Tobias,<br>thank you for writing this. It seems you are speaking from the hearts of many people.<br>thanks,<br>mike<br><br><div class="gmail_quote">On Sat, Jul 23, 2011 at 6:11 PM, Tobias Knerr <span dir="ltr"><<a href="mailto:osm@tobias-knerr.de">osm@tobias-knerr.de</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">I apologize in advance for distracting everyone's attention from mapping<br>
and other pleasures with a long mail about licenses. However, I have<br>
begun to seriously doubt the decision to choose ODbL as the one future<br>
OSM license, and believe we should have another look at the license that<br>
will ultimately be used for the database published by OSMF.<br>
<br>
In a nutshell:<br>
The Contributor Terms give the OSMF the right to publish OSM data under<br>
the terms of the CC-BY-SA. I suggest that the OSMF should use that<br>
right, and *continue to publish the OSM database under CC-BY-SA* after<br>
the end of the license change process.<br>
<br>
So do I suggest to stop the license change process? No, I don't. The<br>
Contributor Terms will solve many problems on their own, so my<br>
suggestion is what could be labelled "CT + CC-BY-SA".<br>
<br>
Do I suggest to abandon ODbL? No. Though I do not like ODbL very much, I<br>
do not oppose making the OSM database available under that license. If<br>
there are people who want to use OSM under ODbL terms - and I expect<br>
there will be some - we can allow them to do so by dual licensing our<br>
database.<br>
<br>
But what I'm asking for is that we continue to offer fresh OSM data<br>
under the most attractive license for many honest and productive users<br>
of our data: CC-BY-SA.<br>
<br>
<br>
Now for the details. In my opinion, CC-BY-SA has very desirable features:<br>
<br>
* It is easy to comply with.<br>
* It is popular and trusted.<br>
<br>
Below, I will explain why - so if something doesn't seem instantly<br>
obvious to you, read on. In addition, I will address the following<br>
topics commonly discussed in the context of our license change:<br>
<br>
* Collective attribution<br>
* Compatibility with other licenses<br>
* Future-Proofness<br>
* Uncertainty and doubt<br>
* Inadequate protection<br>
<br>
<br>
* CC-BY-SA is easy to comply with. *<br>
<br>
This is what I consider the most compelling benefit. After all, if it's<br>
easy to work with OSM, then people will create cool stuff ("products",<br>
"produced works").<br>
<br>
With CC-BY-SA, you add a license note and attribution to your product<br>
and you are done. All other effects are purely legal (letting people<br>
copy and modify your work), and do not require any effort on your part.<br>
<br>
ODbL expects you to do the same, but adds another, far more onerous<br>
requirement: publishing derivative databases used in the creation of<br>
your products. It is hard to understand what these are, and it is often<br>
a challenge to distinguish them from produced works. Publishing the<br>
derivative databases can be a significant burden when compared with<br>
creation and distribution of the product itself. That's not a new<br>
discovery, by the way, and that's why the license attempts to solve the<br>
issue by letting you instead describe the process of reproducing the<br>
derivative database. Unfortunately, that option is not clearly defined,<br>
might not be possible with proprietary software, and can again be a<br>
significant burden for any producer using OSM data.<br>
<br>
* CC-BY-SA is popular and trusted. *<br>
<br>
The Creative Commons brand is well-known. ODbL and OKF are not. CC's<br>
popularity is for a large part due to its presence in art and popular<br>
culture, which is something that a pure database license will never<br>
fully achieve. The openness instantly associated with the CC licenses by<br>
many is an important differentiator for OSM when it is compared with<br>
closed competitors like Google Map Maker. People won't compare two walls<br>
of legal text. But if they learn that OSM uses CC, they know we're the<br>
good guys.<br>
<br>
* Collective attribution *<br>
<br>
We want to make sure that users of the data do not need to list all<br>
