putting limits on it. We can fix this. There are straight forward ways to make a license modification to ensure that our community can evolve with the best most open license.
> 
> It's simplistic to make broad analogies - this isn't Linux or Wikipedia: OpenStreetMap is data, data is useful down to its smallest subsets and unfolding the full power of data is to allow it to be combined and mashed up in any way with other datasets. And for OpenStreetMap to be used as part of everything and used by different communities it needs to be truly open data. By being more open we will grow our community and only improve the data quality.
> 
> And it is this data that is going to change the world. The world is changing fast right now and the very fact that OpenStreetMap is available in a structured format starts to be its biggest advantage. We can see this in humanitarian applications every day. I'm saying this with the deepest respect to all contributors here: What's to gain is OpenStreetMap in a lot more applications than today. With share alike dropped, a huge hurdle for using OpenStreetMap is just gone.
> 
> Alex
> 
> PS: My talk on this very issue at SOTM-US http://stateofthemap.us/session/more-open/
> 
> 
> On Mon, Apr 28, 2014 at 2:42 PM, Steve Coast <steve at asklater.com> wrote:
> http://stevecoast.com/2014/04/28/attribution-is-it-time-to-name-and-shame/
> 
> --
> OpenStreetMap is the global, open and free map dataset that anyone can use. It is created by a huge community of volunteers who pour their time and energy in to the project. It’s also fun, beautiful and cool.
> 
> So it’s sad that people don’t want to respect the license. It asks two very simple things:
> 
> Please say you’re using OSM. This is very simple.
> If you change the map, please give the changes back. This is called “share-alike”.
> Compared to paying a lot of money for incredibly license-restricted data, you’d think people would be ok with these requirements.
> 
> Sadly, this isn’t the case.
> 
> There are those who are now willfully disregarding our tiny little requirements. It’s being framed as some gigantic and unreasonable proposition, asking to say where the data came from or giving data back when you fix things. As if it’s completely bananas to ask such a thing. As if Linux or Wikipedia should be disaster ghost towns while asking for exactly the same thing of their users.
> 
> This is just baloney. The real comparison should be; if you don’t like the license you’re free to use expensive and complicatedly-license data. That’s your option. Those guys are just a phone call away, and will be happy to sell you data. You’d probably find that they have very strong attribution requirements, just like OSM does.
> 
> It is the ultimate disrespect to the volunteers who built the data to not even attribute their contributions. It’s even worse that there are some who’re trying to also own OSM for themselves by taking away the share-alike requirement.
> 
> Is the license perfect? I’m afraid not. Specifically we need more clarification around the technical implementation and use of geocodes, especially in relation to other datasets. It’s hard today to technically comply with some of those edge cases.
> 
> But that’s not what we’re talking about. We’re speaking here about the simple ask, that if you use OSM you please say clearly on the map that it is OSM. You’re getting a great dataset, for free, under an open license, that millions of people are contributing to. We’re not asking for $100,000 license fees, we’re just asking that you say who we are.
> 
> It’s the ultimate human need; I was here. I did this.
> 
> How could you deny people that?
> 
> Apparently, easily and willfully. People within the OSM community have been frustrated and trying to fix it for some time. If we were a proprietary map supplier we’d revoke a license or jump to legal options.
> 
> We are much nicer than that. I propose a four stage plan, organized on OSM’s legal mailing list and tracked on the wiki:
> 
> A polite email, linking to our requirements
> A week later: Another polite email, warning of what’s to come.
> A week later: Another polite email, same as above
> A week later: Very public naming and shaming on OSMs various social media channels and blogs
> Most people who miss our requirements are making a simple error. This is a process that gives three opportunities and an entire month to correct the mistake. This is not a brand new idea or process. The FSF and others have named & shamed (and have even went further) for GPL violations in the past.
> 
> In a narrow way, this all a good thing. It shows the growth and maturity of the project, that there are those out there that want to own it or take all the advantages without even saying where the data came from. But in the end, we have to defend ourselves for what little, tiny things we ask.
> 
> 
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
> 
> 
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20140428/eb69b1b4/attachment.html>