individual contributors. Luckily, contributors now make data available<br>
to the OSMF under terms that don't usually require attribution, and the<br>
OSMF re-publishes it under an license requiring attribution. Therefore,<br>
collective attribution should be safe with a CT + CC-BY-SA solution.<br>
<br>
* Future-Proofness *<br>
<br>
We want to be able to change the license in the future if the geodata<br>
environment changes drastically. This is not related to ODbL, but a<br>
feature of the CT.<br>
<br>
* Uncertainty and Doubt *<br>
<br>
Another reason frequently suggested for the switch to CT + ODbL is<br>
making it easier for users of the data to know what they need to do in<br>
order to conform to the license requirements. The first aspect is "who<br>
do you ask if you aren't sure?" It has been suggested that introducing<br>
the Contributor Terms makes it easier to get a useful response because<br>
you no longer need to ask thousands of individuals for their<br>
interpretations of the license, but this seems to be a property of the<br>
CT, not of the ODbL.<br>
<br>
As far as the licenses themselves are concerned, I observe that CC-BY-SA<br>
seems to be /easier/ for people to interpret. Among the reasons for this<br>
might again be the popularity of the licenses: CC-BY-SA images are used<br>
by thousands of newspapers and websites, so I can just follow their<br>
example when using OSM maps in a similar manner and will likely not get<br>
it completely wrong. But it goes beyond that. The complexity of the ODbL<br>
itself makes it hard to define what you need to do to comply, and we<br>
have alreadly seen some indication of this when people asked what the<br>
ODbL means in detail.<br>
<br>
And, of course, choosing a dual licensing approach would let people pick<br>
the license they are more comfortable with.<br>
<br>
* Inadequate protection *<br>
<br>
CC-BY-SA might not "work" for data. OSM data is not currently abused in<br>
a manner that threatens the project, and that might never even happen.<br>
Nevertheless, it seems wise to make sure that we can either prevent this<br>
or at least react when it happens.<br>
<br>
It is true that, by continuing to offer the database under CC-BY-SA, we<br>
would no longer /preemptively/ address this potential issue. Making<br>
contributors agree to the CT gives us the ability to react *if* legal<br>
weaknesses of the CC-BY-SA are actually abused at some future point,<br>
though, and I believe that this is sufficient.<br>
<br>
* Conclusion *<br>
<br>
The CC-BY-SA is popular, understandable and easy to implement for users<br>
of our data. It does not build legal barriers that make using OSM much<br>
harder than it strictly needs to be, which encourages people to use OSM<br>
in creative, productive and unexpected ways. Continued publication of<br>
the OSM database under CC-BY-SA will therefore help us fulfil our<br>
project's mission, and can be implemented without disruption of the<br>
ongoing licensing process.<br>
<br>
-- Tobias Knerr<br>
<br>
_______________________________________________<br>
legal-talk mailing list<br>
<a href="mailto:legal-talk@openstreetmap.org">legal-talk@openstreetmap.org</a><br>
<a href="http://lists.openstreetmap.org/listinfo/legal-talk" target="_blank">http://lists.openstreetmap.org/listinfo/legal-talk</a><br>
</blockquote></div><br><br clear="all"><br>-- <br>James Michael DuPont<br>Member of Free Libre Open Source Software Kosova <a href="http://flossk.org" target="_blank">http://flossk.org</a><br>
<style>img, #cubbies-overlay{ -moz-transition-property: margin, box-shadow, z-index; -moz-transition-duration: 0.1s; -webkit-transition-property: margin, box-shadow, z-index; -webkit-transition-duration: 0.1s; }
.cubbies-selected{ z-index: 9999; box-shadow: 3px 3px 8px -1px blue !important; cursor: pointer !important; margin: -3px 3px 3px -3px; }
.cubbies-selected:active{ box-shadow: 2px 2px 5px -1px darkblue !important; margin: -1px 1px 1px -1px; }
#cubbies-overlay{ position: fixed; z-index: 9999; bottom: 30px; left: 30px; box-shadow: 0 2px 3px rgba(0,0,0,0.8); border: none; }
#cubbies-overlay:hover{ box-shadow: 0 2px 3px rgb(0,0,0); }</style>